0% found this document useful (0 votes)
448 views364 pages

Sanet - ST - 978 981 19 4340 9

Uploaded by

lrbg nkks
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
448 views364 pages

Sanet - ST - 978 981 19 4340 9

Uploaded by

lrbg nkks
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 364

Management for Professionals

Ganesh Mahadevan
Kalyana C. Chejarla

Lean Management
for Small and
Medium Sized
Enterprises
Adapting Operations to Changing
Business Environment
Management for Professionals
The Springer series Management for Professionals comprises high-level business
and management books for executives. The authors are experienced business
professionals and renowned professors who combine scientific background, best
practice, and entrepreneurial vision to provide powerful insights into how to
achieve business excellence.
Ganesh Mahadevan · Kalyana C. Chejarla

Lean Management
for Small and Medium
Sized Enterprises
Adapting Operations to Changing
Business Environment
Ganesh Mahadevan Kalyana C. Chejarla
Kanzen Institute Asia Pacific Private Ltd. Institute of Management Technology
Secunderabad, India Hyderabad Campus
Hyderabad, India

ISSN 2192-8096 ISSN 2192-810X (electronic)


Management for Professionals
ISBN 978-981-19-4339-3 ISBN 978-981-19-4340-9 (eBook)
https://doi.org/10.1007/978-981-19-4340-9

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Singapore Pte Ltd. 2023
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
To our families…
Foreword

The upheavals across the world over the last three years are unprecedented. Just
as the world started coming to grips with the COVID-19 pandemic, along came
the Ukrainian conflict. We have seen entire industry segments almost wiped out
due to a combination of lock-downs and international travel restrictions. Factories
were shut down in different countries resulting in a huge shortage in key compo-
nents that lead to even the mighty Toyota scaling down their own operations. The
recent rise in fuel process is causing intense pressure on logistics, manufacturing
and processing costs and thereby profitability margins. And yet at the same time,
industry is also having to cope with dampened market demand.
The brunt of these disturbances has been borne by the Small and Medium Enter-
prises (SMEs) who do not have the large reserves needed to cushion such long
term shocks. It is not easy to suspend operations for months together and still be
able to pay off salaries, interest on loans and fixed costs. Several businesses have
folded up or scaled down operations significantly. This does not augur well for the
economy at large since it is a well proven fact that SMEs form the backbone of
nation’s economy and are a significant employer in most nations. So what can an
SME do to survive in this environment and continue to grow in the long run?
In my opinion Lean remains the “Go-to” philosophy in these difficult times
and it has never been more relevant than now. Two key characteristics of a Lean
organization stand out as a beacon of hope – flexibility and productivity. Flexibil-
ity enables the organization to scale up and down rapidly to match the business
demand and volatility in the external environment. Productivity makes us more
efficient and cost competitive. Both these aspects lead towards long term growth
and sustenance. Lean as a management philosophy has been around for over four
decades. Numerous books, articles, and papers have been published and anyone
can just Google to read and learn about Lean concepts, tools, and techniques.
However, the actual rate of success in implementation remains a question mark.
In the case of SMEs, rate of adoption itself has been negligible.
This book therefore has come out at exactly the right time and context. The
authors seek to explain the reasons for Lean “hesitancy” and provide alternative
approaches to break down the barriers to adoption. It addresses, how business
owners and senior executives motivate themselves to invest resources in commenc-
ing the Lean journey. And having commenced, how does the organization sustain
the practices to ensure long term business growth. The main body of this book
vii
viii Foreword

lays out a comprehensive stage-by-stage methodology to implement a well-defined


roadmap concluding with ways to sustain the improvements through culture build-
ing practices. The chapters are peppered with real life examples giving readers
an insight into how Lean actually works and delivers results. A few features set
this book apart from other books on Lean. It provides additional guidance on
how Lean integrates with other management philosophies and technologies such
as IoT. It discussed, how Lean helps cope with prevailing uncertainty. And most
value adding are the case studies spanning diverse situations and industry sectors
which are written from the first-hand experience of the authors.
Over a decade ago I had the good fortune of engaging with Ganesh Mahadevan,
one of the co-authors, during the course of our SME Financing and Development
project implemented by German Technical Cooperation (GTZ). He came out with
a proposal to implement Lean in the pharmaceutical sector at a time when it was
unknown to the entrepreneurs in this sector. The initial seminar to bring about
awareness was well attended and three young second generation business owners
signed up by committing to pay 50% of the consulting fees. Over the next six to
nine months, I was able to witness first-hand, the transformation in their factories
and processes leading to significant and ongoing business gains. More importantly,
Lean was able to positively impact employees in the organization across the hier-
archy right from top management to the plant operators. A decade later, all three
businesses continue to grow. The most consistent adopter of the lot has now grown
tenfold and has become a global exporter with multiple facilities.
The story I have shared above is a teaser of one of the several case studies that
form the latter half of this book. Each case study shares a different perspective on
the Lean implementation journey focusing on how to make Lean work across as
diverse fields as a restaurant, an engineering job shop, high volume manufacturing
set up, a warehouse and an agriculture based processing unit. There are learnings
to be had by one and all whether you own a business, manage and run it or are an
operating level executive.
I spent the intervening years since my first-hand experience of Lean at work as
the Head of the National Institute – MSME, India’s premier knowledge hub for the
development and growth of SMEs. Government officials and SMEs from across
the globe have participated in various NI-MSME programs covering organizational
development, sustainability and growth. But the Lean intervention left an indelible
mark on my mind and continued to remain at the top of my consciousness. And
Foreword ix

so it is with great happiness that I welcome the publication of this book. I urge all
of you to read on…learn …and grow! Best wishes to authors and publishers.

Hyderabad, India Chandrashekhar Reddy


Technical Advisor, Indo German
Vocational Education and Training
Programme (IGVET) GIZ-INDIA:
Former Director-General National
Institute for Micro Small and Medium
Enterprises (NIMSME)
Preface

Lean is becoming an ever more relevant operational strategy in the post-COVID


world due to the changing socio-economic scenario. Global recession and new
work normal such as social distance are forcing organizations to look at ways to
boost productivity, make processes fast, flexible, and Lean. A recent McKinsey
report indicates that the COVID induced disruption may result in a 16% to 26%
of global trade shift. The emerging economies in South and Southeast Asia and
elsewhere are some of the beneficiaries of such shift. However, for the emerging
economies to be competitive, they would need to embrace Lean concepts. SMEs
from emerging economies have a great opportunity to move up to the next level
both in terms of business growth as well as in becoming a part of the global supply
chains.

Perspective of the book

Over the past two decades, Lean management has become one of the most talked-
about management philosophies for business operations. There is a lot of literature
in terms of books and articles, training material, and publicly available informa-
tion on the internet. Millions of professionals world over are training themselves on
Lean and its allied concepts. Yet, when it comes to actual implementation, there
are more failures or sub-optimal implementations than there are successes. It is
now an established fact that merely copying tools and techniques does not deliver
the desired results. The ability to achieve improvements and sustain them depends
on piercing through superficial facts and look in-depth at factors contributing to
long-term successful implementations despite seemingly adverse circumstances.
The failure to sustain is both due to a lack of appreciation of the Lean philos-
ophy as also due to missing out on certain critical implementation nuances. The
primary objective of this book is to bridge the gap of the unavailability of an
implementation-oriented Lean management book, especially one that focuses on
the needs of SMEs. Also, most books on Lean and its implementation are based
on case studies and experiences of Japanese organizations, primarily Toyota, and
of organizations in the Western world (USA and Europe). This is difficult to relate
to for industries in developing economies where the economic, social, and cultural
environments are different. Again, most literature especially the case examples
xi
xii Preface

are related to large organizations and therefore focus on aspects of organization


culture, building Lean champions and teams and looking at long-term Lean and
Continual Improvement (CI) interventions with a horizon of years. For SMEs the
strategic focus is mostly on the here and now. They do not have the bandwidth of
large teams or afford to employ Lean champions. This book is intended to bridge
these shortcomings in the current body of Lean knowledge.
Diverse business situations such as meeting the demand fluctuations, designing
a facility, or improving profit margin, etc. are included in the case studies from
diverse industry sectors, to ensure that every reader finds a situation similar to
his / her organizational situation. While the publicly available literature on Lean
offers a large collection of tools and techniques, given each organization’s unique
context, the choice of the right sequence of tools differs. The book offers guidelines
in terms of which solutions work in which context, backed by real cases, which
is a big help to the resource constrained SMEs. This book is an equally good
resource for the organizations that have already implemented Lean, as it provides
realistic pointers about sustaining, tackling supply chain uncertainties and going
beyond Lean by integrating emerging technologies and management principles.
The following are the highlights of this book.

• Focusing on what Lean means as an operational strategy for the SMEs and how
to deploy and make it a sustained success considering the socio, cultural, and
economic factors unique to developing economies,
• Lean is all about the implementation and chapters are organized in terms of
project life-cycle based flow rather than basis of Lean tools and techniques,
• Diverse real-life SME case studies from an emerging economy like India rang-
ing from discrete to process manufacturing, products to services, and tactical to
strategic implementations,
• Push Lean-driven results to new frontiers by integrating with other operations
excellence initiatives such as sustainability, six sigma, Industry 4.0, theory of
constraints (ToC), and time-driven activity-based costing (TdABC),
• Discuss the factors that support, or diffuse the Lean momentum based on inter-
actions with owners and senior managers of SMEs who have implemented
Lean, and
• Integrate supply chain resilience and Lean management, and present lessons on
how to handle disruptions, a dire need of the hour for SMEs.

The philosophy of the book is distilled directly from the interactions with the
original Kaizen Sensei. The application of well-proven Lean concepts in large
automobile and engineering industries is adapted, tinkered and fine-tuned to suit
the needs and constraints of SMEs. This unique blend stems from the authors’ first-
hand experience as well as access to work done by their organization at over 150
organizations across industry sectors ranging from SMEs to large multinationals
across the world. Experience gained from multiple Lean transformations indicates
that the key to success is in understanding and adapting the philosophy to the
unique human and cultural aspects of each organization and region.
Preface xiii

At the tactical level, the book guides the reader through practical approaches
to deploying Lean. This includes how to set the context for Lean, diagnosing, and
framing a Lean improvement roadmap, approaches to implement the roadmap,
and what works best in which situation. (S) he will also learn from the failures
and successes of organizations that have already undertaken the Lean journey in a
variety of situations.

Organization of this book

The book consists of six modules organized along the typical Lean implementation
life-cycle phases.
The first module of the book (Chaps. 1 and 2) covers introduction of Lean
management to the reader with a brief background of Lean origination at Toyota,
its spread through the rest of the world, and an overview of the current state of
Lean across different industry sectors and sizes. The book then proceeds to set
the context on the imperative of adapting Lean management in the current volatile
environment – the value proposition of Lean implementation. The last chapter
of the first module discusses the core principles of Lean and a conceptual Lean
implementation framework incorporating the stages of stabilization, flow, pull and
the ideal of single piece flow. A brief outline of the key Lean concepts, tools and
techniques relevant to each stage is provided and then we end with the typical
performance metrices used to measure the success of Lean.
Module-II (Chaps. 3 and 4)provides an overview of SMEs covering their role
in economic development, challenges, constraints, government programs, and the
ownership and management styles. Previously published literature on Lean imple-
mentation in SMEs from various sources is reviewed. Important findings such as
identified barriers and enablers to Lean implementations are highlighted. Authors
own experiences in terms of how to overcome financial and non-financial barri-
ers are discussed to give the reader some creative ideas. This module concludes
with a generic framework to implement Lean in SMEs, and discusses typical
implementation model variants.
Module III (Chaps. 5–7) deals with the implementation of Lean and is as such
the core of the book. The first step is to diagnose business operations under
the Lean paradigm. Lean performance metrics such as quality, delivery, inter-
nal productivity metrics such as value-added ratio, people productivity, Overall
Equipment Efficiency (OEE), etc., are introduced to the reader here. Key tools,
benchmarks and techniques for diagnosing different industry sectors and their link-
ages to overall organizational financial performance is discussed. We move on
next to the project planning stage in which the improvement roadmap is framed
that maps business goals, improvement areas and priorities on to implementation
schedule. The two key components of Lean, creating flow and enabling flow, are
discussed with a detailed step by step practical approach for implementing them.
Softer aspects such as the role of champion, communication, feedback loop, and
experimentation in implementation effectiveness are discussed in this Module.
xiv Preface

Module IV (Chaps. 8 and 9) looks at how organizations can stabilize improve-


ments made in the implementation phase through the judicious use of standards,
elements of Autonomous and Planned Maintenance and deploying organization
wide 5S. Stabilization is the key to ensuring the gains from Lean are held over the
long term. Data shows that over half of all Lean implementations do not sustain the
momentum and fall back to old ways within two years of the original implemen-
tation. We track organizations through typical stages of Lean adoption and review
the reasons for lack of sustenance based on direct feedback from a cross section
of SME owners who have been on the Lean path. At the same time, we also look
at possible ways in which SMEs can overcome some of these hurdles and develop
a culture that not only sustains but also continuously improves its Lean status.
Lean is a journey towards perfection and is therefore an ongoing pursuit of
excellence. Module V (Chaps. 10 and 11) provides guidance to the readers with
suggestions on how other emerging management philosophies such as sustain-
ability, Six Sigma, theory of constraints, Industry 4.0, and TdABC integrate with
Lean principles and can be leveraged to achieve further improvements. The book
demonstrates the synergies and necessity of such integrated approaches. Further,
the module describes the increased relevance of Lean in the context of increased
supply chain uncertainties. Lean is known to make processes fast, while being
flexible and thereby enhance supply chain resilience. Solutions such as build-
ing redundant capacity adapted by larger enterprises may not be suitable options
for SMEs. Hence, flexibility-based supply chain resilience solutions that fit the
constraints of SMEs are proposed.
Throughout the book, caselets are provided to illustrate specific topics at rele-
vant places. These caselets are derived from the larger stories of successful Lean
implementations at a SMEs across a range of industry sectors to give readers a
wide perspective on the nuances of Lean implementation. Having gone through
various aspects of a Lean implementation project life cycle, readers will be able
to appreciate and learn from the overall social-cultural-economical context of the
complete case studies in Module VI. Range of business drivers and how Lean
facilitated these drivers are discussed in these cases. Some examples are, how to
grow rapidly, how to meet demand fluctuations, what is the best layout for a new
facility, how to perk-up the eroding profit margins and how to build a motivated
team – to name a few, are covered in detail.

Guide to the Case Studies Section

Over the last two decades, one of the authors has consulted for over a hundred
organizations, mostly Small and Medium Enterprises, helping them improve their
operational and business performance. These SMEs are from a diverse array of
sectors, ranging in turnover from micro, small, medium to large, located across
different regions of India, the Middle East and Southeast Asia and with varied
management styles. The cases presented in this book have been carefully selected
Preface xv

from among these so as to present the reader an experience of wide-ranging


implementations.
The following table summarizes the contours of the seven case studies detailed
in module VI, which will help the reader not only identify key take-aways but
also navigate through the cases to dwell deeper into any particular context of
interest. Each of these Lean stories is unique and throws light on different facets
of implementation and sustenance of Lean.
Each case details key aspects of the Lean journey undertaken by the SME and
showcases the application of relevant Lean tools and techniques in different sit-
uations. Readers are encouraged to refer to the appropriate section in the earlier
modules to refresh themselves about these concepts, tools, and techniques when
required.

Audience and Pathways

SMEs form the industrial backbone of the developing economies providing sig-
nificant employment opportunities and are a vital link in supply chains. SME
owners/entrepreneurs often hesitate to take the plunge into implementing Lean
management concepts primarily due to lack of awareness of benefits and a wrong
notion that implementation is either complicated or costly. Even those who are
aware of general principles of Lean are unable to relate them to their context. This
book is intended to be a practical guide on what to expect from Lean management
and covers Lean implementation steps systematically and effectively in the given
the backdrop of typical constraints of SMEs.
It details an approach that has been tested and fine-tuned at over a hundred
organizations, mainly SMEs, across the emerging economies of India, Southeast
Asia and the Middle East. The approach follows the typical flow of an implemen-
tation cycle and enables the reader to understand the imperative for Lean, how to
diagnose current operations, how to plan and deploy Lean and shows a path for
long-term sustenance. The overall focus of this book is therefore to bring about a
3600 perspective in Lean management implementation at SMEs, right from need
identification through ongoing sustenance.
Senior executives responsible for large manufacturing units and corporate offi-
cers looking at driving Lean across the supply chain would also benefit from the
book, given its implementation-oriented approach.
The primary audience for this book therefore are SME business owners, CEOs,
plant / production managers and professionals engaged in operational excellence
activities. This book would help them understand Lean philosophy and the business
benefits of deploying Lean as an operational strategy. The secondary audience are
researchers and students of operations management. Of special interest for them
would be the enumeration of key aspects for success or failure of Lean inter-
ventions in SMEs and the use of different approaches to Lean implementation in
different situations and industry sectors. How Lean fits in with other emerging
xvi

Table 1 Summary of full cases of lean implementation


Characteristic Company
PAN seeds Linkwell Liberty Toys manufacturer Gubba Pharma cluster GSV
Industry Agri products Electronics Services Consumer Warehousing Pharmaceutical OEM manufacturing
(hospitality) manufacturing
Product(s) Seeds Energy meters Restaurant services Toys Cold storage space Bulk drugs and Fabricated parts for
intermediates motors
Revenue category Medium Medium Micro Small Small Small Micro
Demand pattern Seasonal: peaks Regulatory product: Peaks at weekends Generally, peaks in Inflow and outflow Steady demand Linked to pump
twice a year contract-based and and holidays months leading to peaks at particular through the year OEM which is again
so fluctuates Christmas and new times of the year a seasonal industry
year
Customer Dealers and Government Public Global toy majors B2B (seed, pharma, Pharma majors Pump OEMs
distributors organizations (export) and food industries)
Management style Mix of first and Professionally run Run directly by first Professional Run directly by Mix of first and First-generation
second generation with first-generation generation management (owners second generation second generation owner-driven
owner support oversee finance and
strategy)
Workforce Contract labour Mix of company Mix of full-time Mix of company Contract labour Full-time Full-time employees
gangs + full-time employees and staff (chefs, etc.) employees and gangs + full-time employees only only
supervisors contracted workforce and casual workers contracted workforce supervisors
(housekeeping)
Supply chain Farmer Subcontractors Local groceries Component/RM Only infrastructure RM Sheet metal suppliers
aggregators [mported part suppliers suppliers suppliers—local
suppliers and import
Trigger for lean Systematic Low margins Growth potential Global customer To be world class General curiosity Rapidly rising
working pressure demand
Preface
Preface xvii

management philosophies and technological solutions such as ERP and indus-


trial IoT, theory of constraints, Sustainability also provides useful directions to
the academic and research community.
We hope readers find valuable take-aways from this book.

Secunderabad, India Ganesh Mahadevan


Hyderabad, India Kalyana C. Chejarla
Acknowledgements

The seeds for writing this book were planted when we were under “house arrest”
as the COVID-19 pandemic first broke out. Two months of sitting at home would
have been unbearable had we not chosen to spend this time in penning down case
studies of significant recent work. Reading and rereading some of the best books
on Lean and allied philosophies made us realize that we could still bring to the
world, new aspects and ways of looking at Lean under different paradigms.
One of the highlights of this book is the varied and detailed stories of Lean at
work in different industry sectors. This would not have been possible without the
willingness of the industry owners and their plant managers to share their stories.
We would like to thank Anshuman Marodia (Pan Seeds), Ms. Radha Rani and
Mohan Rao (Linkwell), Vishal Lalwani (Liberty Exclusive), Kiran and Prashanth
Gubba (Gubba Group), Chandrasekhar Reddy (GIZ), Srinivasan (Porus), Krishna
Chaitanya (A.R. Lifesciences), Vamsi Krishna (Fine Group), and Manish Gupta
(Rockwell) for allowing us to write about how Lean has worked at their respective
organizations. We would also like to acknowledge Mr. Vijay Rangarajan (GSV
Industries) for our case study from the Coimbatore Engineering Cluster Lean
implementation program.
Implementing Lean successfully in MSMEs in a developing economy has its
own set of challenges. The approach to changing the mindset is the key. I, Ganesh
Mahadevan, have personally learned how to drive Lean on the gemba under my
guru S. Dorairajan and through the teachings of Sensei (late) Dr. Gondhalekar (Dr.
G), both founders of Kanzen Institute. I am indebted to them for providing me the
opportunity to work with them and be able to develop this body of knowledge. A
special mention to my father Dr. E.G. Mahadevan without whose constant prodding
I would never have attempted to write a book and get it published.
And I, Prof. (Dr.) Kalyana C. Chejarla, am immensely thankful to all my col-
leagues and management at Institute of Management Technology (Hyderabad) for
providing an extremely inspiring and intellectually stimulating academic environ-
ment, without which it would have been unthinkable for me to undertake such
a writing project. Above all, I am especially grateful to almighty for whatever

xix
xx Acknowledgements

courage, intellect, and grit he graced upon me which is the true foundation of
everything worthy that I have ever endeavoured.

Ganesh Mahadevan
Prof. (Dr.) Kalyana C. Chejarla
Contents

1 The Imperative of Lean Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1


1.1 History of Lean Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Mass Production (1908–44) . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1.2 Toyota Production System (1945–75) . . . . . . . . . . . . . . . . . 3
1.1.3 Foray of Lean into USA (1975–2000) . . . . . . . . . . . . . . . . . 4
1.1.4 Diffusion of Lean into Rest of the World
(2000–2020) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 The Imperative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2 Lean Management Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Lean Management Philosophy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Lean Management Maturity Phases . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3.1 Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.2 Pull . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3.3 One-Piece Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3.4 Stabilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.4 Performance of Lean Enterprises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3 Lean Management in Small and Medium Enterprises . . . . . . . . . . . . . . 27
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.2 Overview of SMEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.1 Economic Role . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.2 Unique Constraints of SMEs . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2.3 Growth Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2.4 Barriers and Enablers to Lean in SMEs . . . . . . . . . . . . . . . 31
3.3 Resistances to Initiating Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3.1 Financial Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3.2 Non-financial Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
xxi
xxii Contents

4 Lean Implementation Methodology for SMEs . . . . . . . . . . . . . . . . . . . . . . 39


4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.2 Lean Implementation Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.3 Breaking Down the Barriers—Approaches to Initiating
Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.3.1 Rapid Process Improvement Approach . . . . . . . . . . . . . . . . 44
4.3.2 Problem-Solving Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5 Commencing the Lean Journey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.2 Time-to-Serve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.2.1 Measuring Time-to-Serve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.2.2 Assessing Output Capability . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.3 Cost-to-Serve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.3.1 Measuring Cost-to-Serve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.4 Assessment Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.5 Target Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
5.5.1 Growth-Oriented Targets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.5.2 Profitability-Oriented Targets . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.5.3 Employee Well-Being . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
6 Designing the Lean Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
6.2 Building a Roadmap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
6.2.1 Sequencing the Improvement Projects . . . . . . . . . . . . . . . . . 84
6.2.2 Motivating the Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.3 Developing the Roadmap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
6.3.1 Theme-Based Roadmaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
6.4 Preparing for Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
6.4.1 Focused Improvement Workshops . . . . . . . . . . . . . . . . . . . . . 93
6.4.2 Post-workshop Reviews and Handholding . . . . . . . . . . . . . 96
6.5 Organization Structure for Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
7 Implementing Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
7.2 Begin with Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
7.3 Understanding Basic Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
7.4 Key Consideration for Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
7.5 Creating Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
7.5.1 Stage 1—Refining Process Flow Design . . . . . . . . . . . . . . 106
7.5.2 Stage 2—Redesigning the Layout . . . . . . . . . . . . . . . . . . . . . 109
7.5.3 Stage 3—Implement Redesigned Layout . . . . . . . . . . . . . . 112
7.5.4 Stage 4—Run and Validate Flow . . . . . . . . . . . . . . . . . . . . . . 112
Contents xxiii

7.6 Pull Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113


7.6.1 Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
7.6.2 Storage or Material Handling Containers . . . . . . . . . . . . . . 115
7.6.3 Electronic Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.6.4 FIFO Lanes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.7 Value Addition Must Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
7.7.1 Product is Fixed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
7.7.2 Person is Fixed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
7.8 Flow in Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
7.9 Enabling Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.10 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
8 Stabilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
8.2 Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
8.3 Standard Operating Procedures (SOP) and Work
Instructions (WI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
8.4 Autonomous Maintenance (AM) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
8.5 Planned Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
8.6 5S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
8.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
9 Sustaining Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
9.2 Sustenance Stages of Lean Adoption . . . . . . . . . . . . . . . . . . . . . . . . . . 154
9.2.1 Stage I: Tried and Failed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
9.2.2 Stage II: Successful Pilot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
9.2.3 Stage III: Standalone Intervention . . . . . . . . . . . . . . . . . . . . . 157
9.2.4 Stage IV: Enhanced Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
9.2.5 Lean Thinking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
9.3 Factors Affecting Lean Sustenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
9.3.1 People Related . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
9.3.2 Scaling Across . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
9.3.3 Creating Headroom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
9.3.4 Making Own Lean Operating Model . . . . . . . . . . . . . . . . . . 160
9.3.5 Periodic Lean Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
9.3.6 Top Management Commitment . . . . . . . . . . . . . . . . . . . . . . . 161
9.4 Empirical Study of Lean Sustenance . . . . . . . . . . . . . . . . . . . . . . . . . . 161
9.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
10 Beyond Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
10.2 Sustainability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
10.2.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
xxiv Contents

10.2.2 Integration with Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168


10.3 Six Sigma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.3.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.3.2 Integration with Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
10.4 Theory of Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
10.4.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
10.4.2 Integration with Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
10.5 Industry 4.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
10.5.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
10.5.2 Integration with Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
10.6 Time-Driven Activity-Based Costing (TdABC) . . . . . . . . . . . . . . . . 175
10.6.1 Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
10.6.2 Integration with Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
10.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
11 Leveraging Lean to Tackle Uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
11.2 Definition of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
11.2.1 Sources of Disturbance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
11.2.2 Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
11.3 Recent Supply Chain Disruptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
11.3.1 COVID-19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
11.3.2 Semiconductor Manufacturing Factory Fire . . . . . . . . . . . 187
11.3.3 Brexit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
11.3.4 Suez Canal Blockade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
11.3.5 Drought in Taiwan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
11.3.6 The Texas Freeze . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
11.4 Lean: Counter to Uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
11.4.1 Fire at Aisin Seiki Plant (1997) . . . . . . . . . . . . . . . . . . . . . . . 190
11.4.2 Strike at the US West Coast Ports (2002) . . . . . . . . . . . . . 191
11.4.3 Manufacturing Problems at Freescale (2005) . . . . . . . . . . 191
11.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192

Case Study 1: Seeds of Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193


Case Study 2: Business Transformation Through Lean . . . . . . . . . . . . . . . . . . 211
Case Study 3: The Lean Restaurant—Serving Customers Effectively . . . 247
Case Study 4: Lean Design for New Product Manufacturing . . . . . . . . . . . . 271
Case Study 5: Lean at Gubba Cold Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Contents xxv

Case Study 6: Applying Lean to Problem Solving


in the Pharmaceutical Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Case Study 7: GSV Industries—A Case study on Lean
Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
About the Authors

Ganesh Mahadevan is a partner and Lead Consultant at Kanzen Institute Asia


Pacific Pvt. Ltd. (www.kiap.in), India, an organization co-founded by former
associates of global Kaizen Sensei Masaaki Imai. With a Bachelor’s degree in
Metallurgical Engineering from the Indian Institute of Technology, Mumbai, and
a Masters in Business Management in operations management, his initial years
were spent in global majors in the aluminium industry including HINDALCO
and Pyrotek. After a four-year stint in Deloitte’s process consulting practice he
joined KIAP in 2007 and has since worked extensively on more than sixty com-
prehensive Lean transformations across a wide range of industry sectors, sizes, and
geographies. He is an empanelled Lean Expert with the Government of India and
successfully led lean implementations in several MSME clusters under the NMCP
scheme.

Kalyana C. Chejarla is an Assistant Professor at the Institute of Management


Technology, Hyderabad. He obtained his Ph.D. from the Indian Institute of Man-
agement Lucknow, completed his postgraduate studies from Indian Institute of
Management Ahmedabad and ICFAI Business School, and holds a B.Sc. (Engi-
neering) from National Institute of Technology Jamshedpur. His areas of interest
are operations, logistics and supply chain management, lean and quality manage-
ment, and business process management. Before moving into academics in 2014,
he has worked for about 18 years in the supply chain domain at globally renowned
organizations such as Tata Motors, HCL Technologies, Tata Consultancy Services
and Dell International Services, working extensively with clients in US and Japan.

xxvii
List of Figures

Fig. 2.1 Self-reinforcing virtuous cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10


Fig. 2.2 Lean implementation maturity stages . . . . . . . . . . . . . . . . . . . . . . . 13
Fig. 2.3 Wastes contributing to higher lead time . . . . . . . . . . . . . . . . . . . . . 17
Fig. 2.4 Reduction of wastes reducing the lead time . . . . . . . . . . . . . . . . . 17
Fig. 3.1 Favourable and unfavourable (to Lean) characteristics
of SMEs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Fig. 4.1 Lean implementation pyramid at large enterprises . . . . . . . . . . . 40
Fig. 4.2 Toy manufacturing process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Fig. 5.1 Different types of material wastes . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Fig. 5.2 VA and Muda times for human resources . . . . . . . . . . . . . . . . . . . 65
Fig. 5.3 Typical space-related operating costs . . . . . . . . . . . . . . . . . . . . . . . 67
Fig. 5.4 Vicious cycle of layout and material storage . . . . . . . . . . . . . . . . 68
Fig. 5.5 Material flow diagram for an aluminium products unit . . . . . . . 69
Fig. 5.6 Growth-oriented Lean program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Fig. 6.1 Creating a Lean roadmap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Fig. 6.2 Rockwell manufacturing process flow . . . . . . . . . . . . . . . . . . . . . . . 88
Fig. 6.3 Rockwell: improved clinching operation . . . . . . . . . . . . . . . . . . . . 88
Fig. 6.4 Focused improvement workshop . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Fig. 6.5 Cross-functional team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Fig. 7.1 Flow for a deep freezer manufacturing unit . . . . . . . . . . . . . . . . . 102
Fig. 7.2 Process design for a “runner” product . . . . . . . . . . . . . . . . . . . . . . 107
Fig. 7.3 Biscuit manufacturing—Trolley Kanban . . . . . . . . . . . . . . . . . . . . 116
Fig. 7.4 FIFO-based pull of pizzas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Fig. 7.5 Inventory is analogous to water in the sea . . . . . . . . . . . . . . . . . . . 124
Fig. 7.6 The heart of the refrigerator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Fig. 8.1 The staircase of continuous improvement . . . . . . . . . . . . . . . . . . . 134
Fig. 8.2 Revising process standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Fig. 8.3 Standardization tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Fig. 8.4 Process for milk solution preparation . . . . . . . . . . . . . . . . . . . . . . . 137
Fig. 8.5 Converting standards to sustainable practices through
WIs and visual standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
Fig. 8.6 The supervisor walk! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Fig. 8.7 Steps to implement AM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Fig. 8.8 5S implementation cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
xxix
xxx List of Figures

Fig. 8.9 5S Assessment score card—LE cluster . . . . . . . . . . . . . . . . . . . . . . 150


Fig. 9.1 Sustenance stages of Lean adoption . . . . . . . . . . . . . . . . . . . . . . . . 155
Fig. 9.2 Post-implementation improvements at the lean cluster . . . . . . . 162
Fig. 10.1 Schematic of initiatives complementing Lean . . . . . . . . . . . . . . . . 166
Fig. 10.2 IIoT solutions for different stages of Lean
implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Fig. 10.3 Revenue generation and cost accumulation
in organizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
Fig. 11.1 Sources of disturbance faced by firms . . . . . . . . . . . . . . . . . . . . . . 181
Fig. 11.2 Strategies to be adopted by firms . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Fig. 11.3 World uncertainty index (WUI) on the raise . . . . . . . . . . . . . . . . . 186
Fig. C1.1 Annual cycle of seed process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
Fig. C1.2 Broad roadmap for Lean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
Fig. C1.3 Organization structure—Bardhaman plant . . . . . . . . . . . . . . . . . . . 197
Fig. C1.4 AM implementation schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
Fig. C1.5 Plant process sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
Fig. C1.6 Seed grading and packing process flow . . . . . . . . . . . . . . . . . . . . . 202
Fig. C1.7 Cycle time study of semi-automatic line . . . . . . . . . . . . . . . . . . . . 203
Fig. C1.8 Improvement in Bori filling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
Fig. C1.9 Sunning process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Fig. C1.10 Yard marking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Fig. C2.1 The material and information flow of current process.
Notes Units 3 and 4 make their own schedules based
on the monthly plan requirement for assembly units . . . . . . . . . 214
Fig. C2.2 VSM of existing process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Fig. C2.3 Two-person cell workstation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Fig. C2.4 Earlier layout of the overall production unit . . . . . . . . . . . . . . . . . 219
Fig. C2.5 Simplified layout and improved work processes . . . . . . . . . . . . . 220
Fig. C2.6 High-level material flow between own production units
and subcontractors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
Fig. C2.7 High-level VSM of mechanical units . . . . . . . . . . . . . . . . . . . . . . . 223
Fig. C2.8 Earlier soldering process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
Fig. C2.9 Parallel processing in soldering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Fig. C2.10 Typical VSM of leaded PCB subcontractor . . . . . . . . . . . . . . . . . . 226
Fig. C2.11 a Earlier (above) and b modified (below) layout
of the SMT line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Fig. C2.12 VSM of name plate preparation process . . . . . . . . . . . . . . . . . . . . . 232
Fig. C2.13 Future state VSM of name plate preparation process . . . . . . . . . 233
Fig. C2.14 Operations analysis of bending process . . . . . . . . . . . . . . . . . . . . . 234
Fig. C2.15 Operations analysis of pad printing process . . . . . . . . . . . . . . . . . 235
Fig. C2.16 a Earlier printing room layout. b Modified printing
layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
Fig. C2.17 MRP run and execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
Fig. C2.18 Lean organization chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Fig. C2.19 Declining inventory month-on-month . . . . . . . . . . . . . . . . . . . . . . . 246
List of Figures xxxi

Fig. C3.1 Guest, food, and service staff flows in restaurant


operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
Fig. C3.2 Process flow of current operations . . . . . . . . . . . . . . . . . . . . . . . . . . 252
Fig. C3.3 Existing operations flow and layout . . . . . . . . . . . . . . . . . . . . . . . . . 253
Fig. C3.4 Modified operations flow and layout . . . . . . . . . . . . . . . . . . . . . . . . 254
Fig. C3.5 Existing layout and staff movement . . . . . . . . . . . . . . . . . . . . . . . . . 255
Fig. C3.6 Redefined responsibilities of coordination team . . . . . . . . . . . . . . 256
Fig. C3.7 Post-improvement staff movement . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Fig. C3.8 Value stream of pizza making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
Fig. C3.9 a Earlier (Left) and b improved (Right) layout for live
kitchen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Fig. C3.10 FIFO-based pull of pizzas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
Fig. C3.11 Food preparation process flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Fig. C3.12 Stores layout and an example kit preparation for one
Indian kitchen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
Fig. C3.13 Revised layout (blue dotted line indicates repacking
into small packs which is done during non-peak hours) . . . . . . 265
Fig. C3.14 Preparation cell—vegetables cutting . . . . . . . . . . . . . . . . . . . . . . . . 269
Fig. C4.1 Typical process flow chart for toy manufacturing . . . . . . . . . . . . 274
Fig. C4.2 Rough future state VSM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Fig. C4.3 Final layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Fig. C4.4 Booth allocation and flow balancing in painting process . . . . . 283
Fig. C4.5 Cellular layout at printing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Fig. C4.6 Separator table at carton packing . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
Fig. C4.7 Assembly and packing layout and material flow . . . . . . . . . . . . . 287
Fig. C4.8 Final planning model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
Fig. C4.9 Fast ramp up for from project execution to full
production output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
Fig. C5.1 Plant organization structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Fig. C5.2 Top view of storage layout in a typical floor . . . . . . . . . . . . . . . . 295
Fig. C5.3 Elevator layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Fig. C5.4 Earlier and improved truck turnaround process . . . . . . . . . . . . . . 300
Fig. C5.5 SOP of material inward process . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Fig. C6.1 Current state VSM for product Pavest . . . . . . . . . . . . . . . . . . . . . . 314
Fig. C6.2 Phases of improvement projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
Fig. C6.3 Process stages of Pavest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
Fig. C6.4 Three stages after improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
Fig. C6.5 Grinding and weighing—current state . . . . . . . . . . . . . . . . . . . . . . 318
Fig. C6.6 Improved condition in dry milling . . . . . . . . . . . . . . . . . . . . . . . . . . 320
Fig. C6.7 Reactor time variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
Fig. C6.8 5S—Before and after in stores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
Fig. C7.1 Value stream map of motor guard line . . . . . . . . . . . . . . . . . . . . . . 330
Fig. C7.2 Improvement in stamping OEE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
Fig. C7.3 Motor guard manufacturing process . . . . . . . . . . . . . . . . . . . . . . . . 335
Fig. C7.4 Material handling in cages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
xxxii List of Figures

Fig. C7.5 Material flow in motor guard manufacturing . . . . . . . . . . . . . . . . 337


Fig. C7.6 Cellular layout for two products . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Fig. C7.7 Chute controlling the WIP between processes . . . . . . . . . . . . . . . 339
Fig. C7.8 Machine alignment to enable single worker operation . . . . . . . . 339
List of Tables

Table 3.1 Differences between SMEs and Large Enterprises


(adopted from Berisha and Pula, 2015) . . . . . . . . . . . . . . . . . . . . 29
Table 3.2 Challenges and constraints faced by SMEs . . . . . . . . . . . . . . . . 31
Table 3.3 Unique characteristics of high growth SMEs . . . . . . . . . . . . . . . 31
Table 4.1 Methodology for implementing Lean at SME . . . . . . . . . . . . . . 42
Table 5.1 Takt time calculation for bulk drug manufacturer . . . . . . . . . . . 58
Table 5.2 ECT calculation for bulk drug manufacturing . . . . . . . . . . . . . . 59
Table 5.3 Key resources and their productivity measures . . . . . . . . . . . . . 63
Table 5.4 Shrimp processing diagnostic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Table 5.5 Appropriateness of different tools and metrics
for different industries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Table 5.6 Strategies to increase capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Table 5.7 Examples of profitability improvement targets
for a crockery unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Table 5.8 Comprehensive Lean target table for an LED
manufacturing unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Table 6.1 Impact of reducing 3 M’s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Table 6.2 Theme-based roadmap for an edible oil manufacturer . . . . . . 92
Table 7.1 Product–Quantity (P–Q) analysis and process design . . . . . . . 106
Table 7.2 P-P matrix for “Repeater” products . . . . . . . . . . . . . . . . . . . . . . . 108
Table 7.3 Daily demand and operation cycle times for one group . . . . . 108
Table 7.4 JIT rules for standard inventory points . . . . . . . . . . . . . . . . . . . . 114
Table 7.5 Towering heights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Table 7.6 List of lean tools and the flow impediments they
address . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Table 7.7 The Incomplete Kit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Table 8.1 Ford’s CANDO and Japanese 5S . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Table 8.2 5S related to personal change, lean improvement,
and standardization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Table 10.1 Various studies focusing on sustainability in SMEs . . . . . . . . . 167
Table 10.2 Lean tools and supporting technology solutions
mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Table C1.1 OEE computation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Table C1.2 Comparison of packing line performance . . . . . . . . . . . . . . . . . . 204
xxxiii
xxxiv List of Tables

Table C1.3 Throughput time of sunning process . . . . . . . . . . . . . . . . . . . . . . 206


Table C1.4 Comparison of truck TAT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Table C1.5 Plant performance comparison (paddy
seed—Bardhaman plant) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Table C2.1 Process goals for lean intervention . . . . . . . . . . . . . . . . . . . . . . . . 215
Table C2.2 Comparison of performance of assembly line
and two-person cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Table C2.3 Operations analysis of calibration process . . . . . . . . . . . . . . . . . 221
Table C2.4 Comparison of calibration process
before and after improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Table C2.5 Operations analysis of testing process . . . . . . . . . . . . . . . . . . . . . 224
Table C2.6 Summary of improvements in mechanical unit
assembly subcontractors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Table C2.7 Operations analysis of PCB testing process . . . . . . . . . . . . . . . . 227
Table C2.8 Improvements in output of the PCB subcontractors
after Lean implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
Table C2.9 Initial performance of name plate preparation process . . . . . . 233
Table C2.10 Why-why analysis of process time variation
(component removal) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
Table C2.11 Improvement in kit collection time by SCs . . . . . . . . . . . . . . . . 239
Table C2.12 Why-why analysis for poor OTIF of kits . . . . . . . . . . . . . . . . . . 240
Table C2.13 Consolidated benefits of Lean implementation . . . . . . . . . . . . . 245
Table C3.1 Why-why analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
Table C3.2 Existing and improved cutlery replenishment processes
in the buffet area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
Table C3.3 Stock availability-related issues . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Table C3.4 Storage and issues of Kits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
Table C3.5 BoM for example recipes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Table C3.6 Leftover management board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
Table C3.7 Work checklist for deserts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
Table C4.1 P–Q analysis with Takt time (for moulding) . . . . . . . . . . . . . . . 273
Table C4.2 P-P matrix for “model blue toy” . . . . . . . . . . . . . . . . . . . . . . . . . . 276
Table C4.3 Moulding machine requirements for ‘blue toy’ . . . . . . . . . . . . . 277
Table C4.4 Customer orders in a typical week . . . . . . . . . . . . . . . . . . . . . . . . 285
Table C6.1 Key differences between discrete and process
manufacturing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
Table C6.2 VSM data for Product X . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
Table C6.3 Distillation monitoring data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
Table C6.4 Root causes and solutions to reduce dust spillage . . . . . . . . . . 319
Table C6.5 Summary of improvements achieved by the bulk drug
pharma SME cluster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Table C7.1 Phase-wise implementation roadmap . . . . . . . . . . . . . . . . . . . . . . 331
Table C7.2 Changeover activity allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
Table C7.3 Stamping production—output trend . . . . . . . . . . . . . . . . . . . . . . . 335
List of Tables xxxv

Table C7.4 Material movement between processes . . . . . . . . . . . . . . . . . . . . 336


Table C7.5 Motor guard production trend . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
Table C7.6 Lean organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
The Imperative of Lean Management
1

1.1 History of Lean Management

Two world events played a pivotal role in the development and adaption of Toy-
ota Production System (TPS), a precursor to Lean management. First, the Second
World War ravaged Japan, led to a period of economic distress that became a
source for the creation of ingenious, innovative, humane, sustainable and in many
ways, a common-sense approach to production by Toyota. For instance, general-
purpose machinery, rather than the specialized large machinery favoured by the
then mass manufacturers like Ford allowed production of smaller batches of larger
varieties of components. Or, closer supplier-ties enabled just-in-time supplies and
thereby enabling quicker product changeovers. Further, smaller families, smaller
parking lots, and constrained budgets of customers have led Toyota to build smaller
and fuel-efficient cars, as against the larger models prevalent in the western world.
The goal was then to deliver small numbers of a variety of cars at low costs to
the fragmented Japanese market post the devastation of the Second World War.
Reduced purchasing power in the post-war years meant cars had to be affordable
even for those few people who could still think of owning a car. At the same time,
there needed to be some choice in terms of product to satisfy different needs.
Hence, right from its origin, Toyota’s manufacturing has been oriented to deliver-
ing what the customer wants in the shortest possible time and lowest cost without
sacrificing process efficiencies. By 1970s, Toyota had established their own manu-
facturing philosophy and systems that collectively came to be known as the Toyota
Production System (TPS) making Toyota one of the most productive automobile
companies in the world. Up to this point in time, Toyota was still operating mainly
in the Japanese and Asian markets, and TPS had not yet come into the sights of
the western world.
Second, the decade-long oil embargo placed by OPEC countries on the west-
ern world starting 1973, led to accelerated adaption of Lean outside of Japan.
The embargo induced hike of oil prices for an extended period of time dampened

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 1
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_1
2 1 The Imperative of Lean Management

the demand for automobiles, especially the large gas guzzlers made by western
manufacturers. This led to rise of demand for the relatively smaller and more
fuel-efficient cars made by Toyota. The rest of the world started to take notice of
Toyota, and by the 1980s, researchers and industrialists were realizing that Toyota
continued to make profits while competitors in their own industry were struggling
to contain losses. Toyota was open to visits by outsiders to their plants to study
and understand TPS, and the resultant research gave birth to what was called Lean
thinking. In the successive decades through 2000 and till date, hundreds of organi-
zations have implemented TPS or Lean with different degrees of success. During
the same period, Toyota has shot past to become the world’s number one automo-
tive manufacturer. The main reason for this differing degree of success from TPS in
other organizations is that most of them have tried to implement specific concepts
or individual tools, leaving out the larger spirit of TPS philosophy that includes
elements of organizational culture, people management, and top management com-
mitment. Fairly soon, the momentum lost steam in most of these firms, which then
slipped back to the paths of least resistance. Organizations regressed to old ways
of doing things, giving out excuses such as TPS is a Japanese thing or that it is
suitable for low variety high volume business or meant only for automobile indus-
try. TPS has sustained in those few organizations that have deeply internalized the
philosophy into their organizational routines and made it as their own production
system.
An understanding of the evolution of industrial production over the last hundred
years or so would help us better appreciate the place of Lean in the current era.
The developments can be grouped into four main eras which are detailed in the
following sections. Readers can refer Samuel et al. (2015) for a table of specific
events and developments related to Lean during the twentieth century.

1.1.1 Mass Production (1908–44)

Industrial scale (mass) manufacturing was pioneered by Henry Ford for the pro-
duction of Model-T car through the world’s one of the early “moving assembly
line” method. Model-T, launched in 1908, had a successful run of about 19 years
without any major changes either to the product or process resulting in a very low
cost per car. Till then, most manufacturing was craft based and made using cus-
tom parts, resulting in high unit cost. Principles of assembly line production were
extended to other industries as well, and functional managers and industrial engi-
neers held the managerial control within organizations. Ford expanded to Europe
with the assembly line production methods to avoid the substantial logistics costs
(in exporting to Europe market). However, the management burden of a large ver-
tically integrated firm could not provide enough leeway for Ford to drive product
variety or component innovation.
Alfred Sloan’s corporate structure design at General Motors, with individ-
ual divisions held accountable for financial performance from late 1920s, drove
organizational efficiency and innovation. General managers with overall Strategic
1.1 History of Lean Management 3

Business Unit (SBU) responsibility were in their prime demand during this era.
General Motors led the automobile world for 40 years till 1960s. Both Ford and
Sloan systems are referred to as mass production systems, and both suffered from
poor product quality. However, at this point in time, the Japanese products were
also known to be of below par quality albeit at a higher cost.

1.1.2 Toyota Production System (1945–75)

Toyota Motor Company was formed in 1937 under the aegis of Kiichiro Toyoda,
who started the company mostly with simple machinery. In 1950, Eiiji Toyoda,
Kiichiro’s cousin became the Managing Director of Toyota Manufacturing, and
he travelled to USA and Germany to learn the prevailing production and manage-
ment processes. He was keen to implement mass production concepts at Toyota,
but found it infeasible due to the prevailing capital and market constraints in
Japan. However, the delegation which included Taiichi Ohno learnt the notion of
producing to takt time from their Germany visit (Holweg, 2007).
Ohno who is widely considered as the mastermind behind Toyota Production
System (TPS) started his career as an engineer at the Toyoda automatic looms
in 1932. It is here that the origins of TPS lay, where Ohno experimented with
production autonomation (Jidoka) and learnt its benefits. An example of autono-
mation popularly cited in the literature is that of an automated loom, which stops
production as soon as a thread breaks, instead of continuing to produce bad prod-
uct automatically. Ohno moved to the automotive business in 1943. Autonomation
was the underlying thread (no pun intended) throughout the development of TPS
from 1945 to 75. The beginnings of TPS are traced to 1948, when Ohno started
with just-in-time withdrawals and removed intermediate storages (Ohno & Bodek,
2019). While cell-based production was already in place, a visit to USA during
1956 prompted Ohno to devise Kanban replenishment throughout Toyota supply
chain in the following years. The lot size reduction gained impetus when Shi-
geo Shingo, who was brought on board in 1955, developed a formal methodology
to reduce changeover times of the large presses that were used for sheet metal
stamped parts (Dillon & Shingo, 1985). The smaller lots across the supply chain
then paved the way for yet another masterstroke, production levelling (Heijunka)
by Ohno. Hiejunka helped Toyota to truly match supply with actual customer
demand across models and variants.
American industry/government, as part of post-war support, sent quality gurus
such as Deming to help Japanese industry improve its quality. Japanese industry,
including Toyota, wholeheartedly welcomed and learnt from this intervention, so
as to be competitive globally. TPS tools like error proofing (poka-yoke) and inter-
nal supplier and customer relationships with respect to product quality came out
of this movement and served to strengthen the Jidoka pillar of TPS.
4 1 The Imperative of Lean Management

1.1.3 Foray of Lean into USA (1975–2000)

As stated before, a long oil embargo by OPEC on the western world led global
automobile manufacturers to start looking closely at TPS as a means to bring-
ing down the operating cost. Growing trade deficit, increasing logistics costs, and
strengthening of Yen led Toyota and other Japanese manufacturers to set up both
greenfield and brownfield plants in north America. One such notable brownfield
venture was the New United Motor Manufacturing, Inc. (NUMMI) at Fremont,
California, in collaboration with General Motors. In 1979, Massachusetts Institute
of Technology and few other universities and industry sponsors created a research
program called International Motor Vehicle Program (IMVP) with an intention to
study the differences in production systems at various automobile factories around
the world. This research was instrumental in disseminating the notion of TPS man-
ufacturing as Lean manufacturing to the rest of the world via research articles
(e.g. Krafcik, 1988), and the popular book “The Machine That Changed the World”
(Womack et al., 2007, originally published in 1990). The idea behind phrasing TPS
as “Lean” by the IMVP researchers is to reflect the fact that TPS uses much lesser
resources (people, equipment, space, etc.,) than the mass-producing counterparts.
Further, in 1996, Womack and Jones have packed the Lean concept, approach,
and implementation in an action-oriented book titled, “Lean Thinking” a via set of
diverse case studies.
During this period, many other books were written and a number of consultants
also started to offer Lean implementation services. The adoption of Lean was faster
in the USA, followed by Europe. Further, quality improvement programs in line
with Deming’s quality award in Japan such as Malcolm Baldrige National Quality
Award (MBNQA) have heightened the necessity of improving production pro-
cesses, especially in the USA. Womack and Jones (1996) document the successes
of a few enterprises such as Wiremold, Lantech, Porsche, and Pratt & Whitney on
account of Lean adoption in their organizations. Just-In-Time (JIT) was the fore-
most among the lot in terms of absorption, perhaps because of its highly visible
effects. This also required manufacturers to take a closer look at their supplier
relationships, resulting in lowering of manufacturing Lead times in the USA in
the early 1990. However, this inventory reduction trend began reversing from late
1990s with a steady increase in inventories of manufacturers, retailers, and whole-
salers, reflecting that the true root causes of higher levels of aggregate inventories
were not yet mastered by majority of industrial supply chains.
As India opened up its economy to foreign investors in 1991, there was an influx
of global automobile majors looking to source parts from low-cost Indian manu-
facturers. The quality standards forced progressive Indian organizations such as
the TVS group to start looking to implement Lean under the guidance of Japanese
Sensei (teacher).
1.2 The Imperative 5

1.1.4 Diffusion of Lean into Rest of the World (2000–2020)

The twenty-first century saw a lot of dissemination of Lean principles both to


remaining geographies and to different industry verticals. The diffusion of Lean
to the rest of the world and other sectors followed four different tracks (Samuel
et al., 2015). They are Lean as a generic version of TPS, as process improve-
ment toolkit, as an organizational ideology, and as an academic research area. The
first track of diffusion took the form in which various authors, publishers, speak-
ers, consultants, and other proponents of Lean wrapped the specifics of TPS (be
it tools, culture, or philosophy) with an overarching and a generalized approach
called Lean. Lean rubbed shoulders with its process improvement cousins such
as Six Sigma, Theory of Constraints, Business Process Reengineering, etc., in its
second track of diffusion. The third form of Lean diffusion has been broader in
the sense that (a) it moved beyond the production floors into boardrooms as an
organizational strategy or ideology, (b) service organizations have adopted Lean
principles as ardently as manufacturing firms, and (c) the shift of perspective from
cost reduction to value maximization among the adapters of Lean. The academic
world contributed to final form of diffusion in the form of research specifically in
the streams of operations management and organizational behaviour.
In their Shingo Research and Professional Publication award winning com-
pendium, McKinsey & Company describes how companies from different sectors
such as financial services (Ameritrade, RBS Citizens Financial Group, Axis Bank,
etc.), public services (Swedish Migration Board), and Telecom (TDC) have imple-
mented Lean principles to achieve superior organizational performances.1 From
its homestead of discrete manufacturing, Lean management has been adopted in
process industries such as food, steel, pharmaceutical, textile, chemical, paper,
and sugar (Panwar et al., 2015). In their study of Lean implementations in pub-
lic sectors in 26 countries, (Lukrafka et al., 2020) note that the UK leads with
26 implementations followed by the USA with ten implementations. Europe,
as a continent, leads with 48 implementations followed by Americas with 17
implementations. We discuss the adoption of Lean by SMEs in Chap. 3.

1.2 The Imperative

A Lean organization is characterized by its capability to satisfy customers on


quality, delivery, and price, through fast and flexible processes that quickly
respond to customer needs utilizing available resources effectively, thereby

1 https://www.mckinsey.com/~/media/mckinsey/industries/consumer%20packaged%20goods/

our%20insights/the%20consumer%20sector%20in%202030%20trends%20and%20questions%
20to%20consider/2014_lean_management_enterprise_compendium.pdf.
6 1 The Imperative of Lean Management

incurring the lowest possible cost through operating a visually managed and
self-regulating facility.

At the outset, Lean appears contrary to conventional operational strategy that typi-
cally pitches “cost-to-serve” and “time-to-serve” as conflicting objectives and that
one should be optimized at the cost of another. Lean management shows us the
way to improve on both the objectives, as we shall see in the coming chapters.
In particular, the capability to reach higher levels of output not only without any
capital investments but also at a much lower operating cost (materials, energy,
manpower, etc.,) is highly attractive to the Small and Medium-sized enterprises
(SME). The above stated definition of Lean can be distilled into the following two
key elements.

• Flexibility of the value stream which is the set of processes that deliver the
value in terms of product and/or service to the customer and
• Productivity of the resources used in delivering this value— man, material,
energy, machinery, or infrastructure.

The important component of Lean management is the presence of a definite


goal, utopian it may be, that provides unfaltering direction to the organization
at all decision-making situations without any ambiguity. Managers and executives
who embark on Lean transformation think about three fundamental business issues
that should guide the entire organization. These are:

• Purpose: What customer problem will the enterprise solve to achieve its own
purpose of prospering?
• Process: How will the organization assess the value stream(s) to make sure each
step is valuable, capable, available, adequate, flexible, and that all steps flow in
synchronous manner
• People: How can the organization ensure there is someone responsible for
continually evaluating each value stream in terms of business purpose and
Leanness? How can everyone touching the value stream be actively engaged
in operating it correctly and continually improving it?

Lean management is a world view with a broader perspective involving the


interests of stakeholders starting from customers all the way through to suppliers.
Lean also takes a longer-term perspective as against short term, for example, in
spending time and effort on root cause elimination, even at the cost of seemingly
temporary revenue loss. This perspective always looks to improve business pro-
cesses, develop standards, and implement systems keeping in mind the following
realities that are an inherent part of the world we operate in.

• Uncertainty of future: Future is always uncertain, and the farther one looks at,
more uncertain it becomes. Forecasts, howsoever sophisticated they are, do not
1.3 Summary 7

reduce the uncertainty. Plans generated using the forecasts do not become a
reality.
• Process variability: Processes are variable. Excess capacity or stock building
does not reduce the variability, instead hides the immediate effect of it.
• Everyone in the supply chain needs to make money. Therefore, in the long run, it
is both convenient and profitable to promote transparency, trust, and fair share
among all the supply chain members.
• Resources is a broad term and includes space, people’s effort, materials,
machinery, and utilities. Organizations have a responsibility to acquire, use,
and dispose these resources in the most efficient manner.
• Genuine humility leads to respect for everyone, customers, employees, and com-
petitors, which then leads to accept and work with reality at all levels within
the organization.
• The only constant in the world is change, and so create an organization that
expects change and rapidly learns from it with improvement as the focus.
• Organizations last longer than individual. One needs to think beyond oneself
and build a system that is self-perpetuating rather than one propelled by an
individual or a team.

In the last three decades, many books, articles, and case studies have been
written and published covering Lean philosophy, concepts, popular tools, and tech-
niques and comparing Lean with other management methodologies such as TOC
or Six Sigma. This book is not so much about hair-splitting definitions of Lean
and allied terms, but about presenting solutions that SMEs can use. To that effect,
we integrate relevant notions, principles from other management, and technologi-
cal domains in various chapters and case studies. Lean management is not about
east versus west or Toyota versus GM, or Lean versus some other management
philosophy. In its true spirit, and the way we deal with it in this book is that, it is a
culmination of learnings by and from a multitude of organizations that believe in
providing a worthy goods/service to its customers over their lifetimes, by efficient
and sustainable use of all the resources at its disposal and by endlessly enhancing
the human learning and adaptation potential of its employees. It is a truism that
profits (have to) follow.

1.3 Summary

This chapter traces the roots of Lean management to Toyota Production sys-
tem which was developed in response to the resource-constrained environment
in Japan, post-Second World War. As Toyota mastered the art of producing more,
that accurately meets the needs of the customers using least resources, the western
world plunged into resource-constrained environment in 1970s due to a prolonged
oil embargo by OPEC. This paved way for the American companies to take a
leaf from TPS handbook. The world industrial production systems have evolved
from mass production to mass customization to today’s technology led systems.
8 1 The Imperative of Lean Management

We looked at how Lean evolved through these systems and across geographies.
Finally, we conclude by presenting how Lean maps to the current realities bet-
ter than any other production system, and thus a clarion call for all firms, and
especially SMEs to seriously consider Lean implementation.

References

Dillon, A. P., & Shingo, S. (1985). A revolution in manufacturing: The SMED system. CRC Press.
Holweg, M. (2007). The genealogy of lean production. Journal of Operations Management, 25(2),
420–437.
Krafcik, J. F. (1988). Triumph of the lean production system. Sloan Management Review, 30(1),
41–52.
Lukrafka, T. O., Silva, D. S., & Echeveste, M. (2020). A geographic picture of Lean adoption in
the public sector: Cases, approaches, and a refreshed agenda. European Management Journal,
38(3), 506–517.
Ohno, T., & Bodek, N. (2019). Toyota production system: Beyond large-scale production. Produc-
tivity Press.
Panwar, A., Nepal, B. P., Jain, R., & Rathore, A. P. S. (2015). On the adoption of lean manufactur-
ing principles in process industries. Production Planning & Control, 26(7), 564–587.
Samuel, D., Found, P., & Williams, S. J. (2015). How did the publication of the book the machine
that changed the world change management thinking? Exploring 25 years of lean literature.
International Journal of Operations & Production Management.
Womack, J. P., & Jones, D. T. (1996). Lean thinking: Banish waste and create wealth in your
corporation. Simon & Schuster.
Womack, J. P., Jones, D. T., & Roos, D. (2007). The machine that changed the world: The story
of lean production–Toyota’s secret weapon in the global car wars that is now revolutionizing
world industry. Simon and Schuster.
Lean Management Principles
2

2.1 Introduction

Lean management in its ideal form endeavours to develop a value delivery sys-
tem that continuously learns and evolves, to get closer to the possibility of getting
supply to exactly meet when, where, and how the demand occurs. As firms get
closer to this ideal in execution, they will observe that some of the assumptions,
especially related to forecasts, made in planning phase render themselves unnec-
essary. This will release the need for cushions in planning and make the firm more
responsive to changing market needs, resulting in reducing reliance on forecasts.
This creates a virtuous cycle (Fig. 2.1).
All the principles, tools, techniques, and other paraphernalia are created, mod-
ified, and dropped, if required, to support and sustain such a learning system.
Clearly, the end goal (match demand with supply accurately) is worthy of pursuit,
immutable, true north star, independent of the firms’ managers, and timeless and
provides an unambiguous direction whenever there is a decision dilemma. The
ideal is tirelessly pursued by steadily taking steps towards it. No sooner than a
firm achieves a milestone, there is a next milestone planned in the Lean direction
at the firm. In fact, the true mark of a Lean organization is that it is a learning
organization through cycles of continuous improvement (Mrugalska & Wyrwicka,
2017).
Despite being a logical, sustainable, common sense, and scientific approach,
Lean is usually shadowed by the overhang of mass production and human inertia
to change. A somewhat counter-intuitive (only because of our earlier conditioning)
insights such as avoid/reduce batch production add to the mystery in the SME
owner/manager’s mind. The most unfortunate and yet very common shortfalls in
the Lean implementations are the ones which were limited to some specific Lean
tools, sans the philosophy/long-term direction. While Lean management does use
a lot of tools, good Lean organizations are not fixated by tools themselves, for
they are just what the name suggests, tools! A tool is needed in a certain context

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 9
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_2
10 2 Lean Management Principles

Fig. 2.1 Self-reinforcing


virtuous cycle
Lean
Execution

Lean
Planning

and another one in another context. The smartness of the Lean organization lies in
choosing/adapting or even creating the right tool in right context. Lean is a philos-
ophy based on a paradigm—a way of looking at things. We provide an overview
of the foundational concepts of the Lean paradigm in this chapter.
The pursuit of Lean management may be characterized into four distinct, pro-
gressively maturing stages. The role of a manager in Lean paradigm is that of
a coach who helps the employees master the specific tools, but more importantly
build, sustain, and live the learning culture. This chapter describes the foundational
concepts behind the Lean management as appropriate for different maturity stages
and point out corresponding tools and techniques. Further, the chapter concludes
with a brief overview of performance improvements Lean enterprises worldwide
have and continue to achieve, providing a financial motivation for an SME to adopt
this path.

2.2 Lean Management Philosophy

In any organization, its working philosophy is embedded in its language,


behaviour, communication, shared beliefs, routines, cadence, and career paths.
The vision and mission statements alone cannot give the full picture. Different
organizations will have different philosophies which explain their business pro-
cedures and in turn influence employee thinking patterns. Culture can be viewed
as the explicit and implicit manifestation of organizational philosophy. Top-down
approaches such as mass production assume that not only managers and senior
executives know what is good for the organization in the long run, but they also
2.2 Lean Management Philosophy 11

know exactly how to execute these steps, which they instruct to their subordi-
nates. These approaches rely on narrowly defined department-level economies of
scale-based unit costs missing the non-value-added effort and time the product
undergoes. Further, such approaches miss on the possibility of leveraging the inge-
nuity of operators at the floor in improving processes and thereby inculcating a
sense of ownership in the success of the firm.
On the contrary, Lean management is a true long-term philosophy that is based
on the following tenets:

• Profits and growth (long term/sustainable) are a result of continuously and effi-
ciently matching demand with supply. Any other growth is unsustainable and,
therefore, offers no lasting organizational learning,
• Buffers (capacity, inventory, and lead time) are a result of poor understanding
and management of process and demand variation and not weapons to beat
variation. Therefore, a right approach is to incrementally understand and control
the variation, and critically question all buffers, and
• Finally, developing and adhering to the standards that reflect institutional-
ized and prevailing know-how (i.e. based on controlled experiments performed
on the shop floor by the operators). The standards provide launch pad for
next round of improvement initiatives and are readily changed as soon as an
improved method is tested successfully.

This grass-roots-level fundamental understanding of how the organization


makes money in the long run makes Lean organizations to have targets for efficient
matching of supply with demand, rather than short-term (quarter to quarter) prof-
its and growth. (Please see “Toyota Kata”, by Mike Rother for details on how a
seemingly harmless pursuit of short-term goals by a firm may put it on a path to its
deterioration). Lean enterprises have targets for variability reduction and produc-
tion cycle time matching to demand rate, rather than maintaining or even reducing
buffer levels.
Lean management flourishes in a collectivistic set-up, which is free of indi-
vidual agendas, idiosyncrasies, or fears. Lean management requires that not only
every employee is treated with respect, but is truly considered as a source of human
ingenuity to solve problems at the work area. It is universally accepted that in the
long run, one man wanting to grow at the expense of others, or an organization
wanting to grow at the expense of the customer (by making them pay for its own
inefficiencies) is unsustainable. Clearly, sustaining and surviving in the long run
requires not giving-in to short-term temptations such as quarterly sales goals and
increase profitability by firing employees. Decision-making (especially in the con-
text of improvement investments) according to Lean management philosophy is
thus based on the following basic principles:

• Only customer defines/indicates the value of product/service.


• A decision should align with the long-run vision and yet be efficient in the short
run in that order. This clearly implies that an efficient improvement proposal is
rejected if it is not in alignment with the long-term vision.
12 2 Lean Management Principles

• A decision should be based on the floor experimentation-led knowledge not on


hunch, not on consultant recommendation, not on Sensei instruction, and not
on automation company presentation.
• Between people and machines, people are a more valuable, learning, adap-
tive, and flexible resource, and thus the work allocation should reflect this
understanding.
• Documenting the current state of learning in the form of a standardization doc-
ument is a “must”, before undertaking any experimentation for improvement.
• Sunk costs (pre-existing capacity) do not justify creation of additional waste
(maximizing utilization), e.g. inventory without orders.
• Supplier–customer relation should be established at the smallest of each
handover. The objective of such unitary-level tight coupling is to ensure instan-
taneous transmission of feedback and implementation of corrective response.

2.3 Lean Management Maturity Phases

The end goal of Lean management is to create a supply system (production pro-
cess) that can dance synchronously in lockstep with demand (of product). The
variability in supply and demand needs to be understood, reduced, and responded.
One major source of variability is a high time to respond, a.k.a lead time. Longer
the time to respond, higher are the chances of a different reality emerging than
what was planned and hence a higher variability. Lead time consists of informa-
tion and material lead times. Taiichi Ohno had once famously said that the focus
of the Toyota Production System (TPS) is to reduce the time it takes to meet a cus-
tomer order from the time it is received. Lean’s foundation lies in TPS, and Lead
time reduction is a primarily goal. Shorter the lead time, faster is the response
to customer demand and more adaptable are the processes. The ideal state would
be one where the lead time is a bare minimum or just equal to the sum of the
processing times of all the processes required to make one unit of product.
Just as it takes time, practice, and a focused effort on the part of an ama-
teur dancer to become a maestro, even firms need to approach this ideal (perfect
matching of demand with supply) in a structured manner. Figure 2.2 presents a
conceptual loop of four stages of improvement which are not necessarily sequen-
tial. While adopting Lean, one can enter anywhere in this loop and exit after any
number of iterations, depending on the organizational need and strategy. An exam-
ple of a single iteration would be one in which one may take up a project in
flow/pull phases in order to demonstrate quick results to get the buy-in of rel-
evant stakeholders, as we shall see in Chap. 4. Although achieving one-piece
flow is the ideal, the recursive representation in Fig. 2.2 is to indicate that firms
move on to other supplier–customer binary links in their value chain for Lean
implementation. We describe these stages and provide an overview of the relevant
tools/principles for each stage in the following subsections. Implementation details
2.3 Lean Management Maturity Phases 13

Flow

Stabilization Customer Pull

One-piece
flow

Fig. 2.2 Lean implementation maturity stages

of the tools/principles are described in Modules III and IV, which deal with the
Lean implementation and stabilization, respectively.
The most important element of any of these improvement stages is their sole
focus on customer perspective, i.e. improvement is done “only when” the initia-
tive adds value to customer. This point cannot be overemphasized in the current
environment, where a number of organizational capital investments are fancy and
have a fragile customer basis. This emphasis on everything “revolving” around
the customer is why we have placed customer in the centre of the loop. Pioneer-
ing Lean thinkers have established that, from a customer’s perspective, there are
three types of inefficiencies in any process that organizations have difficulty see-
ing, let alone eliminating. These three are popularly called 3Ms based on their
Japanese names, viz. Muda (waste), Mura (unevenness), and Muri (overburden).
3Ms are explained in detail in later chapters. A lot of attention during Lean-based
process improvements is given to the elimination or reduction of the 3Ms.

• Flow: The flow of material/work through the organization is unhindered, akin


to the flow of a river, without any churning, reversal, or overflows.
• Pull: The trigger to perform action is initiated by the request/order from
downstream customer (internal/external). In the absence of a signal from down-
stream, no action will be performed by the upstream. There may be some
strategic inventories to help maintain flow.
14 2 Lean Management Principles

• One-piece flow: The ideal state where a single piece is made at a time in
response to each customer order just when it is needed. There is no batch pro-
duction. Each process in the chain is working on one piece, and a maximum of
one piece may wait its turn before any process.
• Stabilization: The organization has a complete grip on the quality of its prod-
uct(s), by virtue of clearly defined processes, and their operating ranges,
conditions, and capabilities.

2.3.1 Flow

Removal of obstacles to the smooth flow of the product through various produc-
tion stages is usually the first step in most of the Lean implementations. Flow is
the natural order of things—the best natural systems are designed on flow princi-
ples including all living systems such as human body. Streams and rivulets flow
to the river and rivers flow to the sea—creating great natural wealth along their
journey. Human body itself has multiple flows—air or breathing, food (digestion),
water and blood (life), and brain (impulse). What would happen if any flow is
constrained, blocked, or disrupted?
In business processes, typically, a product undergoes a series of sequential steps
as it is converted from input (raw material) to finished item that is ready to be sold.
In flow phase, firms would identify the impediments to flow and remove them
gradually and steadily. Smooth flow is prevented whenever sequential processes
operate at different rates. Flow is also blocked if large changeover/transfer times
induce production/movement of big batches. Imagine the flow through a water pipe
with different diameters at different cross sections. Conventional locally optimizing
schedules result in inventory getting accumulated before processes with slower
rates, and the equipment after such processes stays idle for want of input materials.
Further, such a schedule results in long flow time for the products.
Before moving further, let us pause for a moment and ask ourselves the ques-
tion—What should flow? The organization is in business to serve and satisfy its
customers—it is customers who pay for enterprise’ goods and services and keep
the business profitable and sustainable. Customers pay for the specific product.
Typically, the raw material gets converted in an incremental manner to the final
product. Each step that changes the material in the direction specified by the cus-
tomer is value adding through change in physical or inherent properties. And by
corollary, any step (including waiting) that is not in the customer direction is a
non-value-adding step, a waste. So, the product needs to flow through these value-
adding steps, i.e. value should flow. A good way to visualize this would be to
imagine oneself as a product moving through different production steps in your
plant. If you are moving from first to the last step without ever stopping any-
where, then you have a perfectly smooth flow. Each place where you stop is an
impediment to the flow, and since there is no value addition during waiting, it is
considered as a waste. You will be surprised to see the number of places and the
2.3 Lean Management Maturity Phases 15

amount of time you will be waiting without value being added. Similarly, there
are other inefficiencies (3Ms) as mentioned earlier which are impediments to flow.
The objective of this stage is to identify, reduce, and eventually eliminate the 3Ms,
as much as possible in structured and documented iterations. For a firm new to
Lean, implementing this phase (flow) is the most involved, as it requires a number
of behaviour altering changes at the shop floor. The following four principles help
one to identify impediments and improve the flow of value through an enterprise.

2.3.1.1 Value Stream Mapping (VSM)


Value stream mapping is one of the key visual tools that helps chart the flow
of material through the plant from material receipt to dispatch of final product.
The information flows needed to ensure material flow are then depicted on the
VSM making it an integrated view of the operations. Based on Toyota’s Material
and Information Flow diagram, Rother and Shook (Learning to See: Value Stream
Mapping to Add Value and Eliminate Muda, Lean Enterprise Institute, 1999) have
developed this tool. Drawing a VSM is a good first step in establishing a flow,
because it offers a clear, unambiguous, comprehensive, and unanimous view of
current flow and its obstacles. With the seemingly confusing product paths sorted
out on paper, it is easy to identify the major impediments. These impediments are
marked as clouds or improvement projects in star boxes and priorities assigned. It
is also important to understand that true flow is said to occur, only when value is
added. Hence, the VSM charts only the value-adding operations and gives a per-
spective on the mismatch of rates of different operations, thereby offering pointers
to the true capacity of the system. The non-value-adding material flows pertaining
to material handling, and transportation within the factory premises is separately
depicted using a Spaghetti diagram or an arrow diagram (described in detail in
Chap. 5). These bring to surface the causes for impediments to flow such as layout
and process design issues.
Reasonable caution needs to be exercised while creating VSM, and yet there
is no need for the team to go overboard in capturing the complete nitty–gritty
of material flows. Likewise, the initial project priorities may have to change, if
some initial improvements result in changing the process behaviour substantially,
rendering either previously established improvement projects as irrelevant or by
presenting completely new, unforeseen, and more critical impediments to flow.
The firm should be prepared to step back and view at the whole VSM every now
and then, critically reassess the priorities of projects.

2.3.1.2 Balance the Production Rate (Cycle Time) in Line with Takt
Rate
Takt time is the inverse of desired production rate (cycle time) to meet the entire
customer demand and can be calculated as the available time divided by the num-
ber of units to be produced (to meet the demand). If a factory needs to produce
960 units in an 8-h shift, the production rate is 120 units per hour, then the takt
time is 30 s per unit and takt rate is 2 units/minute. Derived from a German word
that means “metre”, it is similar to the heart beat of a human being. When demand
16 2 Lean Management Principles

goes up, takt time or target time to produce will reduce and vice versa. One of
the most common blocks to flow in most organizations is due to the differences in
cycle times for different sequential steps in production. Cycle time is the amount
of time taken by the specific operation to complete the processing of one unit of
the product. Clearly, the cycle time of no process can be more than the takt time,
else the factory would be unable to meet the customer demand on a consistent
basis. Similarly, having cycle time for any of the process much lower than takt
time implies idling of resources, an undesirable situation as well.
Detailed time and motion studies help managers to develop a thorough under-
standing of the work performed at a granular (task) level, both by machines and
operators, and the dependencies. For various historical and convenience reasons,
tasks are grouped into chunks at different work stations resulting in different
production rates. According to balancing principle, tasks are redistributed and
regrouped in a way that the time taken at all work stations is nearly equal to the
Takt rate. Some of the useful solutions to balance the rates at different work
stations are decoupling machine and operator work, ergonomic redesign of the
workplace, elimination of uneven (Mura), strenuous (Muri), and wasteful motions,
reduction of set-up time, and transfer batches (Muda). Work stations having longest
cycle times are called constraints or bottlenecks and determine the overall capacity
of the system. Upstream of the bottleneck, the production is pulled by the signal
from bottleneck, and downstream production either flows or is pulled by customer
signal.

2.3.1.3 Waste Elimination (Muda)


Anything that is not value adding from a customer perspective is a wasteful activity
and prevents free flow of value. Such activities/processes should be reduced and
eventually eliminated. Usually, customers value product features (such as safety,
environment friendly, maintainability, ease of replacement), associated services
(such as delivery, service, insurance, reminders), and the experience of purchase
and post-purchase. It is worth the while of a Lean manager’s time to dig deeper and
evaluate whether and how different processes add value. In majority of the systems
we observed, the value-added time was as low as 20% of the overall cycle time
of the component. Understanding value and waste from a customer perspective
within the manufacturing operations is the key. TPS has categorized the wastes
into seven categories, viz. overproduction, storing in inventory, waiting, material
transportation, human motion, overprocessing, and rework.
This Muda has a direct impact on the manufacturing lead time, the reduction
of which is the primary goal of Lean. Figure 2.3 shows a typical production flow
comprising both value adding and wasteful activities.
Now we minimize the waste activities and the time spent on them. Value-adding
time remains the same. We can see that reducing the wastes even without any
investments in speeding up or enhancing capacities of the actual value-adding oper-
ations can directly reduce the lead time and thereby increase capacity as shown in
Fig. 2.4.
2.3 Lean Management Maturity Phases 17

Lead Time = 10 days

Value Added time = 2 days Waste

Fig. 2.3 Wastes contributing to higher lead time

Fig. 2.4 Reduction of wastes


reducing the lead time

Lead Time = 5 days

Value Added time = 2 days Waste

The benefit of capacity increase is only incidental. As per Womack and Jones
(1996), every time an organization quarters its lead time, productivity goes up, and
costs come down by 20%. It should be noted that, not all wastes are alike in terms
of pursuit of their elimination. Context plays an important role in determining
the priority. For example, in a capital intense industry, eliminating overproduc-
tion is less important than the waiting of the machine, whereas in an organization
with mostly manual processes, the waiting time of operators may be used in
some other productive tasks such as maintenance/process improvement, rendering
overproduction as unnecessary.

2.3.1.4 Changeover Time Reduction—SMED


If any concept could be touted as the single most important thing to have helped
establish TPS and implement flow at Toyota, it would be Single Minute Exchange
of Dies (SMED). One of the main obstacles to flow was high changeover times
between product types on many machines. TPS suggests a way of observing all
18 2 Lean Management Principles

the sub-tasks involved with a view to identify sub-tasks that can be done without
having to stop the machine, or by someone other than operator, and any pro-
cess/jig/fixture improvements that can reduce the remainder of the changeover
times.

2.3.1.5 Visual Management (5S)


Visual tools play a very important role in Lean management on the floor monitor-
ing, and control. Visual systems are designed to give immediate cues to operator
and supervisor regarding the status of process. The 5S—sort, set in order, shine
(and inspect), standardize, and sustain—are the classes of actions every operator
undertakes around her immediate work area, every day, to inculcate the behaviour
of observation, action, and visual management.
“Sort” requires that only the tools, materials, bins, and documents that are
needed for value-adding work should be present at the workstation and the
unwanted things removed and are disposed (after the review period). These
required items are “Set” in clearly pre-defined places that enable ease of access
to perform the work with minimum waste. The entire workplace is ensured to
be clean, by “Shining” which also facilitates inspection for any abnormal condi-
tions. The workplace processes are “Standardized” through documented SOPs and
visual standards which help maintain the first 3S. Finally, the individual and the
organizations are expected to reach a level of self-sustenance, where all the above
practices are followed by each and every person on their own and it becomes a
part of their conditioning.

2.3.2 Pull

Conventionally firms produce to forecast, because they have accepted that the
lead times (consisting largely of non-value-adding times) are unchangeable. This
requires various manufacturing resources in the plant to be scheduled (ahead of
time) such that the product becomes available, when it is forecasted as required.
In addition, generally each production resource is considered in isolation in the
material requirement planning (MRP) systems to generate locally cost-optimal
production batches. The production of different batch sizes at different production
stages adds further to process variability and hence to lead time. Lean management
overcomes this quandary by implementing a pull method for production.
Once the firm has achieved a good flow, by reducing and or eliminating as
many wastes, and hence the lead time, it is ready to implement pull. Reduction of
lead time paves way for pull because the downstream process (customer) does not
have to wait too long to receive the input (as against picking from stock). Often
TPS Sensei (teacher/master in Japanese) considers pull as a bridge to achieving
the ideal state of one-piece flow. In the words of Rother and Shook (2003), “Flow
where you can, pull where you must”. By flow, here it is meant one-piece flow.
In practice, pull loops become necessary in the value stream due to batching and
differential rates of production in different product stages. Typically, pull loops
2.3 Lean Management Maturity Phases 19

proceed from customer end and move up the organizational processes until such
point where it is difficult to distinguish the order-level activity. Following are some
mechanisms to implement pull in organizations.

2.3.2.1 Pacemaker
Pacemaker is the last process in the value stream after which there exists only
continuous flow till the finished good stage (Rother & Shook, 2003). As the name
suggests, pacemaker determines rate at which the customer demand is met. Before
the pacemaker, there can be a mix of flows, shared resources catering to multiple
product groups, etc. Pull often begins from the pacemaker process. For example,
in a typical assembled product such as automobile, white goods, pumps, or toys,
the first operation in the final assembly line would be the pacemaker as after this
there is only line flow. The continuous improvement endeavour of Lean organiza-
tions keeps pushing the pacemaker further up the supply chain. Further, recall that
we defined a chain of clear binary supplier and customer relations throughout the
supply chain using VSMs in the flow stage. We make use of these binary links to
send the pull signal all the way to the most upstream supplier via each supplier
in our VSM. The upstream supplier to each customer in the chain will only start
producing when she receives the pull signal from the customer. The limitation to
this is the fact that not all resources belong to a single supplier–customer lineage.
There are usually some shared resources as part of most of process streams. Cen-
tralized planning is inevitable for such resources. For the resources prior to the
pacemaker, pull is typically implemented using Kanban cards.

2.3.2.2 Kanban
Kanban is a visual signalling system that can be implemented by use of cards that
are passed from downstream customer to supplier and vice versa. As observed
previously in Takt rate Sect. 2.3.1.2, production upstream of pacemaker is pulled
by Kanban signal from pacemaker, whereas downstream production is driven by
customer orders. This is implemented for every binary customer–supplier relation-
ship in the enterprise value chain. The literal meaning of the word Kanban in
Japanese is signboard. As the name suggests, Kanban is the manifestation of pull
implementation on the production floor using visual management (typically cards),
i.e. the upstream producer will produce “only” when (s)he (or a machine with a
sensor) “sees” a signal from the downstream customer. There are six inviolable
rules that go with Kanban at TPS, to ensure that good material moves along with
the cards, only when there is trigger. The number of cards in the system not only
controls the extent of inventory, but also the extent of variability propagation up
the supply chain, popularly known as the Bull-Whip effect. While organizations
have developed a number of variants of Kanban, to suit specific contexts, includ-
ing e-Kanbans, the fundamental pull principle remains intact. Given their specific
purpose, Kanban is also classified as production, transport Kanban. Further, pro-
duction Kanban is categorized into signal and production Kanban, depending on
the presence of change-over time and transport Kanban is categorized into supplier
and in-house Kanban.
20 2 Lean Management Principles

2.3.2.3 Production Levelling (Heijunka)


Most organizations produce more than one model or a variant of product/service
using common resources. Traditionally, it is viewed convenient to produce one
model continuously and then switch over to another model and produce it contin-
uously. Such uneven production in batches is an example of Mura. The problem
with this approach is that sales happen at a different rate compared to production
resulting in inventory holding or order backlog. All the associated problems due
to excessive inventory such as feedback delay and lost/additional human effort fol-
low. The solution is to create production runs such that in each run all models are
produced in the same proportion as the overall sales proportion. The production
run size is gradually reduced to minimum by continuously reducing changeover
times (see SMED, Sect. 2.3.1.4), and transport batch is reduced by defining an
appropriate material collection frequency (pitch). The important outcome of Hei-
junka is to define batches such that production effort is levelled throughout the
time horizon. A side benefit of the ability to produce mixed lots is the embed-
ded production flexibility to accommodate any model-mix changes in the demand.
One of the common challenges to implement Heijunka comes from change resis-
tance by employees, understandably so. We, humans, are conditioned such that
we avoid change, if we can. Therefore, it requires conscious effort, coaching, and
behavioural reinforcement to repeatedly change the models. Over a period of time,
this flexibility becomes second nature of the employees at Lean organizations.

2.3.3 One-Piece Flow

This is the perfect or utopian phase of achieving make-to-order, yet with the
efficiencies of mass manufacturing. Existing make-to-order systems mostly have
unacceptably long lead times and are unable to scale up their business as a result.
A single-piece flow (or 1 × 1 or continuous flow) means that in each operation of
the chain of processes that are involved in the making of a product, only one unit
of material is under process (value addition), and at the most, another unit is wait-
ing before the operation. At its most perfectly synchronized level, this would be
akin to the child game of passing the parcel or a set of labourers manually trans-
porting bricks from stacking point to point of use in a construction site. Achieving
single-piece flow means an excellent balance of cycle times among the connected
operations and minimum disruption in the flow of work due to process instabil-
ity, equipment breakdowns, and planning issues such as material availability. All
the initiatives described in flow and pull sections must be at work such that they
support a batch size of one, with minimal wastes. Further, the following initiatives
help organizations to stay on course towards the Lean perfection manifested in the
form of single-piece flow and waste elimination throughout the value chain.
2.3 Lean Management Maturity Phases 21

2.3.3.1 Kaizen
Kaizen is a combination of two Japanese words: Kai—Change and Zen—For the
better.1 As the name suggests, Kaizen is an endeavour by everyone in the orga-
nization from top to bottom to continuously improve within their own sphere of
work. While Kaizen is practised in general in a number of progressive organiza-
tions, the way it is implemented within TPS is quite purposeful. Rother (2019)
in his book, “Toyota Kata: Managing People for Improvement, Adaptiveness, and
Superior Results”, does provide an in-depth account of how are target conditions
determined, the steps to reach there, a cultural cushion that promotes taking smaller
steps in the direction of target condition, and an overall organizational alignment
towards this approach. So, Kaizen combined with kata (a way of keeping improve-
ment steps in alignment with the target condition) to get to a target condition
provides a powerful impetus and direction for organizations strive to achieve the
Lean perfection.

2.3.3.2 Lean Enterprise


It is imperative that for Lean to truly work at an organization, and not backslide
because its customers or suppliers continue to exhibit batch behaviour, Lean prin-
ciples should be extended throughout the value chain, termed as Lean Enterprise
by Womack and Jones (1994). They provide guidance on how the legally different
enterprises in the value chains may still cooperate, enter into transparent agree-
ments, and perform just transactions. An end goal of a Lean enterprise is that all
wastes are eliminated throughout the value chain, entities are rewarded in pro-
portion to their contribution to value, and there is no pushing of the product to
end customer through meaningless discounts. Lean Enterprise is a powerful and
an over encompassing notion, and organizations may implement the same in a
step-wise manner by progressively convincing different entities in the value chain
through demonstrated benefits.

2.3.4 Stabilization

Before the organization hopes to achieve quantity responsiveness, logic requires


that it must ensure that the products/services it produces conform to the quality
standards set as per customer expectations. The purpose of stabilization stage is
twofold, ensures either lock-in of the previously achieved improvements or as a
preliminary step to Lean, and stabilize the product quality. So, in stabilization
phase, the focus is on minimizing the process variability, improving equipment
effectiveness, achieving defect-free production, and ensuring that operators adhere
to set operational procedures. Outcome of stabilization stage for a given pro-
cess is that it is defect free and consistently produces at a predictable rate. This

1 https://www.kaizen.com/what-is-kaizen.
22 2 Lean Management Principles

stage is perfected by preventing defects, and this is achieved by following guiding


principles.

2.3.4.1 Quality at Source


The empowerment of the operator to stop production as soon as a defect is spotted
is achieved by pulling what is called an andon card (in an assembly line set-up).
There are other digital and on the floor variants available for implementing andon
mechanism. Stopping production in itself does not eliminate defects. The supervi-
sor and the relevant operators immediately convene at the location where defect
was found, root cause analysis performed, and the source of the error is fixed,
before resuming production. Once this discipline is maintained (despite the temp-
tation to put the defect away and keep producing), it is easy to see that the time
lost in root cause analysis is gained quickly by avoiding potential and bigger losses
of passing along the defect and the potential rework involved. This is the trigger
to setting the virtuous cycle in motion. Fixing a defect at its root requires prob-
lem solving of higher order, forcing the operators develop a deeper understanding
of the interactions between material, equipment, environment, and other relevant
parameters.

2.3.4.2 Hundred Per Cent Inspection


All production at all processing stages is fully inspected to catch defects, if any,
before being passed on to next stage. Deploying human resources for inspection is
costly, demeaning to individuals, and above all is not completely fool proof. Efforts
to automate inspection using simple solutions are called Jidoka or autonoma-
tion (automation with human mind) in TPS parlance. Poka-yoke or fool-proofing
goes one step beyond and operationalizes various ingenious and cost-effective
mechanisms to prevent errors in processing.

2.3.4.3 Total Productive Maintenance (TPM)


Equipment availability and reliability are enhanced by having the operators under-
take routine checks, preventive maintenance routines, and change of consumables.
TPM makes the operator more sensitive to any irregular behaviour of the equip-
ment and prevent such abnormal symptom building into a more serious problem
later. Equipment conditions impact the process and resultant quality much before
they actually breakdown. TPM makes the production operator to take more
accountability of the equipment they work on and hence take a good care of
them. Chapter 8 details some useful practices related to autonomous maintenance
(AM), planned maintenance (PM), and cleaning, lubrication, repair, and inspection
(CLRI).

2.3.4.4 Standardize
Standardization of a work procedure is the documentation of current, proven best
practice to operate the work station. This serves as the baseline to undertake new
improvement projects, in order to achieve higher performance levels. Standard-
ization refers to both the operating procedures and also the equipment condition.
2.4 Performance of Lean Enterprises 23

Standardization is a necessary step in stabilization phase. A TPS practice called


5S (standing for sort, set in order, shine, standardize, sustain) is an excellent tool
to institutionalize various improvements achieved through Lean initiatives within
the organization. The 5S implementation needs to be strictly in the sequence as
mentioned above for truly firming up the improvements. Typical implementations
of 5S are stage-wise starting from 1S to all the way 5S, augmented by periodic
audits to ensuring adherence.

2.4 Performance of Lean Enterprises

In this section, we present some of the published empirical research on the finan-
cial performance of firms that have implemented Lean vis-à-vis the firms that did
not. Two trends are evident in the published results that Lean not only causes
financial performance improvement, but also is self-perpetuating, i.e. it feeds on
its success and delivers even better results every successive year.
Dieste et al., (2021) conducted a systematic literature review of 24 research
articles published on the impact on financial performance due to Lean imple-
mentation. For the purpose of comparison, they grouped Lean initiatives into
four bundles. They are “JIT” consisting of pull/Kanban, small lots of production,
SMED, continuous flow production, and cellular layouts, “TQM” consisting of
Kaizen, customer involvement, visual management, statistical process control, and
5S, “TPM” consisting of preventive and autonomous maintenance, and “HRM”
consisting of multiskilled workforce, employee involvement, and Lean leadership.
They found that JIT and TQM seem to enable a better financial performance,
than TPM and HRM. Interestingly, they have also found evidence of unsatisfac-
tory financial performance, where Lean was implemented half-heartedly. Similarly,
Hofer et al. (2012) observed that a complete (internal and external oriented) Lean
implementation gives higher performance benefits as against implementation of
selective tools.
According to a research survey conducted by Arnaldo Camuffo in 20162 at
100 SMEs based in Italy who have seriously embarked on the Lean journey, the
EBIDTA margin and ROIC have shown substantial and nonlinear improvements
vis-a-vis their non-Lean counterparts. EBIDTA margin has increased to 11% in
year 3 and to 54% by year 4. Likewise, ROIC has improved from 4% in year 1 to
nearly 1.5 times by year 7. There was however a small dip in the initial year or so
on EBIDTA margin, which is understandable since it takes that long for the new
ways of operating (such as avoiding overproduction) to settle. In another study on
North American SMEs, Olsen (2004) determined that not only the cash-to-cash
cycle and RoE are better for Lean firms, but also synergistic in nature. Based
on the survey findings from 187 manufacturing firms in Malaysia, Iranmanesh
et al. (2019) observed that Lean culture of an organization has a positive effect

2 https://planet-lean.com/financial-performance-sme-research/.
24 2 Lean Management Principles

on the sustainability performance of an organization due to its process, equip-


ment, and supplier relationship management practices. Boyd et al. (2006) used
the Data Envelopment Analysis (DEA), a linear programming-based efficiency
measurement tool that considers a range of inputs such as inventory, labour, and
other assets, to produce outputs such as net sales, gross profit, EBIDTA, EBT, and
net income. The authors applied this comprehensive measure to 18 manufacturing
firms to conclude that the Lean management implementations help firms achieve
superior technical efficiencies. Lastly, in a comprehensive study of the effect of
Lean implementation at Delphi automotive, and Johnson Controls, both driven to
Lean in the second wave by their mother companies, Kocakulah and Upson (2004)
observe that driving culture change and supplier practices are the two main iner-
tias preventing these companies achieve the desired levels on performance metrics
such as inventory turns, quality performance, improved delivery, production lead
times, conversion costs, product introduction times, smaller plants, and improved
productivity. These authors also note that it takes a while for the network effects
to show. After all, it took more than 50 years for the Lean leaders (Toyota) to
achieve the current levels of performance.

2.5 Summary

In this chapter, we presented a comprehensive framework of Lean implementa-


tion with different maturity stages, namely flow, pull, and one-piece flow, and
stabilization. Each stage and the interrelations among them was described from
a conceptual standpoint. Relevant tools and techniques as appropriate for each of
these four stages are introduced. A global review of performance improvement
reported on account of Lean implementations by Small and Medium Enterprises
is presented.

References

Boyd, D. T., Kronk, L. A., & Boyd, S. C. (2006). Measuring the effects of Lean manufacturing sys-
tems on financial accounting metrics using data envelopment analysis. Investment Management
and Financial Innovations, 3(4).
Dieste, M., Panizzolo, R., & Garza-Reyes, J. A. (2021). A systematic literature review regarding
the influence of lean manufacturing on firms’ financial performance. Journal of Manufacturing
Technology Management.
Hofer, C., Eroglu, C., & Hofer, A. R. (2012). The effect of lean production on financial per-
formance: The mediating role of inventory leanness. International Journal of Production
Economics, 138(2), 242–253.
Iranmanesh, M., Zailani, S., Hyun, S. S., Ali, M. H., & Kim, K. (2019). Impact of lean manufac-
turing practices on firms’ sustainable performance: Lean culture as a moderator. Sustainability,
11(4), 1112.
Kocakulah, M. C., & Upson, J. (2004). Lean manufacturing: Selected financial performance of rec-
ognized lean manufacturers. International Business & Economics Research Journal (IBER),
3(12).
References 25

Mrugalska, B., & Wyrwicka, M. K. (2017). Towards lean production in industry 4.0. Procedia
Engineering, 182, 466–473.
Olsen, E. O. (2004). Lean manufacturing management: The relationship between practice and
firm-level financial performance. The Ohio State University.
Rother, M. (2019). Toyota Kata: Managing people for improvement, adaptiveness and superior
results. MGH.
Rother, M., & Shook, J. (2003). Learning to see: Value stream mapping to add value and eliminate
Muda. Lean Enterprise Institute.
Womack, J. P., & Jones, D. T. (1994). From lean production to the lean enterprise. Harvard
Business Review, 72(2), 93–103.
Womack, J. P., & Jones, D. T. (1996). Lean thinking: Banish waste and create wealth in your
corporation. Simon & Schuster.
Lean Management in Small
and Medium Enterprises 3

3.1 Introduction

Exposure to constraints such as dependence on banks for capital, lack of stable


demand, poor visibility to demand, presence of floating and relatively unskilled
workforce, and far upstream positions in supply chains, make Small and Medium
Enterprises (SMEs) particularly vulnerable to vagaries of social, economic, and
political uncertainties. However, these very challenges make SMEs the natural can-
didates to pursue Lean practices. SMEs already have the favourable characteristics
for Lean such as smaller, closer, and visible shop floors, fewer management layers,
closer relationships among employees, higher product variety and the neces-
sity to closely meet customer’s needs, and make-to-order systems (Mrugalska &
Wyrwicka, 2017). They also have unfavourable (to Lean) characteristics such
as investment in terms of loss of production during training, owner’s long-term
ambiguous commitment to Lean, and the impatience for the success to solidify as
a permanent culture (Djassemi, 2014). In our view, the favourable characteristics
outweigh the unfavourable characteristics as visualized in Fig. 3.1.
Many SMEs implemented Lean management to overcome business environment
challenges and to improve profitability. For example, Abdul-Nour et al. (1999)
observe that some asset base is shifting from larger customer enterprises to smaller
supplier SMEs, in order to reduce investment risk. Their research explores ways
to reduce the break-even risk among SMEs using Lean techniques such as mixed-
model scheduling (leading to improved basic, system, and aggregate flexibility),
joining a JIT network, and continuous improvement.
And yet, most literature on Lean is about implementation at large enterprises.
Approach to Lean implementation needs to be modified to suit the needs, con-
straints, and styles of SMEs. In this chapter, we review the literature on nuances
of SME sector with a focus on identifying the barriers and enablers of Lean
implementations. For implementing Lean management, SMEs need to overcome
some unique challenges, which one may not be able to learn from the traditional

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 27
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_3
28 3 Lean Management in Small and Medium Enterprises

• Loss of production during


training,
• Ambiguous commitment to
lean, and
• Impatience to sustain
success

• Smaller shop-floors,
• Fewer management layers,
• Closely knit employees,
• Product mix,
• Meet customer’s needs
accurately, and
• Make-to-order

Fig. 3.1 Favourable and unfavourable (to Lean) characteristics of SMEs

Lean success stories of larger enterprises. This chapter concludes with a descrip-
tion of these challenges, laying a foundation for the rest of this book on Lean
implementation at SMEs.

3.2 Overview of SMEs

What is an SME? Unfortunately, there is no common definition. Ayyagari et al.


(2007) lists the definitions used by many countries for their SME sectors in order
to collect, process, and publish the data on SMEs. Apart from a few quantitative
measures, such as revenue and number of employees, there are a number of qual-
itative characteristics that define SMEs. Berisha and Pula (2015) have highlighted
the qualitative differences between SMEs and large enterprises as shown in Table
3.1.
We describe the perspectives on economic role, unique constraints, barriers
and enablers of Lean adoption, and growth culture with respect to SMEs in the
following sub-sections.

3.2.1 Economic Role

SMEs form the backbone of most economies and are providers of large-scale
employment. A thriving SME sector is a positive economic indicator. In their
study involving 76 countries, Ayyagari et al. (2007) show that the organized SME
sector gains importance as the countries become richer. They also observe that
better business environment comprising healthy competition and a good credit
eco-system contributes to the well-being of SME sector.
3.2 Overview of SMEs 29

Table 3.1 Differences between SMEs and Large Enterprises (adopted from Berisha and Pula,
2015)
Category SMEs Large enterprises
Management • Proprietor-entrepreneurship • Manager-entrepreneurship
• Functions linked to • Division of labour by subject
personalities matters
Personnel • Lack of university graduates • Dominance of university
• All-round knowledge graduates
• Specialization
Organization • Highly personalized contacts • Highly formalized
communication
Sales • Comparative position not • Strong competitive position
defined and uncertain
Buyers’ relationships • Unstable • Based on long-term contracts
Production • Labour intensive • Capital intensive, economies
of scale
Research and development • Following the market, • Institutionalized
intuitive approach
Finance • Role of family funds, • Diversified ownership
self-financing structure, access to
anonymous capital market

Considering the important role played by SME sector, and yet their limited
access to capital, and advanced knowledge, many governments have come up with
a number of support systems. Firstly, most governments have dedicated ministries
or other bodies and also policies to support SMEs (viz., Small Business Services
(UK); SME Promotion Office (Thailand), Small Business Administration (USA),
Bureau of Small and Medium Business Development (Philippines), Ministry of
Micro, Small and Medium Enterprises, India, etc. Stephen Ezell (2011)1 observes
that the national manufacturing support of the countries (Argentina, Australia, Aus-
tria, Canada, China, Germany, Japan, Korea, Spain, and United Kingdom, and
the United States) for SMEs is shifting from continuous productivity improve-
ment (Lean, quality, Six Sigma) to innovation and growth (Technology adoption,
new age manufacturing, connecting SMEs). Babu et al. (2016) review the positive
effects of various government support initiatives under the aegis of National Manu-
facturing Competitiveness Programme (NMCP) starting 2007 on the MSME sector
in India. One of the nine components of NMCP is a Lean manufacturing scheme
under which the government financially supports various SME clusters across the
country in implementing Lean. Further, development of clusters fosters building
horizontal alliances, shares common resources and achieves access to global value
chains, in the presence of co-operative competition (Herr & Nettekoven, 2018).

1 https://itif.org/publications/2011/09/14/international-benchmarking-countries%E2%80%99-pol
icies-and-programs-supporting-sme.
30 3 Lean Management in Small and Medium Enterprises

Taiwanese PC cluster, Chilean salmon cluster, and Indian textile cluster are a few
examples of clusters supported and developed by the respective governments.
Researchers also have recommended that governments should take active role
in supporting SMEs. For example, Tambunan (2005) recommend that government
policies should enable an effective market (domestic and global) linkage. Fur-
ther, Lim and Kimura (2010) observe that increase in globalization is providing
unprecedented opportunities for SMEs to participate in global value chains. They
note that for SMEs to continue to flourish, policy measures to strengthen tech-
nological, human resources, and better access to finance must be taken by the
respective governments.

3.2.2 Unique Constraints of SMEs

Shocks such as BREXIT, shifting orientation global value chains, COVID-induced


recession, and supply disruptions, growing nationalism, and other environmental
uncertainties cause strain especially on SME sector since most of them are at the
upstream end of the supply chains, and by the time the true demand reaches them,
they usually would have spent resources on production (based on forecast).
Naradda Gamage et al. (2020) observed that SMEs are increasingly affected by
global challenges such as global market competition, global finance and economic
crises, information communication technology, the emergence of multinational
corporations, transnational corporations, consumer changes and especially their
preferences, trade dumping, international terrorism, and religious conflicts and
trade wars. According to them, some useful survival strategies that SMEs can adopt
are-expanding dynamic capabilities, investing in research, technology innovations,
new partnerships, getting into global supply chains, and skill development.
Table 3.2 summarizes the constraints and challenges faced by SMEs as
documented in literature.

3.2.3 Growth Orientation

Herr and Nettekoven (2018) classify SMEs into three categories, namely those
that start with an innovative combination of production factors—Schumpeterian
SMEs (Joseph Schumpeter, 1934), those reacting to normal mismatch between
supply and demand—normal SMEs, and those that start in response to lack of bet-
ter economic opportunities—poverty-driven SMEs. Further, the authors note that
the SMEs belonging to first two categories are open to innovation and are rela-
tively fewer in numbers, more so in developing nations. Large number of SMEs,
belonging to the last category, strive neither for innovation nor to become Lean.
Table 3.3 summarizes the unique characteristics of high growth SMEs.
3.2 Overview of SMEs 31

Table 3.2 Challenges and constraints faced by SMEs


Researcher Source Constraints and challenges
Siegel et al. (2019) 45 published academic research Manpower and financial constraints,
articles poor management and leadership,
lack of strategy, resistance to
change, and improper project(s)
selection
Beck (2007) Enterprise Surveys database of the Cost of finance, tax rates,
World Bank macroeconomic stability, economic
policy uncertainty, corruption,
access to finance,
anti-competitive/informal practices,
tax administration, electricity, crime,
theft, disorder, skills of available
workers, legal system, customs and
trade regulations, access to land,
licensing and operating permits,
labour regulations, transport, and
telecommunications

Table 3.3 Unique characteristics of high growth SMEs


Researcher Source Constraints and challenges
Smallbone et al. (1995) High growth SMEs in comparison – Continuously expand the core
with others product offerings to more
value-added versions, thereby
increasing wallet share of
existing customers and
acquiring newer customers
– Added to employment,
increased productivity,
improved production processes,
achieved growth organically,
and
– Increased bandwidth from
management
Bigliardi et al. (2011) Innovative SMEs in comparison Enhance product offerings,
with others improve process capability, and
integrate innovation into the
business strategies

3.2.4 Barriers and Enablers to Lean in SMEs

Common critical success factors (enablers) for Lean implementation as noted by


a number of researchers are organizational learning, culture of innovation, lead-
ership and constancy of purpose, processes and information-based management,
customer focus, strategic direction and planning, training by an external consul-
tant, top management support, communication, supplier relationship, linking Lean
32 3 Lean Management in Small and Medium Enterprises

to business strategy, and customers (Achanga et al., 2006; Hu et al., 2015; Lande
et al., 2016). Likewise, Thanki and Thakkar (2018) in their study of interdepen-
dence of critical success factors for Lean and green implementation among SMEs
have found that effective leadership and management’s customer focus, communi-
cation of goals, linking improvement initiatives with objectives, top management
commitment, clear organization strategies and policies, and government support to
be the driving CSFs.
Unique constraints faced by SMEs in their Lean adoption are lack of Lean
knowledge across the organization, cost reduction through economies of scale
mentality, poor (if any) linkages with suppliers and customers, external drive for
Lean (typically by customers), lack of qualified staff, and lack of access to exper-
tise (Matt & Rauch, 2013). Based on a detailed literature review and Lean experts’
interviews, Belhadi and Touriki (2017) identify lack of management involvement,
lack of adapted methodology of Lean implementation, short-term vision, fear and
resistance to change, and lack of understanding of Lean as the top five barriers to
Lean adaptation. Moreover, they also note that commitment and participation of
management, adoption of simple measurement and KPIs, development of organi-
zational learning culture, early deployment of Lean culture through training, and
allocation of sufficient time and resources are the top five solutions to overcome
the above barriers. Thanki and Thakkar (2014) identify inadequate Lean training
and lack of Lean awareness programs for employees, poor application of statisti-
cal tools for process improvement, and uncertainty regarding the appropriate Lean
tool as the barriers to adapt Lean among SMEs. Choi (1997), while discussing
the efforts of seven small automotive parts suppliers to implement continuous
improvement (CI), shows that only three had any degree of success. He identi-
fies three major pitfalls for small- to medium-size companies when implementing
Lean production: (a) alienation of line leaders, (b) treating CI simply as a problem-
solving activity or as something to do when there was a spare moment, and (c)
viewing CI as either a management program or a worker program.I am not able to
locate where I picked this from. Perhaps the editor team can help me find the refer-
ence?Please ignore this comment. We mentioned this reference in the References
list.
In their case study analysis involving a medium-sized Swedish enterprise,
Assarlind and Aaboen (2014) have identified positive and negative forces that push
forward or pull down an organization, respectively, in its Lean journey. According
to them, it is equally important for an organization to weed out the negative forces
as much as it is to strengthen the positive forces. Various frameworks are pro-
posed for implementing and sustaining Lean such as tree of Lean implementation
by Thanki and Thakkar (2011), Lean staircase by Hu et al. (2015), and a three-
phased approach by Djassemi (2014). Hu et al. (2015) in their literature review
covering over 100 Lean papers noted that most implementations are tool centric,
and efficiency focussed, and are not driven by an effective long-term roadmap.
3.3 Resistances to Initiating Lean 33

3.3 Resistances to Initiating Lean

It takes patience and trust-building with the SME owners and managers to elicit the
exact challenges specific to their organization. During our numerous interactions
with them, we were able to identify a few common resistances to initiating Lean.
These resistances belong primarily to two categories, financial and non-financial,
and we discuss them comprehensively here.

3.3.1 Financial Resistance

The foremost resistance put forth by SMEs to adopt Lean in their organizations is
to do with the cost of the intervention. This resistance manifests in various ques-
tions such as—where would the funds come from, how to reduce the initial outlay,
how soon would they recover the investment, how many times over the initial
investment can be recovered etc,. Most SMEs operate with limited funds, generally
internally generated or highly restrictive bank loans as the primary source of capi-
tal, and hence, this resistance is understandable. So, the SME owners’ first lookout
is at the payback of such an initiative. “If I invest X this year, how many times X
will I realize and when”. While it is easy enough for the SME owners to visualize
the payback of hard assets such as machinery or people, it is difficult for the own-
ers to visualize outcomes of an intangible (and typically ongoing) management
approach such as Lean.
Further, a sustainable Lean intervention is more of a teaching and training exer-
cise, which can be successful only with willing-to-learn students, and less of a
consulting exercise in the conventional sense. Considering the all-pervading mar-
keting glitz of making promises of quick fixes and magic pills by various solution
providers including consulting and IT, it is a tough task for the Lean consultant to
get the SME owner to understand that taking ownership of Lean is the only way to
make it generate sustained benefits. Please see “Skin in the game” for our recent
experience of this phenomenon.

Skin in the Game


A few years ago, we were referred to a defense electronics manufacturing
unit where there was an urgent need to ramp up the output to meet customer
demand. The owners, a technocrat couple, felt that they had implemented the
best manufacturing system and technology. We noticed a typical batch type
operations with a possibility of 50% increase in output through implementing
basic Lean concepts. And we went ahead and put the same in our proposal to
the client along with a reasonable fee structure considering their SME status.
We included local references of our work.
After going through the proposal, the owners called us for a discussion.
The gist of the discussion was that we should take only 25% of the fee
34 3 Lean Management in Small and Medium Enterprises

for implementation and balance would be paid to us only on the basis of


achieving the targets. We explained to them the Lean philosophy and how
we would only be the guides to their internal team and that it is the internal
team which would take ownership for the initiative and also credit for its
results.
The next day we received a two-line email from the Managing Director.
It said either accept their terms or forget the assignment. We thanked him
for his time and wished them well for the future.

Lastly, many SMEs live from year to year; strategizing and planning five or ten
years ahead is rare. When the business (planning) horizon itself is not very long, it
becomes difficult for them to commit to a year-long Lean program. Few, if at all,
SMEs do a structured budgeting exercise and use it for planned expenses. Absence
of planned budget adds to the difficulty of raising funds, out of the blue, to pay
for a Lean intervention.

3.3.2 Non-financial Resistance

After the financial resistance is softened or overcome (some ideas on how this
is done are described in Chap. 4), a variety of non-financial obstacles begin to
surface. Most common among them are described below.

3.3.2.1 Divergent Thinking


A lot of SMEs are family owned and are managed with fathers, uncles, sons,
and daughters overseeing different aspects of the business. Often, the younger
generation is enamoured by Lean, learning about it during their studies or from
books or during business networking meetings. However, they find it difficult to
get the buy-in of their “senior” family members who often hold the reins of major
decision-making. Please see the box “Weaving strands together…or not”.

Weaving Strands Together…or not


A former student of one of our consultants invited us to their manufactur-
ing unit involved in making high strength plastic ropes. She was keen to
have Lean implemented to increase their capacity in-line with their ambi-
tious growth plans. We agreed to do a chargeable diagnostic visit. Once
there, we met her brother who had been given charge of plant operations
and their father who was the founder and overall boss.
The brother said he wanted the unit to have a world class look and feel
through implementing Lean as it would attract more international buyers.
The father did not set any goals but said he supports whatever his children
do and he wanted them to take charge of the business.
3.3 Resistances to Initiating Lean 35

We concluded the three-day diagnostic laying out a clear roadmap with


the expected benefits. We even took up a couple of areas and demon-
strated application of Lean techniques like flow and cycle time reduction
and showed immediate results on a pilot basis.
But the implementation assignment never came through. We learnt later
that the father was not convinced about Lean despite seeing the results.

3.3.2.2 Resource Time


After the financial and philosophical issues are ironed out, operational challenges
are brought to the fore. Typically, SMEs are structured as flat organizations with
just a couple of key people running the show. It is no surprise then that they are
usually busy with routine work and time is at a premium. The minute Lean and its
emphasis on involving all the people is laid out, and the management gets worried
about their key resource time getting blocked. A frequently asked question which
most Lean consultants would have heard many times over is “Can you deploy your
people to run and implement the Lean program. Our people do not have the time”.
However, this is not an option as the sustenance of the program depends only on
the internal team taking ownership for the Lean initiative right from day one.

3.3.2.3 Employee Turnover


A closely related challenge is the continuity of employees with Lean knowledge
within the enterprise. The salaries being conservative in SMEs, some employees
(especially the good ones) are looking to gain some experience before moving on
to a higher paying job or secure job with a better established companies. There
is a genuine worry that having been trained on Lean, such employees become
more marketable and may soon leave for better opportunities. Another continuity-
related challenge especially in developing nations is the dependence of the SME on
contractual “migrant” workforce to run their operations. In South Asian countries
like India, these workers have unique vacation patterns that depend on myriad
of factors such as monsoon, harvest season, festivals, and marriage functions, at
their “native” states. It is quite normal to see a whole group of people absenting
themselves for weeks at a time during such occasions leaving the SME struggling
to run day-to-day production. Please see the box “People are the key”.

People Are the Key


We recently commenced an engagement at an engineered stone manufac-
turing unit located in a remote area. Within the first two months, half the
supervisory staff had left, the engineer given role of Lean champion had to
take over supervisory duties and was placed in the night shift. The plant man-
ager went on a long leave of absence. One of the family members became
plant manager. And Lean came to a grinding halt.
36 3 Lean Management in Small and Medium Enterprises

At this stage, the owners asked us the inevitable question “Can you deploy
one of your people to continue Lean”.

3.3.2.4 Trust
Openness to share information and data is another important barrier, especially
in recipe-based industries such as food and pharmaceuticals. Hiring an outside
Lean Sensei or consultant requires opening the shop floor and operational data
to this external scrutiny. Family-owned SMEs are conservative by nature and are
not comfortable with sharing the operational data or process knowledge with an
outsider.
With these many barriers, both real and imaginary, in their minds, it is no
wonder that most SMEs do not opt to even commence their Lean journey in spite
of its widely known and proven benefits. We will present a few practical ways
based on our experiences, to counter, soften, and overcome these resistances and
make a start of Lean, in Chaps. 4 and 5.

3.4 Summary

This chapter discusses the unique characteristics of SMEs and their differences
vis-à-vis larger enterprises. It surveys the state-of-the-art literature to present the
economic role, unique constraints, and the growth pathways taken up by SMEs.
Further, the literature review identifies the documented enablers and barriers, for
Lean implementation at SMEs across the world. The chapter is concluded with
the empirical findings from the authors from their various experiences in terms
of financial and non-financial resistances posed by the SMEs to the idea of Lean
implementation.

References

Abdul-Nour, G., Drolet, J., & Lambert, S. (1999). Mixed production, flexibility and SME. Com-
puters & Industrial Engineering, 37(1–2), 429–432.
Achanga, P., Shehab, E., Roy, R., & Nelder, G. (2006). Critical success factors for lean implemen-
tation within SMEs. Journal of Manufacturing Technology Management.
Assarlind, M., & Aaboen, L., (2014). Forces affecting one Lean Six Sigma adoption process.
International Journal of Lean Six Sigma.
Ayyagari, M., Beck, T. H. L., & Demirgüç-Kunt, A. (2007). Small and medium enterprises across
the globe. Small Business Economics: An International Journal, 29(4), 415–434.
Babu, D. J., Neerackal, J. J., & Balamurugan, R. N. (2016). Make in India—A prospect of devel-
opment for micro, small and medium enterprises. Journal of Contemporary Research in
Management, 11(1), 23.
Beck, T. (2007, April). Financing constraints of SMEs in developing countries: Evidence, deter-
minants and solutions. In KDI 36th Anniversary International Conference (pp. 26–27).
References 37

Belhadi, A., & Touriki, F. E. (2017). Prioritizing the solutions of lean implementation in SMEs to
overcome its barriers: An integrated fuzzy AHP-TOPSIS approach. Journal of Manufacturing
Technology Management.
Berisha, G., & Pula, J. S. (2015). Defining small and medium enterprises: A critical review.
Academic Journal of Business, Administration, Law and Social Sciences, 1(1), 17–28.
Bigliardi, B., Colacino, P., & Dormio, A. I. (2011). Innovative characteristics of small and medium
enterprises. Journal of Technology Management & Innovation, 6(2), 83–93.
Choi, T. Y. (1997). The success and failures of implementing continuous improvement programs:
Cases of seven automotive parts suppliers. Becoming lean: Inside stories of US manufacturers
(pp. 409–456).
Djassemi, M. (2014). Lean adoption in small manufacturing shops: Attributes and challenges.
Journal of Technology, Management & Applied Engineering, 30(1).
Herr, H., & Nettekoven, Z. M. (2018). The role of small and medium-sized enterprises in develop-
ment: What can be learned from the German experience? (No. 53). Global Labour University
Working Paper.
Hu, Q., Mason, R., Williams, S. J., & Found, P. (2015). Lean implementation within SMEs: A
literature review. Journal of Manufacturing Technology Management.
Lande, M., Shrivastava, R. L., & Seth, D. (2016). Critical success factors for Lean Six Sigma in
SMEs (small and medium enterprises). The TQM Journal.
Lim, H., & Kimura, F. (2010). The internationalization of small and medium enterprises in regional
and global value chains.
Matt, D. T., & Rauch, E. (2013). Implementation of lean production in small sized enterprises.
Procedia CIRP, 12, 420–425.
Mrugalska, B., & Wyrwicka, M. K. (2017). Towards lean production in industry 4.0. Procedia
Engineering, 182, 466–473.
Naradda Gamage, S. K., Ekanayake, E. M. S., Abeyrathne, G. A. K. N. J., Prasanna, R. P. I. R.,
Jayasundara, J. M. S. B., & Rajapakshe, P. S. K. (2020). A review of global challenges and
survival strategies of small and medium enterprises (SMEs). Economies, 8(4), 79.
Schumpeter, J. A. (1934). The theory of economic development. Harvard Economic Studies.
Siegel, R., Antony, J., Garza-Reyes, J. A., Cherrafi, A., & Lameijer, B. (2019). Integrated green
lean approach and sustainability for SMEs: From literature review to a conceptual framework.
Journal of Cleaner Production, 240.
Smallbone, D., Leig, R., & North, D. (1995). The characteristics and strategies of high growth
SMEs. International Journal of Entrepreneurial Behavior & Research.
Tambunan, T. (2005). Promoting small and medium enterprises with a clustering approach: A
policy experience from Indonesia. Journal of Small Business Management, 43(2), 138–154.
Thanki, S. J. & Thakkar, J. (2011). Lean manufacturing: Issues and perspectives. Productivity,
52(1), (p. 12).
Thanki, S. J., & Thakkar, J. (2018). Interdependence analysis of lean-green implementation chal-
lenges: A case of Indian SMEs. Journal of Manufacturing Technology Management.
Lean Implementation Methodology
for SMEs 4

4.1 Introduction

Doing is the essence of Lean. This spirit is captured perfectly by former Toyota
President Fuji Cho, when he stated

We place the highest value on actual implementation and taking action. There are many
things that one doesn’t understand and therefore we ask them, why don’t you just go ahead
and take action; try to do something?…So by constant improvement, or should I say, the
improvement based upon action, one can rise to the higher level of practice and knowledge.

The essence of a management philosophy such as Lean is that there are many
possible approaches to implement. Organizations mistakenly viewing Lean as a
collection of tools, tend to focus on implementing the tools with progressive dif-
ficulty in adaption. When Lean is seen as a philosophy, then there exists a logical
progression of phases and the corresponding steps that ensure sustained results.
Over the last three decades, organizations have undertaken the Lean journey
through many different routes with differing degrees of success. Even within orga-
nizations that consider Lean as a philosophy, what succeeds at one organization
may not work for another. This is due to the myriad of social, economic, and
cultural diversities across the implementing organizations.
Hence, while implementing Lean, one has to follow a certain broad framework
within which the concepts, tools, and techniques are defined. But, what to do
when, where, and how is customized or fitted based on the specific needs and
current status of the individual organization.
This book focuses on the needs of the Small and Medium Enterprises and
standalone units. Many of them are family-managed businesses, and their unique
features have been discussed in the previous chapter. In this chapter, we propose
a methodology for implementing Lean at SMEs. A few outstanding organizations

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 39
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_4
40 4 Lean Implementation Methodology for SMEs

have gone through the entire journey and developed an internal culture of contin-
ual improvement. However, most SMEs find it difficult to envision the outcomes
and are hesitant to commit resources to undertake this long journey at the outset.
We propose various approaches that can give them a taste of Lean and whet their
appetite for the long haul. These are based on our experience of over a hundred
interventions during the past two decades.

4.2 Lean Implementation Methodology

Large organizations driven by strategy, foresight, and budgetary planning are able
to initiate and embark on Lean journey in a planned-ahead manner. There are sev-
eral case studies and books written about TPS and Lean at such organizations.
Many of these organizations have their own internal teams often guided by exter-
nal Lean Sensei as they walk along the Lean path. The overall methodology for
implementing Lean is pretty much streamlined, and we have discussed this in
an earlier chapter. The pyramid framework for Lean is founded on the principle
of process stability. Lean begins with validating process stability against existing
standards as the first step to ensuring processes are running in a stable and stan-
dardized manner. Implementation then follows the steps up the pyramid as shown
in Fig. 4.1.
However, with SMEs, as we have discussed in the previous chapter, there are
several barriers and constraints to adopting Lean. We saw that these include a num-
ber of financial as well as other factors. With their short-term planning horizons,
financial constraints, and people stability issues, it is usually difficult to excite
SME managements to take up Lean journey through the pyramid model. A Lean
implementation methodology that shows some quick and substantial gains is more
likely to motivate the SMEs to go in for the long haul.

One
piece flow

Pull

Flow

Stabilization

Fig. 4.1 Lean implementation pyramid at large enterprises


4.2 Lean Implementation Methodology 41

Wading through the various approaches that can be taken, we have been able
to arrive at a broad methodology of the flow of activities for implementing Lean
across organizations.

Learning from the Largest Formal Lean Manufacturing Program for SMEs
In 2008, the Government of India, unveiled a National Manufacturing Com-
petitiveness Program (NMCP) with a separate track on implementing Lean
manufacturing in MSMEs. There was a belief that Lean would help in
improving profitability and growth of SMEs, but that they would need some
support in getting access to reputed Lean Sensei. The program was designed
using a broad 5-stage framework spread over a 12–18 month time frame.
One of the authors has worked with three clusters comprising 29 SME orga-
nizations under this program and adapted the lean methodology to this
framework:

• Stage 1 — Completion of Diagnostic Study Report (DSR) that includes


current operations assessment and time bound targets for achieving
improvements,
• Stages 2 and 3 — Applying Lean tools and techniques to improve
processes as per roadmap,
• Stage 4 — Standardizing the improved processes,
• Stage 5 — Long-term sustenance which was aptly captured in the pro-
gram guidelines stating “However, it is required that the beneficiary units
will follow the Lean Manufacturing techniques after the exit of Government
of India program”.

Over a decade, this methodology has been rigorously applied across organi-
zations of all sizes, industry sectors, and geographies and improvised with the
learnings from each successive implementation. The version presented here in
Table 4.1 has been validated over the past five years through multiple Lean
interventions and depicts one complete cycle.
This methodology is a variation of the Lean implementation pyramid model dis-
cussed earlier. Firstly, process stabilization is now worked upon post-improvement
Stages 2 and 3 which focus on flow and pull, respectively. The reasons for this
fundamental change are: processes in most SMEs have a lot of inherent waste, and
there is little point in stabilizing these only to alter them again in the next stage,

• Stabilization of processes needs patience and discipline, takes some time to get
right, and may not lead to striking gains; SMEs may not have the inclination to
invest resources,
• Stabilizing processes after improvement sets the ground for next iteration of
improvement projects which are taken up by the sustenance team formed in
Stage 5.
42

Table 4.1 Methodology for implementing Lean at SME


Timeline (months) Stage Focus areas Key Lean tools & techniques Expected outcomes
1 1.Diagnostic “Know 1. Understanding lean • Value Stream Mapping (VSM) • Know current capacity
Current State” paradigms • Material Flow Diagrams • Set Targets for key parameters
2. Assessing current state • Measuring resource productivity – Productivity, Quality,
vis-a-vis organizational goals • Overall Equipment Delivery, Cost
3. Identifying opportunities Effectiveness (OEE) • Lean Roadmap to enable
for improvement operations to meet the set targets
2 to 4 2.Create Flow “Quick 1. Connect processes with • Takt time & Process Design • At least 20% increase in
Wins” each other • Layout design and throughput
2. Improve material flow implementation • Reduction in WIP and
through value stream • Eliminate 3Ms (Muda, Muri and throughput time
Mura) • Minimized material handling
• Line Balancing • Increased space utilization
4

5 to 8 3. Enable & support flow Debottlenecking • Workstation improvement • Increased output


“Removing hurdles” Enabling conditions to ensure • SMED - Reduce Changeover • Reduced rejection and improved
24 x 7 flow time FTR
3. Implement Zero Defect • Problem-solving techniques • Further reduction in WIP
Systems • Jidoka and Error proofing • Increased flexibility with shorter
• Pull Systems – KANBAN lead times
9 to 12 4. Standardize “Bringing 1. Improve equipment and • Key elements of Total • Increased equipment availability
stability” system reliability Productive Maintenance (TPM) – further gain in output
2. Maintain improvements – AM & PM • Maintain the improved metrics
through standards • Workplace standards through 5 • Self-managed “de-skilled”
S operations
• Visual SOP
13 to 18 5. Sustain “Eye on the 1.Install harmonized support • Continual Improvement (CI) • Internal trained resources to
future” systems (Synchronization) structuring manage the CI journey
2. Ensure continuity of Lean • Lean (5S) assessments • Foundation for Building Lean
initiative • Kaizen programs culture
Lean Implementation Methodology for SMEs
4.2 Lean Implementation Methodology 43

Secondly, one-piece flow is not explicitly indicated in the methodology. It is


expected that an organization reaching the sustenance stage will keep improving
the flow towards the ideal of one-piece flow in later iterations.
Lean is all about flow. The originators at Toyota laid emphasis on achieving
single-piece flow as the purest and ideal state of Lean. They also took the tough
path of radically altering the processes to create this flow right at the outset and
then overcoming the obstructions and constraints to this flow by solving the issues
as and when they arose. Our methodology is a slightly diluted approach in the
sense that it recommends creating flow wherever possible to make some quick
gains. We then work to maintain this flow by solving the issues related to defects,
line imbalances, changeovers or breakdowns, process variations, and administra-
tive issues. At times, when the issue is not quickly surmountable and flow gets
disrupted repeatedly, we may take a step back to define standard WIP inventory
and institute pull-based schedules to maintain the overall system output in line with
customer demand. But, we keep working in parallel to solve such issues perma-
nently. Further we would need to synchronize our resources, be it man, material,
or machine so that flow is smooth at all times.
Once issues have been solved and flow is streamlined, we write the improved
method of operations into new standards and make them visual. Equipment
standards for maintaining reliability and performance are also developed and
implemented at this stage. This ensures a smooth operation day after day without
any surprises or blockages.
In the last stage, we see how the internal team can take the learnings from one
such cycle of Lean implementation and devise ways to repeat such a cycle again
and again. Lean assessments are a powerful tool which can aid the Lean champions
in ensuring this sustenance. Very few organizations have actually evolved to this
level where Lean cycles are repeated year after year using such a framework. In
those mature organizations such as Toyota, it is part of not only the organization’s
DNA but also of its value chain partners.
In the next few chapters, we will be going through each of the stages of this
Lean implementation framework in detail with an emphasis on application in
diverse industry situations. Here, we would only like to add that this “generic”
methodology lays out the broad framework of a Lean implementation cycle span-
ning approximately 18 months. However, it is not meant to be rigid template to be
adhered to at all times. As we go along, we will understand the nuances of using
this methodology in different situations which are dependent on a combination of
factors such as business maturity, industry sector, management outlook, and social
and cultural factors. Let us first look at some ways to use this methodology to
initiate Lean at SMEs.
44 4 Lean Implementation Methodology for SMEs

4.3 Breaking Down the Barriers—Approaches to Initiating


Lean

Over the past three decades, Lean implementation has been led or guided by
experts who have learnt about it during the course of their employment in orga-
nizations like Toyota and its brethren. The automotive majors are the source of
a bulk of Lean experts who have then become independent consultants or joined
large global consulting companies that focus on big ticket clients. Their approach
to lean implementation begins with a detailed current state assessment and fram-
ing a roadmap with defined goals. However, this step itself may take several weeks
and involves a significant outlay by the client organization in terms of resource
time and consulting fees. With so many barriers already in the minds of a typical
SME, this approach has proven to be a non-starter. In this case, size does matter,
and we need to tailor the shoe to the foot.
The surest way to begin is to try to give a taste of the good things to come.
It is the initial sip of a good wine or coffee that slowly envelopes your senses
and makes you eager to have the whole glass and more. Likewise, we need to
begin with specific aspects of Lean that convince the management to invest in the
initiative, while at the same time exciting the employees and make them want to
be a part of the journey. We present a couple of approaches that have done just
that.

4.3.1 Rapid Process Improvement Approach

The current state assessment and roadmap are no doubt a logical way to go. But,
“how fast, can we move onto implementation, and show some results?”, is a ques-
tion asked by most SME prospects. Say, if within a week’s time, we take up a
couple of key areas identified during the assessment and rapidly improve them
using appropriate Lean tools and show results; it is sure to catch the attention of
the SME management. These “low hanging fruit” should culminate in quick wins
within 3–4 days. Through this approach, the SME would not only have a broad
roadmap for the Lean journey, but also would have already tasted the fruits of
Lean implementation all within a week’s time. Sounds great, does it not? Here is
a case example of “The Cable Guys”, a medium-sized organization in the UAE,
where this approach worked.

The Cable Guys: How to Improve Within a Week!!


About a decade ago, the real estate market was booming in UAE, and TKB
was well placed to supply a bulk of the cables required for the electricals
of these new constructions. It was expected that the boom would last just
4.3 Breaking Down the Barriers—Approaches to Initiating Lean 45

a couple of years and TKB would need to move fast to take advantage of
the situation. But, the management was in a dilemma as investing in new
equipment could prove unproductive once the market starts slowing down. It
was at this stage that the Lean consultants came in.
We met the top management and briefed them on Lean and its benefits.
While they were aware about Lean through reading and hearing about it,
very few organizations in UAE had attempted to deploy it. Hence, a lot of
scepticism on the outcomes. We did not have a resident consulting team in
the UAE and would have to travel there for this assignment. We then came
up with this idea of a week’s visit from Saturday to next Thursday (Friday
being the weekly holiday in Middle East). During the first two days, we
would do a Lean awareness session for the functional heads and key process
leaders and involve them in a diagnostic exercise. The next three days, we
would do improvements in a few identified bottleneck processes through
cross-functional teams. On the last day, we would present the overall Lean
roadmap, while the teams would present the results of the projects taken up.
TKB’s management agreed to this immediately; they would only have to
commit time and resources for a week.
And it went to plan. Four teams were formed, one for each product family.
Value stream maps were completed by each team and bottlenecks identified
in two days. The diagnostic showed scope for improvement in reducing raw
material inventory holding (44 days) and finished good inventories (23 days),
copper wastage in drawing processes and increasing the extruder OEE from
current level of 30%. Each team then took up one such area for improvement.
Within a week, the teams have managed to show a 30% reduction in copper
wastage at the bunching process, reduce the extruder changeover time from
35 to 10 min, and clear the testing bottleneck for the heavy fire-resistant
cable product line.
TKB now had a taste of the power of Lean and a roadmap to execute for
the next 6–9 months. Did they go for it? You bet they did. Albeit through
their own internal team. As consultants, we probably shot ourselves in the
foot by “revealing too much too soon”. But, the job was done—scepticism
was replaced by belief in the power of Lean. And, it gave birth to this
new approach for SMEs and Lean sceptics—the rapid process improvement
approach.

Over the last decade, this approach has been refined continuously through the
learning from each client experience. In the pandemic period, industries and Lean
experts are facing another set of pressures. Uncertainty in market conditions for the
industry. Travel and time spent at the clients are at a premium for the consultants.
Hence, the rapid diagnosis and quick implementation approach has become all the
more relevant as organizations continue to count their costs and remain unsure of
committing to approaches; they do not understand up front. This recent example
of how a ceramic industry owner converted to Lean in the middle of the pandemic
year is another testimonial to this quick-win approach.
46 4 Lean Implementation Methodology for SMEs

The Broken Cup… Fixing It Quickly!


Yash, a second-generation entrepreneur had been handed over the man-
agement of the ceramic crockery business by his father. A combination
of low raw material costs, availability of cheap labour, and high volumes
had so far kept the business profitable. But, Yash had set up a comprehen-
sive plant performance tracking information system and was unhappy with
the high percentage of rejections across the process. The manufacturing of
ceramicware comprises of two major stages:
Green (pre-firing) stage: any rejected product here can be reused by
charging it back into the ball mill.
Whiteware (post-firing) stage—rejection here is a complete loss, and the
item has to be either sold as seconds or scrapped fully based on the defect.
Yash was looking to improve margins by reducing these unwanted rejec-
tion costs. At a casual dinner with one of his friends, the friend spoke about
how Lean implementation at his plant was yielding positive outcomes. This
piqued Yash’s interest. Though he had never ever heard of Lean or Kaizen,
he got in touch with us immediately. He was not sure how it would help him
in his operations, but he was clear on his goal—reduce costs. As a qualified
finance professional, Yash was also very particular that any amount he spent
on Lean had to have a clear payback.
We went with the quick diagnostic and improvement approach. On day
one of a three-day visit, we organized the team of process in charges and
a factory manager into three teams, one each for the mixing, green form-
ing, and whiteware sections. Capacity not being the focus in this case meant
we could dispense with conventional current state mapping and assessments.
With the simple paradigm that any “touch” of the delicate ceramicware pro-
vided an opportunity for creating a defect; the teams assessed the current
state through a gemba walk and measured material handling in terms of
transportation, number of touches (pick up and place), and motion of peo-
ple. Simple computation highlighted that there were on average four touches
per value-adding operation and that people were moving over a kilometre a
day for material handling.
A roadmap was prepared for the next six months. Yash who was per-
sonally involved through the diagnostic exercise was excited to learn the
potential for improvement, but still sceptical on how it could actually be
realized. While we presented him a six-month roadmap, we decided to take
up a couple of quick wins immediately to convince Yash. Within a month,
projects were taken up in terms of productivity improvement in the final
packing line and reducing handling of fired ceramicware. In a second three-
day workshop, the packing team converted the batch process to a flow line
making some changes in workstation arrangement and layout. Productivity
of the team shot up by 50%. Similarly, by connecting the unloading of fired
ceramicware, to the inspection and subsequent design pasting processes in
4.3 Breaking Down the Barriers—Approaches to Initiating Lean 47

Component making Painting / design Assembly and


through Injection printing on packing of
Moulding components finished toy

Fig. 4.2 Toy manufacturing process

flow to replace the existing independently operating sections, the number of


touches was slashed. Within 15 days, the results were evident. A 50% drop
in handling damages and rejection of whiteware!
Having seen the actual improvement and the results with his own eyes,
Yash was now keen to move ahead on Lean His next question—“When can
we work on the next process?”

4.3.2 Problem-Solving Approach

A popular saying goes “The way to a man’s heart is through his stomach”. Well,
the way to an SME entrepreneurs’ mind is through solving his or her immediate
concerns. And if you can do that through Lean, then the seed of Lean thinking has
also been planted. While Lean is a journey, if application of a particular concept
or tool can solve a pressing problem, the management is likely to accept it and
undertake a more comprehensive journey. The generic methodology can always
be applied after alleviating the immediate concern. If the concern and roadmap
born out of the assessment coincide, then so much the better! The case of this
medium-sized toy manufacturer is a perfect example of this approach (Fig. 4.2).

Toy Story
The Chairman of a large conglomerate had recently paid a rare visit to one of
the oldest factories of their toy division. The toy division contributes just 5%
of the turnover and is a standalone business that rarely merits his attention.
But, this visit turned out to be a wake-up call. The division was in the process
of setting up a new factory to run a new product line for a global customer.
The Chairman was extremely critical after his visit to the existing factory
and made scathing observations about the inventories lying all over the plant,
man and material movements, and lack of visible operational controls. While
leaving, he told the CEO—“I will not allow you to go ahead with the new
factory if it is going to be run like this. Ensure that things are properly planned
and designed”.
48 4 Lean Implementation Methodology for SMEs

A chance meeting of the CEO with one of the group’s ex-employees,


now a technical adviser got us referred to him. The CEO had no idea about
Lean. He told us about the comments of his Chairman and asked for help in
designing the new factory operations. We were up to it but said we need to
have a look at the existing factory to understand the processes and operations
in this industry. A two-day visit threw up the same observations already made
by the Chairman. A typical engineering product set up—the operations were
divided into three stages.
It was a classic case for applying the typical Lean approach and using all
the TPS tools and techniques. But, the immediate problem was to design the
new factory. We gave a broad approach on how we would go about doing
that. The CEO was ready to have us go ahead with the project and was
impressed with our quick understanding of the nuances of the process. We
began with process design; the product analysis based on volumes agreed
with the customer was inserted into a product process matrix and equipment
capacity calculations were done.
At this stage, there was a problem in the new factory registration process
and a hold up in the expansion plan. A few months delay was anticipated
by the CEO. But by this time, he was convinced on what Lean could do to
improve operations in terms of productivity, inventory reduction, and space
utilization. He asked us to make a roadmap and initiate Lean in the existing
factory which had upset the Chairman, in the first place.
We were on for a full-scale implementation! The minute the CEO realized
how Lean could solve his problems, he had no hesitation in taking on an
attempt to transform the old factory.

The potential for starting a Lean initiative through problem solving is immense;
but, the key lies in understanding the problem and then relating it to the application
of Lean concept or tool. Many times, the SME owners are able to define their
problem only through vague symptoms. We need to sift through these to identify
the root cause of the problem so that we can then solve it through Lean. In this
case of a seafood unit, the owner’s problem was that his team seemed unable to
perform to earlier levels. Read on to see how Lean provided the solution in this
case.

Growth Pangs in Feed


A few years ago, we met the Managing Director of a seafood company
through a common reference. The seafood industry had been ravaged by a
global recession due to disease, and the company had been operating at just
10% of it’s capacity in the past two years. The MD explained his problem
to us.

“About three years ago, we set up a new prawn feed mill to enhance our
existing plant capacity by about 33%. Then global recession struck the
4.4 Summary 49

industry and we were literally shut for almost two years. Now demand has
started picking up to the earlier levels, but my team is unable to manufacture
the required quantities. In fact, the factory is struggling to reach 75% of
the earlier output and the team has asked me to add another mill to meet
targets. I am unable to understand what is happening with our factory ”.

The feed plant was operating from a remote location in proximity to some
of the natural agri-based raw materials. On our visit, there we looked at the
daily production data captured in plant logs. And quickly compiled a rough
Overall Equipment Effectiveness (OEE) calculation for each of the mills
(a mill is an independent feed processing line). The OEE was just about
50%. We knew where we had to focus. The next step was to understand the
functioning of the plant team. A quick two-day diagnostic workshop with
the functional heads got us to observe them closely. And what struck us
immediately was the extent to which they were working in their functional
silos. It appeared that two years of inactivity had led to an undercurrent of
conflicts that were now manifesting themselves on the work front.
Our task was cut out. We need to quickly show an improvement in the
OEE. And we could do this only by getting the people to work together in
a focused manner. This meant forming cross-functional teams to selectively
apply Lean tools. And it worked. Initial focus was on process inconsisten-
cies or Mura which was impacting the quality of feed. This in turn had led
to the operators lowering the feed rate of the mill resulting in a poor per-
formance ratio. As teams went onto the gemba together to observe the 3 Ms
and implement countermeasures, they rediscovered their bonding and things
started happening.
In six months, the OEE was at 79%! The journey continues.

4.4 Summary

Lean implementations at large enterprises begin by stabilizing the current pro-


cesses, followed by flow, pull, and single-piece flow, given the TPS heritage and
the general approach followed by big consultants. Further, the project initiation
is itself a 2–3-month affair to do a detailed current state, future state mapping,
and the gap analyses. Considering the various constraints facing the SMEs, it is
understandable that such a front-loaded methodology has not found favour. We
have come up with a more practical implementation methodology that is judicious
mix of delivering quick results, building excitement, and also preparing the SMEs
for the long-term discipline that Lean entails. We have used two ice-breakers to
showcase the power of Lean at SMEs equally successfully; one that shows a rapid
improvement of a bottleneck process and thus improve throughput, and another
50 4 Lean Implementation Methodology for SMEs

that solves an immediate problem facing the SMEs. While, these approaches gen-
erally result in the SME signing up for the full-scale implementation, there is an
occasional risk of client dropping out after this phase, as their immediate concern
is resolved, and they have learnt to take Lean forward themselves. In any case,
the larger purpose of implanting Lean into the minds of the SME would still have
been served.
Commencing the Lean Journey
5

5.1 Introduction

In the previous chapter, we laid out the methodology best suited for designing and
implementing a sustainable Lean intervention in SMEs. The journey begins with a
diagnostic, and in this chapter, we explain how organizations can assess their cur-
rent capabilities in context of their business goals. As we saw earlier, organizations
often lack clearly defined objectives, may err both in the choice of metrics, and
in the way they are measured and reported. When designing a Lean intervention,
one would do well to fix the destination first and then go about defining the route
to get there. Implementing Lean should not be like taking ones’ sports car or bike
out for a spin because one feels like it. It needs a clearly stated purpose, which is
captured in a set of measurable goals. The current state of operations should then
be assessed with respect to these goals and a roadmap drawn up to achieve them.
Lean’s biggest attraction lies in its promise of “Doing more and more with
less and less”. The generic definition of productivity is output/input, and conven-
tional focus for improving productivity has been in trying to reduce resource(input)
and thereby cost. Lean management adopts a holistic perspective with the cur-
rent state assessment and target-setting activities focusing on both the value
generation(output) and the resource consumption (input).

• Value: Value comprises of everything that the customer desires (quality, quan-
tity, features, responsiveness, etc.) Time-to-serve is a direct metric that measures
the organization’s ability to deliver value to customers. We measure what the
current output is and target how much more can be done with the existing
resources.
• Resources: Resources refer to the people, machines, materials, space and time
that are utilized to generate value. Cost-to-serve is a comprehensive measure

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 51
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_5
52 5 Commencing the Lean Journey

of all the resources consumed in delivering value to the customers. We mea-


sure quantum of resources consumed and target how much we can reduce thus
consumption for the same or enhanced value.

These two parameters mirror the entire operational performance of a given


enterprise. In a well-thought Lean intervention, only those resource optimizations
are pursued that do not hurt the company’s ability to deliver value to its customers
in the short and long terms. We need to have a frame of reference to set out
on our Lean journey and the following paradigm succinctly describes what a Lean
organization should possess:

The capability to satisfy customers on quality, delivery, and price


through
fast and flexible processes that quickly respond to customer needs
utilizing
available resources effectively, thereby incurring the lowest possible
cost and
operating
a visually managed and self-regulating facility.

Two essential goals emerge from the understanding provided in this frame of
reference:

Flexibility of the set of processes that together deliver the value in terms of
product and/or service to the customer and best measured through time-to-serve.
Productivity of the resources that go into delivering this value—be it man,
material, machinery, or infrastructure and best measured through cost-to-serve.

In fact, right from its origins in Toyota in the 1950s, Lean has been about
delivering what the customer wants in the shortest possible time and lowest cost
without sacrificing process efficiencies.
In this chapter, we discuss how to diagnose the current state of operations and
establish the targets that would go into the framing of an improvement roadmap,
the next stage in the proposed Lean implementation methodology.

5.2 Time-to-Serve

Reducing the time-to-serve or lead time for customer requests has a number of
advantages. Lead time is defined as the total time taken from the start to the end
point in a business process. In manufacturing, this is generally the time taken from
the raw material receipt to the finished good stage. Common sense tells us that
shorter the lead time faster is the response to (changes in) customer requirements.
5.2 Time-to-Serve 53

It is important to understand that lead time is a direct function of the inventories


lying in the value stream.

The Office Commute


Take the example of a daily morning peak hour commute from your home to
office covering a distance of, say, eight kilometres (km). The average speed
considering the speed limits and traffic lights on the road maybe 40 km
per hour, implying one can reach office in about twelve minutes. But in
reality, the journey takes a good 25 min. Why? Well, you find that you have
to pass through a series of traffic lights at major junctions. At each traffic
light, you wait anywhere from three to five minutes, before going through,
depending on the length of the queue (inventory) of vehicles in front of you.
You cannot reach earlier, because you can go through only after the vehicles
already in front of you. Now say you have switched to an early morning
work schedule. At this hour, there are hardly any vehicles in front of you
at any of the lights and you sail through in fifteen minutes. The reduction
in wait time has enabled you to be much faster and therefore you would
be much more responsive in the case of an emergency. This same logic (less
inventory⇒ less lead time) holds for all the material and information flows
in business processes.

Taiichi Ohno simply summarized TPS in this one sentence:

All we are doing is looking at the timeline from the moment the customer gives us an order
to the point when we collect the cash. And we are continuously reducing that timeline by
reducing the non-value-added wastes.

The benefits of lead time reduction are also captured in this observation by
Womack and Jones in their book, “Lean Thinking: Banish Waste and Create
Wealth in Your Corporation” that, “every time a company quarters its lead-time
it can expect to double its productivity and reduce costs by 20%”. Let us see how
Rockwell, an SME, turned around its operations by reducing lead time.

Lead Time Impact on a Seasonal Industry


Rockwell manufactures commercial cold storage products such as freezers,
bottle coolers, and cold chests. With an established brand, the company
was blessed with a growing demand but unable to meet it during the peak
February–May season which accounts for 75% of the annual turnover.
54 5 Commencing the Lean Journey

Rockwell therefore implemented a strategy to manufacture additional


machines of various product variants in advance of the season and main-
tain adequate stocks. For this, additional warehouse space had to be leased
and machines moved from the factory to this warehouse at the company’s
cost. The owners also started planning expansion of the factory to increase
capacity. At this juncture, they were referred to a Lean Sensei and decided
to take his help.
The current state assessment began with measurement of the manufac-
turing lead time which was found to be about ten days. So, orders for a
not-in-stock product variant therefore required about two weeks to fulfil and
that too only if all the assembly parts are available. The marketing team was
losing several orders to competitors as a result of this lead time. The Lean
initiative was designed to first attack and reduce the lead time.
Two years into the lean journey, the lead time was slashed to less than two
days, off season stocks kept to a minimum defined level, the external ware-
house discontinued, and expansion plans deferred. The pressure had shifted
to the marketing team who had to ensure they placed only confirmed orders
to the factory as any product could be made and delivered within the week.

5.2.1 Measuring Time-to-Serve

Measuring the lead time is not as simple as it sounds and is ridden with a num-
ber of practical difficulties. Some approximation techniques relevant for different
contexts are illustrated below to get an indicative value of this number. Of course,
in keeping with the Lean principle of “Go and See”, the preferred technique is
direct observation.

5.2.1.1 Direct Observation


A fundamental principle of Toyota Production System (TPS) is Genchi Gembutsu
(Go See for yourself ). One way to do this would be to make an indelible mark
on the primary raw material before it is issued to the first stage of manufacturing
or on the document as it starts its journey through the value stream. The date and
time of issue are recorded. The same marked item is also recorded at the end point
on becoming a finished good/item. The lead time is simply the end time and date
minus the original issue time and date.
This technique can be used in some discrete manufacturing industries where
such a mark can be traced as well as for document or information processing
service sectors that generate a time stamp at each process. In today’s world, use of
Industry 4.0 technologies such as barcodes, QR codes, or RFID tags enable this
traceability with ease and accuracy .
5.2 Time-to-Serve 55

5.2.1.2 Batch Traceability


There are many industry sectors where the processes are hidden from direct view,
and it is impossible to “mark” the material. The chemical, pharmaceutical, met-
allurgical, and food processing industries are prime examples where we cannot
observe most of the processes with our own eyes in the first place.
However, these industries already adhere to a strict batch traceability standard,
thus making it easy to compute lead times directly by using the time stamps in
the records. In such cases, we try to trace the lead time through available batch
records. The snippet below is an example of how the aluminium rolled products
industry ensures traceability.

Aluminium Coil Identification System


Way back in the 1980s, the aluminium rolled product industry had devel-
oped a dual identification system combining “marking” and “recording”. The
slab/sheet coming out of the initial casting process is physically “marked”
with an ingot/coil number and the date and time of production. Each coil
also had its own Coil Card or product route sheet with a provision to record
details at each stage. After each rolling or heat treatment stage, the mark
gets erased, and the coil number, date, and time of completion of that opera-
tion are marked once again using permanent markers. More importantly, this
information is also recorded in the Coil Card at each stage. Post the final
finishing stage operation, the Coil Cards return to the central Production
Planning Cell (PPC) who were then able to calculate the lead time directly
by the time difference between the first and last entries.

5.2.1.3 Value Stream Mapping—Computational Method


In “Learning to See”, Rother and Shook explain a logical way to compute the
lead time from the value stream map (VSM) of a product family. Organizations
producing single product families using dedicated resources will find the results
from VSM to be a reliable indicator of their lead times. Early in this chapter, we
have seen how lead time is directly related to the inventories, and this is in fact the
basis for the VSM method of computing lead time. A few definitions pertaining
to the VSM are in order:

• Total Lead time or Throughput time = Sum of processing lead times + Sum
of inventory (waiting) lead times across the value stream,
• Processing lead time = Sum of cycle times + Sum of changeover times,
• Cycle time is defined as the time between two consecutive parts coming out of
a process.
• Changeover time (Setup) is the time between last good product of one variety
to the first good product of another variety in the same process,
56 5 Commencing the Lean Journey

• Inventory Lead (Wait) time = Quantity of inventory/Daily customer requirement


or Quantity of inventory in the value stream x Highest cycle time (bottleneck).

This snippet “Plastic Toy Assembly line” illustrates how to compute lead time
using VSM.

Plastic Toy Assembly Line


The dedicated line has five operations, A, B, C, D, and E with observed cycle
times of 50, 60, 45, 55, and 60 seconds, respectively. The total inventories
across the line were physically counted to be 100 pieces. In any value stream,
the process(es) with highest cycle time decides the rate of output, and in this
case, these are processes B and E which produce at the rate of one toy per
every 60 s. With this information, we now compute the lead time for the
line, using the definitions above as follows:
Sum of all cycle times per toy = 50 + 60 + 45 + 55 + 60 = 270 s or 4.5
min
Changeover time is zero, as it is a manual process with dedicated line.
Processing Lead time = 4.5 + 0 = 4.5 min
Total Inventory Waiting time = (100 toys × 60 s per toy (60 s/ minute) =
100 min
Each toy waits in the inventory for an average 100 minutes
Therefore, manufacturing lead time = 4.5 + 100 = 104.5 min

The accuracy of this method is better for high volume (“runner”) products with
dedicated facilities. The computation gets complex with the increase in product
groups with different cycle times and machine set-ups running through the same
set of machine centres. We may add here that the lead time is not just a measure
of the level of flexibility inherent in the current process but also an indicator of
the output capability and we will discuss this in the following section.

5.2.2 Assessing Output Capability

Time-to-serve also captures the current throughput levels or the quantum of


product or service delivered to customer. This is the value generated by the organi-
zation. We now assess the capability of the current process to deliver the required
output, using the following three steps
5.2 Time-to-Serve 57

1. We first calculate the takt (German word for meter) time. Defined as the rate of
customer demand, it is simply calculated as the available working time/demand.
So, if the demand is 450 pieces per day and the factory works 7.5 h (actual work
time), then takt time is 7.5 × 60/450 = 1 min or 60 s per piece,
2. We then calculate the expected output under current conditions. As we saw
previously, VSM is a structured tool that requires us to measure cycle times for
each operation. We use this data to identify the “bottleneck” operation which
decides the output capability of the entire value stream. Normally, the operation
with the highest Effective Cycle Time (ECT) is the bottleneck and this ECT
determines the current output capability. the ,
3. We now compare the ECT with the takt time; ECT ≤ takt time implies that
the current value stream has the capacity to deliver the targeted output.

The method of computing ECT from observed operation cycle time and using it
to calculate expected output varies from industry to industry and depends on fac-
tory specific factors. The following subsection presents a few scenarios to illustrate
the differences.

5.2.2.1 Discrete Processes


If the same operation is being done on multiple machines or production lines, the
ECT will factor in the available capacity both within a machine and across the
parallel machines. The example below shows how this was done at a small ceiling
fan manufacturing unit.

Motor Winding at a Ceiling Fan Manufacturer


In a ceiling fan manufacturing unit, the motor winding process is critical.
The coil winding machine winds two coils at a time. The unit has two such
coil winding machines in use.
Observed Cycle time = every 150 s, two wound coils are unloaded from
each machine
Effective Cycle Time (ECT)
= {(observed cycle time/ output Per cycle)} /number of machines
= {(150/2)}/2 = 37.5 s
This means that coil winding process has an ECT of one-fourth of observed
cycle time and a corresponding capacity. Since this process has the highest
ECT, it is the “bottleneck” and decides the current capacity of the unit. The
factory operates a single shift at present. Taking the actual working time of
7 h, the capacity would be:
Capacity Operating time / ECT = (7 × 60 × 60) / 37.5 = 672 motors per
day.
58 5 Commencing the Lean Journey

5.2.2.2 Batch Production


In chemical and metallurgical processes, the value-adding process is often a chem-
ical reaction which changes the internal properties of the material including the
mass of the product. For example, in a drying process, moisture is removed result-
ing in weight loss of the obtained output. In reaction processes, the output may
be more or less than the input depending on factors such as process yield and/or
molecular weights of the compounds. In order to find out the bottleneck, we would
therefore need to normalize the output of each stage in terms of a common defined
output while making the VSM—this is generally done through a conversion factor
we call the batch equivalent.
Let us take this example of a bulk drug manufacturing unit which follows a
batch process. Here, we have converted all interim outputs in terms of finished
product batch equivalent output of 500 kgs. This is done through the following
steps:

Step 1: Calculate the takt time (in terms of hours per batch) based on target
monthly demand (tonnes/month) and available working hours as shown
in Table 5.1.
*Assuming the production runs 30 days in a month and 24 hours a day

Step 2: Yield factor for each stage is known based on chemical formulae. We
now work backwards from the final stage computing the input at each
stage as being equal to the output/yield. We need to calculate the input as
in this industry the capacity of the equipment (reactor) is defined by the
quantity charged into it.
Step 3: The input is then used to compute the batch equivalent of each stage
based on number of equipment available at that stage. Continuing the
same example if reactor capacity is 1200 kg and this input gives us 1000
kgs or 2 batches of final output and we have two such reactors, then the
batch equivalent for this stage is (# of batches of output per input batch
x # of reactors) = 2 × 2 = 4 batches of final output.
Step 4: The Effective Cycle Time (ECT) per batch = {Operation cycle time/batch
equivalent}. For the same bulk drug manufacturing unit, this is calculated
as shown in Table 5.2.

To reiterate the computation method let us look at Drying II process. Here, the
output per drier is 350 kgs, the finished goods batch size is 500 kgs (Table 5.1),
and there are 3 driers available. Hence, then batch equivalent = {(350/500) × 3}
= 2.1. The average cycle time for a drier based on the observed data from last

Table 5.1 Takt time calculation for bulk drug manufacturer


Current Target
Requirement A (Tons/month) 28 40
Batch size B Kgs/batch 500 500
Demand rate* C = {A*1000/B}/30 Batches per day 1.9 2.7
Takt time* D = 24/C Hours per batch 13 9.0
5.2 Time-to-Serve 59

Table 5.2 ECT calculation for bulk drug manufacturing


Operation activity Available equipment Observed cycle Effective cycle
time time
Nos Batch equivalent (Hours) (Hours per batch)
1. Reaction 5 5 60 12.0
2. Neutch filter 1 1 11 10.6
3. Dissolution 1 1 6 5.8
4. Carbon 3 1.5 18 11.7
5. Acidification 3 3 16 5.3
6. Drying I (350 kgs 4 2 20 10.0
per drier)
7. Methanol purification 1 1 11.8 12
8. Filtration 3 3 21 7.0
9. Drying II (350 kgs 3 2.1 21.4 10.2
per drier)
10. Milling 1 1 2 2

20 batches was 21.4 h. So, the effective cycle time (ECT) = {Observed cycle
time/Batch equivalent} = {21.4/2.1} = 10.2 h per batch.
Further, we can see reaction (process number 1 in the above table) has the
highest ECT of 12 h per batch. This is then the bottleneck operation having the
current output capability of (24 h/day)/(12 h/batch) × (30 days/month) = (60
batches/month) × (0.5 tonne/batch) = 30 tonnes of finished goods per month.
We compare the current ECTs with the target takt time of 9 hours per batch
(Table 5.1), and see that the three processes (1. reaction, 6. Drying I, 9. Drying II)
are going to be bottlenecks to achieving the target output of 40 tonnes per month.
These three processes are highlighted in Table 5.2.

5.2.2.3 Factors beyond the ECT


In a stable process, the operation with the highest ECT should ideally determine
the capacity. However, other factors such as equipment downtime due to break-
downs or changeovers, imbalances in material planning or fluctuations in operating
staff productivity, and raw material quality also contribute to the degradation of
actual output of an operation.
Therefore, the actual output capability of the bottleneck operation should be
computed giving due consideration to such factors. The following examples illus-
trate the adjustments to be made to ECT-based capacity calculations in two
different scenarios—one where the process is predominantly manual and the other
where it is equipment driven.
60 5 Commencing the Lean Journey

Manual Process—Pump Assembly Line


MWM pumps assemble monoblock pumps on a manual assembly line. The
pump testing operation was found to have the highest ECT of 4 min. The
employees work from 9 am to 6 pm with an hour’s break for lunch and tea.
Expected Output = Available time/ECT
= (8 hrs per shift × 60 mins /hr)/( 4 mins per pump)
= (480 mins / day) /4 mins/ pump)
= 120pumps per day
However, in practice, it was observed that actual testing operation began
at 9.15 am and ended by 5.45 pm. Also, the testing stopped for 10 min every
couple of hours to clean the water lines (3 times in an eight-hour shift)
Actual run time = Available time - losses at shift ends
- losses due to water line cleaning
= 480 min - (15 + 15 ) min - (10 × 3) min = 420 min
This purely manual process demands a fatigue allowance of 5%
which means an operational efficiency of 95%
Expected gross output
= Run time × Efficiency / ECT = {420 min/shift × 95 %}/ (4mins / pumps)
= 100 pumps/ shift
The past data showed that on an average 5 pumps were rejected each
day after testing process.
Daily expected output = 100 - 5 = 95 pumps
This is the daily expected output under current conditions, and we can
see that it is 20% lower than that computed based solely on the ECT. In this
case, the output closely matched the actual daily recorded output, and the
testing process which had the highest ECT was confirmed as the bottleneck
process even after the above adjustments.

Equipment-Driven Process—Welding
In equipment-driven processes, the Overall Equipment Effectiveness (OEE)
is a key metric to arrive at the output capability of the bottleneck process.
5.2 Time-to-Serve 61

A battery manufacturing facility works two shifts of 7 h each of actual


working time. The manufacturing process steps include—electrode plate
formation, frame welding and pressing, plate grouping, cell assembly, for-
mation, battery assembly, testing, and packing. After preparing the VSM, the
team initially identified the plate grouping activity as the bottleneck with the
highest ECT of 24 s per plate based on the cycle time of a standard 13 plate
combination. This operation runs for 13.5 h as half an hour is lost for the
shift change activities, and the current expected output is based on 13.5 h
run time.
Capacity = Actual Run time / ECT = {13.5 hrs / day × 3600 s /
hr} / (24 s / plate)
= 2025 plates per day
The team then verified the expected output from each of the processes in
a similar way to rule out any other possible bottlenecks. And it so turned
out that the frame welding process also turned out to be a bottleneck process
in spite of a lower cycle time of 22 s only. The computation for frame
welding:
Recorded average breakdown in month = 8 h
Daily downtime = {8 hrs / month × 60 mins / hr} / {24days / month}
= 20 mins / day
Hourly stop for cleaning of electrode : 5 min
Average number of stops per day for electrode cleaning = 12
Actual Run time = Available time - downtime losses due to electrode
cleaning
- breakdown losses = (13.5 hrs / day × 60 min / hr)
- (12 stops × 5 min / stop) - 20 mins = 730 min
Capacity of frame welding = 730 mins / day × 60 s / min} / (22 s / plate)
= 1990 plates per day
Thus, the frame welding turned out to be the real bottleneck in this case,
even though it does not have the highest ECT.

It is important therefore to calculate the output capability for all operations


regardless of their ECTs. The Effective Cycle Time gives the expected time per
62 5 Commencing the Lean Journey

unit output and we then factor in equipment/human availability net of stoppages


and quality rejections to arrive at the expected output at each process. The bot-
tleneck process is the one which delivers the lowest output in current conditions.
Tip: We validate this through direct observation at the gemba (Japanese word
meaning “real place” or place where value is created). Operations before which
large inventories are piled up are likely to be bottlenecks.
The expected output should more or less match with the actual recorded output.
If the actual output is lower, the reasons are likely to be found in unrecorded
performance issues such as short stops or misaligned supplier processes. These
issues (termed as clouds in VSM parlance) need to be further analysed and are to
be included in the improvement roadmap.

5.3 Cost-to-Serve

After time-to-serve, we now move our attention to cost-to-serve. Fundamentally,


resources are consumed to produce the output, and the resources cost money. Some
of the resources have long been paid for (fixed assets such as plant, machinery,
warehouse), some are paid as they are used (materials, direct labour), and some
resources cannot be directly costed such as clean air, water, and environment.
Though we will restrict our discussion to the first two categories of resources, we
should note here that amongst all production systems, Lean implemented in its
true spirit results in better environmental sustainability.

5.3.1 Measuring Cost-to-Serve

The highly evolved automotive industry targets reduction in overall manufacturing


cost by a certain percentage each year regardless of any rise in raw material prices.
These ambitious targets are also passed on their vendors, the auto-ancillary units.
So, how are they able to keep reducing costs year on year? The answer is that Lean
has become an integral part of their culture over the last few decades, leading to
an ever-increasing productivity of their key resources.
Better productivity implies an effective utilization of available resources to meet
the target output. An important aspect to note here is that effective resource uti-
lization is viewed through the "Lean Lens" of Value Addition versus Muda (or
waste). For all practical purposes a value-adding process can be simply defined as
"A process step that transforms the shape, property, characteristic or feature of a
product or service in the direction specified by the customer". Any process step
that is not value-adding is by default a Muda or waste activity. All activities incur
some expense. In our parlance, the portion of expenditure spent on value-adding
activities is considered as cost and the rest is wasteful expenditure. Therefore by
reducing Muda, we can minimize wasteful expenditure and thereby the overall cost
of manufacturing. Table 5.3 summarizes the key resource productivity measures
and their impact on manufacturing cost.
5.3 Cost-to-Serve 63

Table 5.3 Key resources and their productivity measures


Resource Productivity Measures Impact
Time Lead time Delivery performance
Value Add Ratio Response to customer needs
Material Inventory turns Cost of inventory
Rejections & Losses Cost of Poor Quality
People Productivity Expense on non-value adding
% Value Add people
Infrastructure (facility) Proportion of value adding Expense on non-value added
space
Distance moved by material Expenses due to unwanted
Number of touches handling and transport
Equipment/machinery Overall Equipment Effectiveness Unutilized (lost) machine
(OEE) hour expenses
Loss of opportunity costs

The advantage of using these performance metrics is that they do not require to
be benchmarked with others. An organization striving to improve on it’s current
performance can consider itself to be moving in the right direction. The goal of
current state assessment is to collect data, calculate process performance on these
metrics, and call out the cost impact of current performance levels. We will now
discuss the approach used to compute each of these metrics.

5.3.1.1 Time
We have earlier discussed the importance of lead time, the various methods used
for measuring it as well as how capacity is a function of time. The value stream
map (VSM) is the assessment tool which primarily deals with time as a resource.
A key metric is the value-adding (VA) ratio; this is the proportion of the time when
the material is undergoing actual transformation (value addition) versus total time
spent by it in the factory. A higher VA ratio indicates that the material is flowing
relatively faster through the value stream to reach the customer.

VA Ratio
Total sum of cycle (processing) times of all operations to deliver unit output
=
Order to Dispatch lead time
The VA ratio is a good indicator of the level of Muda (waste) across the
processes as this example of a toy assembly line illustrates.

Toy Assembly and Packing Continued…


There are five operations with observed cycle times of 50, 60, 45, 55, and
60 s, respectively, and the count of inventories across the shop floor was
100 units. Recall from the previous mention of this example (Sect. 5.2.1.3)
that a 100-unit inventory implies 100 min of inventory/waiting time, given a
maximum cycle time of 1 min.
64 5 Commencing the Lean Journey

Total sum of cycle (processing) time per toy


= 50 + 60 + 45 + 55 + 60 = 270 s or 405 min
Lead time = Waiting Time + Processing Time = 100 = 4.5 = 104.5 min
VA Ratio = Processing Time = 4.5 / 104.5 = 4.3 %
In other words, the material spends 95.7% of the total lead time without
any value addition and is just idle inventory. This WIP popularly termed
as “Waste-in-pipeline” leads to lowering the inventory turns and increasing
wasteful expenditure of storing, handling and transporting the material.

5.3.1.2 Material
From a manufacturing industry perspective, it is evident that customers pay for the
product or material that meets their specifications. The factory transforms the raw
materials to the customer-specified finished product through a series of processes.
All value addition is defined on the basis of transformation of material. According
to Lean paradigm, any material that does not reach the hands of the customer
is a waste or unproductive use of resources. During current state assessment, we
measure this fraction of material as rejection and/or loss.
Let us take the example of a sheet metal fabrication unit where the first pro-
cess is a CNC laser cutting process for cutting discs out of a 25 mm thick steel
plate; these discs are later welded to form the base for a solar pump panel support
structure. Figure 5.1 depicts the different types of material losses that occur in this
process.
The yield loss is the part of the steel plate that is not included in the discs,
and it depends on how best we design the cutting pattern to maximize sheet uti-
lization. Design loss is the material removed from inside the circle to give the
ring shape and this is due to inherent design of the product. In fact, this is where
the additive manufacturing technologies such as 3D printing have an advantage as
it is based on adding the material as opposed to removing material to form the
part. These new technologies are yet to catch-up on scale and cost of production.

Scrap

Rework

Final Output
Yield Design Rejection
loss loss

Fig. 5.1 Different types of material wastes


5.3 Cost-to-Serve 65

Rejection, one of the seven wastes of Lean, is loss due to defects where the output
is not in accordance to customer (next in line process) specification. The rejection
may be dealt with in different ways—it could be reworked, sold as seconds, or
scrapped, depending on the type and severity of the defect. The yield computa-
tions are slightly different in chemical and process industries, as explained in the
section on machinery.

5.3.1.3 People
Human resource productivity has been measured right from the start of the indus-
trial engineering era. Typically, output per person per day is used as a productivity
measure across industries, e.g. number of toys per person per day, litres of oil
packed per person per day, kilograms of seed processed per person per day, etc.
But these measures are based on a narrow definition of output and do not provide
direction without benchmarks with best-in-class performance levels at that point
in time. People productivity measurement under Lean paradigm addresses these
limitations and is based on the broader understanding of value addition. Let us
recollect here the fundamental principle of Lean—in a product environment, cus-
tomer is paying for the material, and therefore, only those activities that directly
transform the material are seen as value adding. All other activities by default are
considered as waste (Muda). By this definition:

Human Resource Utilization = Time spent on value addition/Total time


available

Aggregating this for the total human resource employed across the organization
tells us what proportion of the total manhours is spent in wasteful activities and
can potentially be redeployed into value-adding work, thereby raising the overall
productivity. Figure 5.2 illustrates this segregation of value-added and Muda times.
In organizations, people work typically in core operations or in support roles.
Core operations involve directly transforming the material, information or ser-
vice, a value-added activity. Support functions include finance, material handling,
administration, quality control, and shop floor supervision. Note here that a QC

Human Resource

Core Operation Support Activity

Value Adding Time Muda time Muda

Fig. 5.2 VA and Muda times for human resources


66 5 Commencing the Lean Journey

inspector does not add any value but only performs the function of gate keeping.
Under Lean paradigms, the entire time of the people involved in support functions
is Muda as none of them are directly involved in transformation. Further, even in
core operations, the part of the time spent on actually transforming the material
is only considered as value-adding and the time spent in other activities includ-
ing waiting or idling time is Muda. This example from a shrimp processing unit
illustrates the concept (Table 5.4).

Shrimp Processing Plant


Within the core operation, we can see that there is cycle time variation among
the operations, and the flow is determined by the slowest operations (2. cut-
ting and 4. sorting and setting). The people working in other operations will
therefore have some slack or idle time (to the extent of difference of cycle
times). For example, the idle time of 1st operation (Peeling) = 46–42 = 4 s.
Total cycle time = 42 × 3 + 46 × 4 + 42 × 2 + 46 × 8 + 36 × 9
= 1086 person - seconds
Total wait time = 4 × 3 + 0 × 4+ 4 × 2 + 0 × 8 + 10 × 9
= 110 person - seconds

Table 5.4 Shrimp processing diagnostic


People
Process Support Operation Cycle time (sec) Idle time (sec)
1. Peeling 3 3 42 4
2. Cutting 3 4 46 0
3. Soaking 2 2 42 4
4. Sorting and setting 8 8 46 0
5. Freezing, packing, and storage 6 9 36 10
Total 22 26 210 18
Dispatch section 20
Receiving section 5
TOTAL 47 26
5.3 Cost-to-Serve 67

Repairs & maintenance Rental cost

Space Costs

Housekeeping

Lighting, air conditioning /


heating /cooling

Fig. 5.3 Typical space-related operating costs

So, value-adding ratio within core operation = Value-added time/(value-


added time + wait time) = 1086/(1086 + 110) = 90.8%
Proportion of people involved in core operation = 26/(47+26) = 36%
Overall value-adding ratio for People = 90.8% of 36% = 32.7%
This means that of the total manhours available, only 33% is spent on
actual value addition.

5.3.1.4 Facility Infrastructure


Facility infrastructure refers to both the total space as well as the equipment avail-
able at the factory, and we would like to assess the utilization of the built-up area
provided for operations. Two kinds of costs an be attributed to the facility infras-
tructure; the first is the cost of the space itself, and the second is the expenditure
related to material handling and transport within the facility which is a direct
function of the design (layout) of the space. We discuss each of these elements
below.

Space Utilization
People can be often heard grumbling “We do not have any space to work” or “We
can make more but there is no space available, it is so cramped”. This invariably
leads to management decisions such as taking an extra space on lease or invest-
ing in factory expansion to overcome these space constraints. Before making such
an investment, it is important to assess the current space utilization as the exist-
ing space incurs operating expenses as seen in Fig. 5.3 and creating new space
would require a substantial investment.
The value-adding ratio for space is the proportion of the total available built-
up area that is being used for actual value-adding operations. The total value-
adding area is the sum of all machine and workstation footprints each of which
are physically measured. The “mandatory” space needed for operations including
for gangways, electrical panels, safety, and environmental requirements is also
worked out. The rest of the space is now potentially “available” as it is either idle
68 5 Commencing the Lean Journey

Fig. 5.4 Vicious cycle of


layout and material storage Processes located
far apart

Space blocked by Material needs to


WIP be transported

Material has to be Increased handling


stored -waiting of material

or occupied by the Muda of inventory or non-useful items.

Effective Space Utilization Ratio


Total value adding area + Mandatory space area
=
Total available factory space area

The Layout Effect


The way the equipment and workstations are arranged determines the extent of
material handling and transportation and movement of operators in an operating
unit. It is now an established fact that the facility layout is the single most impor-
tant factor in determining the level of Muda related to material storage, handling,
and transport. Close interlinking of supplier and customer processes is the key to
smooth material flow through the value stream. The layout can ensure the physical
proximity required for this interlinking. All factories grow, morph, and upgrade
over the years and processes, if located farther apart lead to a vicious cycle of
increased material handling, transport, and storage (see Fig. 5.4) resulting in a
further blocking of space which in turn leads to new machines being placed even
further away.
The “Muda Walk” is a tool used for assessing the impact of the current layout.
The idea is to put oneself in the shoes of the material (in a manufacturing unit) or
of a customer or service provider in a service facility. You then walk through the
value stream from start to finish tracing the exact route taken by the material or
the customer/service provider, and on the layout chart, develop the material flow
diagram.

The material flow diagram is popularly known as the Spaghetti diagram. In


most legacy plants that have grown over time, the crisscross flows of the
material traced through the Muda Walk often resemble a bowl of noodles.
5.3 Cost-to-Serve 69

Machine Shop Anodising Section

Machining Drilling machines Compressor


Hard Bath 2
Centre 1

Receipt of Gripper rails


White Bath 2

Hard Bath 1
Machining
Centre 2

Receipt of carding plates


Buffing
4 section
Pattern
2 5
machining Machining
section Centre 3 (4)
White Bath 1 (2)

3 (3)
1
Degreasing / 6
Bend/Twist removal Neutralizing Packing / Dispatch area
section section

Tube receipt,
Store Store Office
dispatch and
cutting section
Receipt of tubes (1)

Tube movement

Carding Plate movement

Fig. 5.5 Material flow diagram for an aluminium products unit

Figure 5.5 illustrates a material flow diagram for an aluminium products unit.
Layout effectiveness is best measured through these two parameters:

1. Distance travelled by the material: as we trace the route of the material through
Muda Walk, we keep count of our footsteps and convert this to an approximate
distance travelled by the material within the premises, For example 10 steps
from Process A to an intermediate storage point is equivalent to 20 feet of
material movement.
2. Number of touches: one of the fundamental lean principles is “One touch to
Value-Add”. This means each time the material is picked up, it should be for
the purpose of adding value. Hence, the ratio of number of touches to number of
value-adding operations is an indication of the extent of extra material handling.
The ideal ratio is of course one.

A layout causing greater material handling will result in the following,

• Higher resource requirement—people, space, handling equipment,


• More opportunities for damages/quality problems.
70 5 Commencing the Lean Journey

This case example of a crockery manufacturer shows us the importance of


measuring and working on the layout.

Broken Cups and Saucers


We recently worked on a Lean assessment for a ceramic crockery manufac-
turing unit. The process is divided into two main sections—the green stage
and the whiteware (fired product). While rejections in the green stage are
reprocessed, any defect in whiteware leads to the item being either scrapped
or sold as seconds. The whiteware rejection was a whopping 20% with dam-
ages and chipping contributing significantly to it. The layout assessment in
the whiteware side of the operation threw up the following data:

Distance travelled by one cup or saucer = 220 ft


Inventory points per operation = 4
Number of touches was (15 for three value-adding operations) = 5
touches per operation.

The ‘Muda walk’ clearly brought out the fact that the layout had led to mul-
tiple storage points which in turn resulted in increased handling. Interlinking
the operations in flow sequence was the project accorded priority. Within
three months, the team modified the whiteware section layout and reduced
handling damages by 50%.

5.3.1.5 Machinery
About a decade ago, a plant’s performance was defined by capacity utilization or
operating efficiency. A plant operating at 80% efficiency simply meant that it was
running 80% of the time, while 80% capacity utilization meant that it produced
80% of rated output. Lean however focuses on the effective utilization of plant and
machinery rather than solely on efficiency. The difference in these two approaches
is captured succinctly in the following statement.

“Efficiency is doing things right while effectiveness is doing the right


things…. efficiently”

Overall Equipment Effectiveness (OEE)


OEE is the single comprehensive metric for the assessment of automated and con-
tinuous process plants such as chemical, pharmaceutical, metal, oil & gas, food
processing. In discrete manufacturing, OEE should be measured mainly for the
identified bottleneck operations, as a lower OEE for a non-bottleneck operation
does not imply reduced capacity (unless its ECT becomes higher than that of the
5.3 Cost-to-Serve 71

bottleneck operation due to reasons mentioned in Sect. 5.2.2.3). In line with the
Lean paradigm of value-adding ratio, OEE is simply the VA ratio for a plant or
machine. It is a measure of the proportion of total available time spent by the plant
or machine in actually transforming material as per the customer specification. The
three main components of OEE are availability, performance, and quality which
measure the following six major losses.
Availability is the time the machine actually runs net of downtime. The two
losses causing downtime are:

• Equipment failure,
• Set-up or changeover.

Performance is how much the machine delivers against its rated output and
depends on the speed losses such as:

• Idling and minor stoppages,


• Reduced speed.

Quality is the good output that actually meets customer specifications and excludes:

• Process defects,
• Yield losses.

Computation of OEE is illustrated in this example of an injection moulding


machine.

OEE Calculation Method for an Injection Moulding Machine


The machine runs 24 h a day and OEE is also computed on a daily basis.
Yesterday there was a breakdown for 90 min and a die changeover due to
die wear-out which took 150 min. We first compute the availability ratio.
Total available time = 24 × 60 = 1440 min
Actual run time = Available time - Breakdown time - Changeover time
= 1440 - (90 + 150) = 1200 min
Availability ratio= Actual run time / Total available time
= 1200 / 1440 = 83 %
Next we compute the performance ratio. The machine actually produced
1080 components and has a rated cycle time of 60 s per piece.
So, rated output in actual run time = Actual run time/Rated cycle time =
{1200 min × 60 s/min)/(60 s per piece) = 1200 pieces.
Performance ratio = Actual output/Rated output in actual run time =
1080/1200 = 90%
72 5 Commencing the Lean Journey

And finally the Quality ratio. A total of 30 pieces was rejected due to
defects.
Quality ratio = Good output/Total output = (1080-30)/1080 = 97%
OEE = Availability x Performance x Quality = 83% × 90% × 97% =
72%
So, effectively the machine spent only 72% of total available time in value
addition.

Let us delve a little deeper into the practicalities of computing the OEE.
Availability parameters: the availability parameters including breakdowns,
changeovers, and planned downtimes are generally captured in a similar man-
ner across most industries. Hence, the calculation of availability ratio is pretty
straightforward as we saw in the injection moulding example.
Performance parameters:

1. Rated Speed—Defining Rated Speed can be a little tricky. Typical Doubts that
Arise Are:
• Is it the speed as demonstrated by the supplier during performance guaran-
tee?
• Is it the speed stated in the machine manual?
• Or is it the best achieved practically demonstrated speed or cycle time?
• Speed is different for different products—how do we consider this during
machine performance ratio calculations?
While there is no one correct answer that fits all circumstances, the following case
example can point us to appropriate directions to take in different situations.

Aluminium Strip Casting


In continuous casting of aluminium alloys for sheet applications, the ideal
casting speed varies based on the alloy and the thickness of the cast strip.
There are no short stops in this operation as even a one second stop of
the casting machine will lead to metal leakage and complete breakdown
of the process for several hours. The performance ratio has to be therefore
calculated for each run based on casting speed alone. The rated speed is
based on the supplier’s performance guarantee as well as mentioned in the
machine manual.
For example, a 1XXX series alloy cast at 8 mm and 6 mm thickness
should be run at speeds of 1.2 m/min and 1.4 m/min, respectively, while
an 8XXX alloy cast at the same thicknesses has to be run at slower speeds
of 1.10 m/min and 1.25 m/min, respectively, due to increased alloy hardness.
5.3 Cost-to-Serve 73

In practice, parameters such as mould setting accuracy and molten bath tem-
perature also determine the actual running speed. This means speed varies
through the day with variation in process conditions.
Performance ratio is therefore computed as equal to actual output for the
day/output at rated speed.
Output at rated speed = (Rated Speed (m/min) × Strip width (m) ×
thickness (mm)/1000) × Density (kg/cu.m) × actual run time (minutes).
Actual output is a sum of produced good coil weights and rejected
material weight (is also weighed before remelting).

2. Short Stops

The second loss factor for performance is short stops. Rarely does anyone
record these one- or two-minute stoppages. If we do not have an auto-recording
mechanism, we can identify the contribution of short stops to performance loss by
working backwards from the performance ratio as we can see from the following
example of an edible oil packing machine.

Edible Oil Packing


A standard Form Fill and Seal (FFS) machine is used to pack one litre of
sunflower oil at a rated speed of 15 pouches per minute. The machine does
not have any provision for recording short stops, and one operator handles
three machines. We need to assess if there is a significant impact on per-
formance due to short stops. Following data was collected for a one day’s
production:

Running time: 6 h
Output as per rated speed = 15 pouches per minute × 360 min = 5400
pouches
Actual recorded output as per machine counter = 4800 packs
Performance ratio = Actual/Rated = 4800/5400 = 89%
Actual speed at the which machine was run (as seen from control panel)
= 14 pouches per minute
Performance that should have been obtained at this speed = 14/15 = 93%
or 7% speed loss.

Hence, 93 − 89% = 4% is the loss due to short stops. If this is significant


in a given context, as a next step, we may go for direct observation to record
number of stoppages and their reasons.
74 5 Commencing the Lean Journey

In today’s world, several plants have started recording short stops without
operator involvement. Industry 4.0-based Internet of Things (IoT) solutions can
automatically capture any machine stoppage with provisions for recording the
reason for stoppage as well.
Quality parameters: there are primarily two types of quality losses—rejections
and yield losses.

1. Rejections: every industry records rejections and most factory teams even drill
down the rejection data to perform a root cause analysis. But one should be
also able to identify in-process defects which are sometimes not clearly marked
as rejections but taken as part of the process losses. This example illustrates
how rejections are identified in a metal casting industry and incorporated into
the quality ratio.

Aluminium Strip Casting


In a casting operation, rejections have a huge cost impact. On a particular
day, it was observed that 60 MT of molten metal was supplied to the casting
machine. This should have produced 12 coils of 5 MT weight each.
No coil was rejected for quality reasons. So, the quality ratio should be
100%. However, on checking, it was found that the total net weight of the 12
coils produced was 59.2 MT. Further examination of the log book showed
that due to a surface defect, one coil was truncated and the defective layers
taken out separately as a mini coil and sent for remelting.
So, the quality ratio here is actually 59.2/60 = 98.6%.

Yield losses: material yield information is sometimes not readily available. In


chemical and process industries, the effectiveness of value addition is diminished
by yield losses. In cases not involving chemical reactions, yield is simply material
output/input. Where there is a reaction, it would be the actual output for given
input as against the theoretical output based on the chemical formula. Let us look
at a few examples on how to factor these losses into the quality ratio.

Prawn Feed Operations


In the prawn feed mill, the final product (feed) is obtained after sieving the
extruded, conditioned, and dried material. Here, the material falls through
a set of vibrating sieves of different mesh sizes and is separated into three
parts:

Right size—good material that is packed


Oversize—collected and reprocessed
Undersize fines—collected and disposed.
5.4 Assessment Criteria 75

Now as per the process design, all the raw material fed into the die can be
extruded to required size. So, while there may be no rejection of packed feed,
the sieves are preventing the undersize and oversize from getting packed.
On a day when 100 MT of material was fed to the die, 93.6 MT was
packed. So, quality ratio is 93.6% as this proportion of the raw material was
converted to product for the customer. Fines collected in bins weighed 1.4
MT, and the remaining 5 MT was reprocessed.

Oil Yield Loss


Going back to our earlier example of an edible oil packing machine, the
quality ratio is a result of rejections and yield losses. Let us say a packing
line comprising of a set of packing machines has packed 60,000 one-litre
pouch packs in a day. At a density of 0.9 g/cc, this should be 54 MT of
good output. At the end of each day, the holding tank levels are recorded.
Oil fed to packing line = Opening Level – Closing Level + Oil inflow
quantity.
We see that the oil fed to packing line was actually 60,500 L. The reasons
for this difference include:

Defective pouches—the oil is later reprocessed by cutting open the


pouches. Some oil remains adhered to the plastic film and cannot be
recovered.
Oil spillage/leakages in the line or machine.
Oil “giveaway” which is the excess oil filled into a pouch due to variations
in the pouch filling process. The customer is paying for 1 l of oil which
weighs about 0.91 kgs. However, there is a variation in pouch-to-pouch
oil content which is seen when the operator weighs a set of 5–10 pouches
every half an hour. So, the actual weight or oil filled could for example
be 0.915 kgs.

It is difficult to measure weight of each pouch and calculate the actual


“giveaway” as also the “dead loss” or irrecoverable losses. Hence, qual-
ity ratio is computed simply as Packed Output/Input Oil quantity =
60,000/60,500 = 99.17%.

5.4 Assessment Criteria

So far, we have looked at the various diagnostic tools that can be used to assess
the current state of any operation and measure the key metrics of time-to-serve and
cost-to-serve. Assessing current state and preparing a roadmap typically take about
76 5 Commencing the Lean Journey

Table 5.5 Appropriateness of different tools and metrics for different industries
Industry sector VSM MFD OEE Key productivity metrices
Discrete/engineering E E Ua Time, space
Assembly lines U L L People
Chemical/process L L E Material yield, equipment
Predominantly manual U E L People, time, space
Job shop/custom built L E Ua VA ratio
Services U E L Time, space, people
L—low relevance, not required
E—essential, has to be done
U—useful in specific places
Ua —useful for bottleneck equipment

three days of focused engagement. The key to a quick and effective diagnostic
study is to be clear about the metrics to be captured and the tools required to assess
the process. How do I decide what is or is not relevant? A fair number of lean
practitioners think that the Lean tools are generic and can be applied everywhere.
While this may be so, both the effectiveness of diagnostic tools and the relevance
of the metrics are dictated by factors such as the type of industry and the situation
prevailing at that point in time.
We have summarized the relevance of the diagnostic tools and metrics across
industry verticals based on our experience with well over a hundred organizations
across multiple industry sectors in Table 5.5. The three tools considered are the
Value Stream Map (VSM), Material Flow Diagram (MFD) and Overall Equipment
Effectiveness (OEE). The goal is to have a focused and value-adding diagnostic
exercise where we assess what is truly important and measure the impact param-
eters. This will in turn help us prepare a focused improvement roadmap that may
significantly cut down on lean implementation time and resource.

5.5 Target Setting

Having understood and assessed the current situation using appropriate metrics,
one can proceed to set targets for the organization. Lean targets need to be
ambitious yet achievable—only then will the organization be motivated to think
differently as per Lean paradigms. These targets are primarily of three types:

1. Growth oriented—how to produce more with the existing resources and


facilities,
2. Profitability oriented—how to minimize resource consumption for existing
output,
3. Employee well-being—improving the working life and morale of the employ-
ees.
5.5 Target Setting 77

5.5.1 Growth-Oriented Targets

The target should consider the projected demand over the next three years. A
gap between actual current output and this projected demand points us towards
making an in depth analysis on the capability of the value stream. Closing this
gap involves a progression of improvements using Lean tools. The first step is to
ensure actual delivered output matches the expected output under current condi-
tions as described in Sect. 5.2.2. The next step is to meet the installed or rated
capacity of the current value stream which is generally the capacity of the bot-
tleneck process, be it equipment driven or manual process. And finally, if the
projected customer demand (target) is beyond the rated capacity, we look at ways
and means to augment capacity. The typical roadmap for a growth-oriented Lean
program is depicted in Fig. 5.6 with the actions to be taken mentioned under each
"step" that help the organization move up to the next level of output.
Some of the principles that are applied in prioritizing improvement areas :

• Improve internal processes first; then work with vendors and sub-contractors,
• Look through the eyes of the customer—hence focus on downstream operations
first,
• Focus on the “pacemaker” operation—this is the last process in the value stream
after which there exists only continuous flow.

For SMEs looking primarily to enhance output, it is advisable to strike first at the
bottleneck operation wherever it may be located, even be it at a sub-contractor or
vendor process. After the bottleneck is resolved, one can aim to quickly stream-
line the flow across the value stream. Once the process is delivering at its rated
capacity or more but still falling short of the target, the management may look at
alternate strategies to close the remaining gap. These strategies vary from situation
to situation and also on the extent of shortfall with respect to the target demand.
Some typical strategies that SMEs can look to implement under different scenarios
are shown in Table 5.6.

5.5.2 Profitability-Oriented Targets

The popular cost reduction is re-interpreted in Lean terms as “expense reduction”.


All expenditure is not cost, only that portion that is actually incurred on value
addition is treated as cost. Since the primary goal of Lean is minimizing waste-
ful or non-value-adding activities, the consumption of resources that go into such
activities is reduced and the expenses also proportionately come down. Hence,
through Lean paradigms, the organization can enhance profitability even without
top line growth.
Earlier in this chapter, we have assessed the value-adding ratio for each major
resource, be it man, machine, material, or facility. Now, targets can be set on each
78

Gap? Target Output

Rated capacity

Augment Capacity
Potential output after
improvement
Long term measures
Expected output in
current state - Zero break downs
Improve – Attack - Defect free operations
Actual Output (current
Clouds
state)
- Increase availability
Process Stability
5

- Reduce rejections
- Minimize disruption
- Reduce cycle time
- Reduce short stops

Fig. 5.6 Growth-oriented Lean program


Commencing the Lean Journey
5.5 Target Setting 79

Table 5.6 Strategies to increase capacity


Gap with target output
Strategies to be explored <20% 20–50% >50%
Increase shifts/hours of work Y
Sub-contracting part/whole of the job Y Y
Cycle time reduction through adding people Y Y
Low-cost automations Y Y
Machine speed increase Y
Adding extra resources (people/facilities/machines) Y Y
Expanding the plant Y

Table 5.7 Examples of


Parameter UOM Current state Target
profitability improvement
targets for a crockery unit People Nos per person per 137 173
productivity day
Material yield % reprocess 8.5% <5%
% rejection 20% <10%
% rework 11% <5%
Facility Distance travelled 0.8 <0.5
utilization (km)
Employee Man movement 219 <100
strain (km)

or any of them, in terms of improvement in VA ratio, in other words, increase in


resource productivity.
A typical target setting done for a small-sized crockery manufacturing unit is
shown in Table 5.7. The processes in this industry are largely manual, and cur-
rent rejection levels are high. Hence, the focus is on optimizing the man and
material resources. Also, note the human well-being factor of employee strain
reduction has been included as part of the roadmap.

5.5.3 Employee Well-Being

To sustain any Lean initiative, involvement of all the people working in the orga-
nization is a must. A good to way to get the buy-in of people especially those
actually working on the floor is to include improvement projects that focus on
strain reduction, safety, working conditions and environment. In fact, many of
these are the outcome of Muda activities and so it is often possible to meet two
objectives with the same improvement project.
Take the example of the crockery roadmap in Table 5.7. The current state assess-
ment showed that collectively people were moving 219 km within the factory floor
80 5 Commencing the Lean Journey

Table 5.8 Comprehensive Lean target table for an LED manufacturing unit
Section Paramter UOM Current Target
Materials management Space utilization % utilized 75% >90%
(CFT)
Customer service % OTIF kit issue <60% 100%
level
Stores Visual No Search free
management
WIP inventory No. of days 35 days <7 days
levels (LED)
Final assembly Productivity Nos/person/day 309 368
Material transport Distance moved 392 ft <200 FT
Bulb handling No. of touches 26 <20
Rejection %/PPM 2% <0.5%
MI line and packing Wave soldering Ratio 48% >75%
performance
MI line Nos/person/day 333 417
productivity
PCB handling No. of touches 20 <8
Rejection %/PPM 1262 <500
People cartons/man-hour 4 >10
productivity
Total factory People Value adding % 135 160
productivity
Space utilization VA % 41% >60%
Facility itilization Distance travelled 680 <400 ft
Customer service % OTIF
(delivery)
Quality defectives % defects >5% <2%

each day. Bulk of this movement also involved handling and transporting material
between processes. More handling of crockery increases opportunities for dam-
ages such as chipping and breakages. Hence by taking up a project to reducing
handling and movement, reducing human strain and reducing rejections can both
be achieved, and the workers are enthusiastic to work on more such projects.
Finally, to summarize, a comprehensive target setting for an LED light manu-
facturer that incorporates all the three types of goals described in this section is
shown in Table 5.8.

5.6 Summary

This chapter is all about performance measurement. What to measure, the rel-
evance of different measures in different business contexts, how some of the
5.6 Summary 81

practical difficulties in measurement can be overcome, etc. are discussed. The mea-
sures belong to primarily two categories: time-to-serve and cost-to-serve. All the
measures in Lean paradigm are clearly defined from customer value-addition per-
spective. Meaning, whether or not the time and cost is being incurred in adding
value to the product from a customer view-point or not. All the time and cost
components that do not add value become candidates for elimination, which is the
topic of the later chapters. Different organizations adopt Lean for different rea-
sons, such as achieve sales growth or increase profitability, and improve working
conditions for the operators. This chapter gives an indication of what measures are
suitable for different strategies.
Designing the Lean Intervention
6

6.1 Introduction

We have seen in the previous chapter that current state assessment starts with
understanding of management goals, defining corresponding performance metrics,
and capturing the current state of the process with respect to the chosen metrics.
Post assessment, all these need to be consolidated into a coherent roadmap that
defines a stage-wise path towards achieving the stated goals under Lean paradigms.
Suffice to say, the roadmap should be simple, reasonable, unambiguous, and
easy to monitor. Documentation should not become a barrier to action. Lean is
all about doing and the quicker we start implementing the required improvements,
the sooner we begin to see the fruits of undertaking this journey. This chapter
discusses roadmap creation, the pitfalls, and how to avoid the same.

6.2 Building a Roadmap

Figure 6.1 indicates the broad stages of framing a Lean roadmap.


The gap between desired performance levels and the current performance levels
needs to be bridged. The roadmap details the steps to be taken in sequence with
approximate timelines in order to achieve the desired performance levels. Each
step may comprise a set of improvement projects that can be executed using spe-
cific Lean concepts, tools, and techniques. What differentiates a Lean roadmap
is the underlying Lean philosophy that drives the prioritization, sequencing, and
focus on standardization of the gains from individual improvement projects. The
following are the key components of a Lean roadmap:

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 83
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_6
84 6 Designing the Lean Intervention

Management Goals

Current State Diagnosis Gaps Process Level Targets

Focus Areas
What to improve
Expected Outcome
Lean Roadmap – Improve,
Standardize, and Sustain

Project Sequence
People Involvement
Execute, Monitor and Manage

Fig. 6.1 Creating a Lean roadmap

How much do we target in terms of performance


What to improve (themes)
Where to focus our attention (processes)
Why do we need to improve this (outcomes expected)
When priorities and timelines (schedule)
Who will work on these projects (teams).

In the previous chapter, we spoke about target setting in Sect. 5.5. The bulk of
the chapter was devoted to the assessment process that enables us to understand the
“what”, “where”, and “why” components. We now crystallize this understanding
through a documented roadmap and make it actionable by finalizing the “when”
and the “who”. We delve deeper into these two aspects here.

6.2.1 Sequencing the Improvement Projects

In an earlier chapter, we saw that SMEs in general have a shorter leverage as their
business horizon dictates focusing on immediate concerns. It is essential that we
keep this in mind while finalizing our Lean roadmap. Project priorities need to be
carefully fixed for both the process and result criteria. The generic methodology
discussed in Sect. 4.2 (Table 4.1) lays out a recommended sequence of activities to
implement Lean. In the first step of implementation (Create Flow), the methodol-
ogy emphasizes on “quick wins” or projects that show immediate gains to motivate
the SME owners to continue on the Lean path. Some typical types of projects that
meet these twin requirements are as follows:

• Solving an immediate pressing problem,


• Increasing output to cater to business growth/market demand,
6.3 Developing the Roadmap 85

• Improve productivity and lower costs.

Understandably, most roadmaps for SMEs are designed to start with implementing
flow, line balancing, and cycle time reduction at bottleneck operations, because this
is where immediate gains can be made with practically no additional investment.
We have discussed several case examples of such beginnings to Lean implementa-
tion in Chap. 4. In Sect. 6.3 below, we will describe how additional improvement
projects can be mapped to the later phases of Lean implementation.

6.2.2 Motivating the Team

People are the key to success in any intervention and more so, in a grass-root-level
intervention such as Lean. Aligning all the stakeholders to a common vision is
known to be the starting point for ensuring success of any change-related initiative.
Most often, SMEs may not have a clearly articulated vision that people can align
to. The Lean intervention should be designed such that it excites and motivates
people in the organization to take up, implement, and sustain Lean. Designing a
right mix of projects at each stage of the roadmap to benefit all the stakeholders
will have a big impact on implementing and sustaining Lean.

6.3 Developing the Roadmap

In Chap. 2, we introduced the core Lean concepts of value and the three types
of waste popularly known as the 3 Ms—Muda, Muri, and Mura. As a rule of
thumb, a roadmap with projects that address all the 3Ms will find buy-in from
all stakeholders, given specific benefits to the individual stakeholders as shown in
Table 6.1.
Given the lack of awareness about Lean management and its benefits amongst
most SMEs, it is essential to choose a mix of projects in the initial stages that can
generate interest among all the stakeholders. These snapshots from a mini-cluster
of bulk drug manufacturers highlight the impact of having the right mix of projects
on the final outcomes of a Lean intervention.

Bulk Drug Cocktail-1


About a decade ago, three bulk drug units took the bold step of attempting
Lean when it was almost unheard of in the industry. This meant that it was
essential to ensure people on the shopfloor buy into this concept early on.
At Porus Drugs, a six-month roadmap was framed with focused interven-
tions every six-weeks. The set of projects to be taken up in the very first
intervention included the following:
86 6 Designing the Lean Intervention

Table 6.1 Impact of reducing 3 M’s


By reducing Who benefits How
MUDA (waste) Entrepreneur customer • Improves business results—top
line and bottom line
• Lesser consumption of scarce
resources
• Quicker service/delivery
MURI (physical strain) Workers • Improvement in working
conditions and work methods
help reduce the strain of daily
working
• Able to focus and provide
better quality
MURA (inconsistency in Managers/engineers/supervisors • Streamlined work flow reduces
process) stress involved in managing
operations
• Better control on process =
more clarity on service
delivery and capacity
management

• Reducing variation in the bottleneck operation (reaction) cycle time with


the goal of consistently maintaining the lowest achieved cycle time: This
would directly increase throughput from the unit. This was to be achieved
by optimizing process parameters, and a team of qualified production and
quality personnel was formed to tackle this project.
• Reducing cycle time in drying process: The project was originally envis-
aged as a Muri project to reduce strain during loading and unloading of
tray driers. This emphasis on reducing physical strain found favour with
the workmen there as they were struggling with the manual loading and
unloading. Also, the delay in completing the drying was stressing out the
laboratory technicians who were forced to test multiple rounds of samples
to confirm whether the required moisture content has been reached. The
end result of this project would be reduced cycle time and higher output
while reducing stress and strain to staff and workers.
• Rearrangement of stores on 2S principles focusing on spare parts and
consumables: This project brought in the involvement of stores and main-
tenance personnel, thereby ensuring people across various functions are
exposed to Lean concepts early on.

Hence, through a judicious mix of projects we were able to provide the


right challenge to the qualified professionals, allay the anxieties of techni-
cians and the strain of workmen, and bring in people across functions to
appreciate Lean. By the end of the journey, the unit had recorded a 20%
6.3 Developing the Roadmap 87

increase in output, lead time reduction from 10 days to 4 days, and the
drying process had actually been eliminated!!

Bulk Drugs Cocktail-2


The second of these enterprising units was KRR Drugs. The primary goal
of Lean implementation here was to identify and minimize waste at the bot-
tleneck processes of reaction and distillation. But the roadmap was drawn
up consciously to work on improvements at the milling area—the last stage,
where dried granules are milled to a fine powder and packed for dispatch to
the customer plant. This process was actually one of the lowest cycle time
operations in the value stream. But the operation generated huge amounts of
dust making working conditions extremely difficult for the operators.
The strain on the two workmen was unbearable. In spite of wearing all
the Personal Protective Equipment (PPE) including boots, overalls, masks,
Googles, and helmets, they were covered in fine product dust from head to
toe. Visibility in the room was so low that even the camera lens was fogged
within seconds of attempting to take a video of the current state. A catchy
and bold target was set for the team working on this project-operators should
be able to work in the room without having to wear a mask. In other words,
dust levels in the atmosphere should be zero.
Sources of dust escaping from the process were identified, solutions dis-
cussed and implemented within three days, with direct inputs from the
operators. On the last day, a complete video of the process could be shot
with the workmen operating the process without masks. They were literally
in tears as the project concluded, and the senior operator gave an emotional
speech in front of the entire staff during the Lean project sharing session.
Needless to say, the entire workforce strongly endorsed the Lean initiative
and was eager to join and implement subsequent projects.

The importance of getting the buy-in of the workers who are the actual
value-adders in any manufacturing organization cannot be overemphasized. For
workmen, company profits, turnover, or productivity is of marginal interest unless
there are incentives. Even incentives most often reward local or individual produc-
tivity and not the throughput from the entire value stream. Since worker acceptance
of Lean is crucial to its sustenance, the roadmap should prioritize projects that can
directly impact their day-to-day working life. Muri or strain reduction is the best
bet as the previous drug manufacturing cases and following example from a white
goods manufacturer shows us.
88 6 Designing the Lean Intervention

Rockwell—90° Makes a Sea of Difference


At this commercial refrigeration unit, makers of deep freezers, and chest
coolers, we completed the diagnosis including a VSM and material flow
diagram. Management goal was to rapidly ramp up output by 50% to meet
market demand. The bottleneck processes were identified from the following
overall manufacturing process flow depicted below (Fig. 6.2).

Fabrication of Body PUF Final Assembly &


sheet metal Assembly insulation Testing

Fig. 6.2 Rockwell manufacturing process flow

Process observations confirmed that the body assembly section was the
bottleneck as sufficient bodies were often not available for the customer PUF
process. The PUF jigs were found to be empty intermittently during pro-
cess observations. So, our roadmap prioritized improving the body assembly
operations which also happened to employ the maximum number of workers
in the factory.
To reduce cycle time, the workstations needed to be balanced and flow
established. However, we had earlier observed that the initial clinching oper-
ation was being done in an unconventional manner. The smallest built worker
was assigned to the two-man team doing this operation. His job was to
crouch inside the space made by the two bent sheet metal parts (named Big
C and Small C after their appearance) and use the clinching machine to join
them together to form the outer shell of the cooler. Once these were joined,
the clinching operator was trapped inside. The second operator would then
lift the entire shell over the head of the clinching operator and place it on
the next station, thus freeing the operator to move about (Fig. 6.3).
Small C

Big C

Earlier Process Improved Process


(Top View) (Front View)

Fig. 6.3 Rockwell: improved clinching operation


6.3 Developing the Roadmap 89

Though this was not the highest cycle time operation, we still added this
as a first priority project in the roadmap. The project goal was to avoid
the operator from having to clinch by sitting inside the shell in an uncomfort-
able position. Within days, a better method was envisaged and implemented
using suggestions by the operators. The existing clinching method was
replaced by clinching on a table with the big C being supported in a aligned
vertical position using magnets and a fixture. The operators stood facing each
other on opposite sides of the table and clinched by running the machine
comfortably along the table platform at waist height. This new method was
so operator friendly that the entire assembly workforce were delighted and
wholeheartedly participated in the assembly cycle time reduction projects
at other workstations. Three months after Lean began, the body assembly
section provided 200 bodies per day to the PUF process against the previous
peak of 120 bodies per day.

One important aspect of diagnosis is getting a feel for the mood of the employ-
ees. For an external Lean Sensei to be accepted by the grass-roots operators, this
is all the more critical. Quite often we find employees comfortable with status quo
and unwilling to try and experiment or pursue a different way of thinking. There-
fore, the roadmap should have a couple of initial breakthrough projects, which can
cut through the natural inertia and reluctance of such employees and get them to try
something new. One of the easy ways to do this is to break a few walls…literally,
as we can see from the following case let!

Breaking Walls… and Mindsets


RDN was looking to balance its working capital burden while meeting
increasing customer demand for its fabricated solar application structures.
Over the past decade, the factory had expanded laterally from its single fac-
tory shed premises by renting adjacent properties and adding machines and
facilities in them.
Current state diagnosis showed a lot of criss-cross material movement in
between various fabrication and welding processes located across the three
adjoining galas (narrow independent constructed sheds). The Lean roadmap
for a first cycle of implementation spanning about nine months included
projects on layout, reducing bottleneck process cycle times, defect-free weld-
ing, pull systems for outsourced galvanizing process, and FG warehouse
arrangements on 2S principles. During the diagnostic study and framing of
roadmap, we observed that most of the team of young engineers and older
supervisors seemed equally reluctant to accept the need to make improve-
ments. Realizing the potential barriers to the Lean journey, we gave the first
priority to layout changes in our roadmap.
90 6 Designing the Lean Intervention

The proposed flow layout required a section of the wall between two
adjoining sheds to be demolished to help drastically slash material handling.
The management was historically not known for any radical actions, and the
team was also comfortable with this inertia. But we were able to convince
the management to give permission and the wall came down overnight. This
single action helped break through the barriers in the minds of the employees
as they realized the seriousness with which their management viewed Lean.
The psychological effect of a wall coming down is well known; the Berlin
wall coming down in 1991 to unify long separated people being a prime
example. In this case too, resistance to change crumbled just like the wall,
and Lean implementation was well and truly underway.

6.3.1 Theme-Based Roadmaps

It is always impressive when a roadmap depicts a clear path to achieve stated


measurable business goals. Increasing output, productivity, profit or reducing rejec-
tions, or wasteful expenditure all sound good and look good on paper to the top
management. However, such roadmaps appear to be “dull” and “unexciting” to the
majority of the employees in an organization and even more so if it happens to be
an SME. Too much emphasis on business performance results may also result in
a negative impact; employees may start questioning “What’s in it for me?”
In our experience, an effective roadmap that resonates with operators’ concerns
is to have themes that can excite, motivate, and align them to the Lean initiative.
The end goals are already clear and defined in terms of metrices, and we also
have identified projects/focus areas, the execution of which will help achieve these
goals. What remains is to present these projects in a theme-based view that help the
employees see the benefits of improvement with respect to the immediate issues
concerning them. Themes can be as simple and straightforward as zero defect or
zero breakdown or as varied as “dust-free factory” or “strain-free operations”.
Through these examples, let us understand how such themes actually help
in developing Lean thinking.
“No Material on Floor”—in simple language, material is money in a manufac-
turing set up, as the customer pays for finished material while the organization has
paid for the input materials. When we develop a theme like “No material on floor”,
we are asking people to give due respect to the material that provides them their
livelihood. This is the emotional connect. Now from a rational Lean perspective,
operations carried on at waist height are strain free and most productive. Most
machines have a waist height loading and unloading point. To pick up material
from or place it on the floor adds to worker strain and increases operation cycle
time. This strain may in turn lead to defects/damages or hygiene issues all of which
are Muda. Hence, no “material on floor” can positively impact all the 3 Ms and
therefore motivate all the stakeholders.
6.4 Preparing for Implementation 91

“One touch”—hygiene is a basic need in segments like food processing, phar-


maceutical and consumer products and every human being can relate to it. In Lean,
we say “One touch to value-add” because any additional touches will only result in
extra handling or transport which is a waste. From the hygiene perspective, every
touch increases chances of contamination or defects and people understand this.
By aligning shop floor people to this theme, every additional touch can be iden-
tified and solutions implemented to eliminate them, thereby reducing the overall
waste (Muda) in the process.
A single roadmap can also have multiple themes, each one aligned to a different
leg of the journey as in this case of an edible oil manufacturer who was looking
for a Lean-based continual improvement (CI) program over a one-year time frame.
This industry has two major processes—refining and packing. Refining is a con-
tinuous process with multiple stages such as deodorization and filtration. Packing
on the other hand is a mass manufacturing operation with high-speed unit pack
machines followed by manual methods of secondary packaging. We needed to
have a common theme that could resonate with all parts of the plant and be able
to incorporate all relevant Lean tools so that people could be trained on using
them. The core 3 M concepts were ideal as they were generic and all-purpose
. Business goals of reducing losses, material wastage, and increasing throughput
were translated into process-level targets (see Fig. 6.1) which are seen in Table
6.2, and each target is linked to a set of projects under one of the three themes.

6.4 Preparing for Implementation

Having completed the framing of the roadmap, the next step is to actually imple-
ment Lean through the projects identified in the roadmap. Up to this stage, Lean
paradigms have been introduced to the key people in the organization, and they
have been trained to relook at existing processes under these paradigms. At the
same time, they have not been asked or told to actually change anything. But the
seeds of Lean thinking if planted properly at this stage do help when it actually
comes to implementing Lean.
Lean implementation is all about “doing” process improvement which means
re-examining existing processes under Lean paradigms and making changes, pri-
marily in the methods or ways of working. In older factories, this could mean
enabling people who have been working in a particular way for the past several
years to now do the same work differently. Opportunity as they say often knocks
only once, and the same may be true of Lean as well. There are several instances
of organizations who have tried and failed in implementing Lean and have now
become wary of repeating the exercise.
Hence as we prepare to kick off the implementation phase to create flow,
we need to define an approach which can generate the required positive energy
and momentum to pull the organization through this journey. We discuss briefly
in these subsections the key aspects for preparing a successful approach to
implementation.
92 6 Designing the Lean Intervention

Table 6.2 Theme-based roadmap for an edible oil manufacturer


S. No. Theme Objectives Goals No. Continuous
improvement
project
1 Reduction in Mura Minimizing (A) To reduce i. To reduce weight
(inconsistencies in process the excess variation to
process) inconsistency in weight within 4 gm
hydrogenated oil giveaway in range as per
plant pouches machine
capability
(B) To reduce ii. To eliminate
the pouch film length variation
loss from 2% and sensor
to <0.5% related issues
iii. To reduce pouch
loss during start
up
Eliminating (A) To improve i. Reduce
process refinery deodorisation
inconsistency in throughput steaming time
refinery plant variations and get
100% batches
within spec of
2.5 h
ii. Study and
minimize
variation in
hydrogenation
time and catalyst
use
2 Reduction in Eliminating (A) To achieve i. Reduce heat
Muri—strain strain reduction the temperature losses in
in bakery fats in blending tank blending area
plant area below
45 °C
(B) To reduce ii. To minimize
the strain during effort and time
changeovers required for
changeover of
micron filter
iii. To minimize
effort and time
required for
changeover of
strainers
(continued)
6.4 Preparing for Implementation 93

Table 6.2 (continued)


S. No. Theme Objectives Goals No. Continuous
improvement
project
3 Reduction in Reducing (A) To reduce i. Online pouch
Muda—wastage residual oil residual oil cutting at
present in scrap going to scrap packing area and
from handling of oil
10 g/pouch to ii. Redesign
<2 g/pouch recovery tank
using steam/hot
air to recover
balance oil stuck
inside pouch
Eliminating (A) To reduce i. Observe and
carton and carton damages correct transfer
product wastage during transfer issues at
in new 2000 conveyor, spiral
warehouse boxes/month to chute
500 ii. Observe and
boxes/month improve manual
handling
conveyor to
storing point and
storing point to
truck loading

6.4.1 Focused Improvement Workshops

Quick results matter to the SME organization, and hence, the pace of implemen-
tation needs to be fast at the start of the Lean journey. Anything with a long
gestation period and likely to engage the time and effort of the SME’s resources
is not going to find favour. Hence, the methodology we choose to implement the
roadmap should be tuned accordingly.
One universal fact is that people love celebrating a festival or an occasion be it
a marriage or a birthday. We see this all the more in traditional communities that
have been in existence for hundreds of years. On such occasions, people come
together with families and friends and work towards a common purpose within
a well-defined date and time. The comradery between the people many of whom
may have not met each other for a while drives the work forward. In fact, this
“work” of preparation and organization of the event is so often as much fun as the
actual event itself.
Generating this same spirit in the approach for implementing Lean is the key to
success. The focused improvement workshop does just this. It creates an occasion
for people to come together and work towards implementing a successful project.
Our experience of using this methodology at SMEs over the last two decades
94 6 Designing the Lean Intervention

Do It
Again 9.Celebrate
1. Observe the
Process
8. Make this
the Standard
2. Identify
Waste, Variations,
Strain

7. Measure
Results

3. Plan
Countermeasures

4. Reality 5. Make Changes 6. Verify Change


Check

Fig. 6.4 Focused improvement workshop

has served to reiterate its effectiveness in quick improvements leading to lasting


results. The typical cycle for a workshop is shown in Fig. 6.4.
Each workshop is normally built around a theme such as “single-piece flow” or
“SMED” or “problem solving” and should incorporate the following elements.

6.4.1.1 Cross-Functional Teams


Figure 6.5 illustrates the typical composition of a cross-functional team.
Typically, team leaders are the process owners at the project location as they
are responsible for the day-to-day functioning of the concerned process. Sup-
plier team members get to understand the impact of their work on the customer

Process to
improve
CFT

SUPPLIER TO PROCESS CUSTOMER TO


THE PROCESS Neutral
OWNER THE PROCESS
members

NEUTRAL
MEMBER

Fig. 6.5 Cross-functional team


6.4 Preparing for Implementation 95

(project) process, while customer team members appreciate the difficulties faced
by their supplier process in serving them. The neutral member is akin to an umpire
or referee. This team member has no direct link to the process being taken up
for improvement and hence lacks any bias. She or he is expected to question
everything, thereby opening up new avenues of thought to the rest of the team.

6.4.1.2 Short Duration Event


A workshop ideally should last from three to five days at the most; anything longer
is bound to create fatigue and disruption in the regular work.

6.4.1.3 Focus
During such workshops, team members are expected to keep off routine work and
devote their time and attention fully to the project being taken up. An apt analogy
is the focus of sun’s rays on a piece of paper. Normally, a piece of paper exposed
to the sun does not hange in any manner. But if the rays are focused through a lens
to a single point on the paper, it soon catches fire. With focus, the team is therefore
expected to achieve in days what would normally take weeks and months.

6.4.1.4 Management Commitment


Full-time participation of the SME owners and partners in the workshops demon-
strates their belief and commitment to Lean and sends a clear message to their
employees. While they may not necessarily be a part of any project team, they
are expected to listen to the team’s progress at the end of each day, give required
resource support, and take decisions that will help the team execute their project.
Finally, they play a big role in motivating their teams.

6.4.1.5 Facilitator
Running a workshop and delivering significant results within days is not an easy
job, and the SME would do well to have the right Lean Sensei do this job. The
Sensei trains the teams on relevant Lean concepts, tools, and techniques, keeps
the projects moving as per plan, moderates issues that may arise, and acts as a
bridge to the SME management. Like any event, the facilitator has a major part in
ensuring success.

6.4.1.6 Clear Action Plan


Not everything can be executed in three days; there are always ideas that need
external intervention such as purchasing some parts or fabricating something out-
side and hence need more time to complete. All these are crystallized into a
clear-cut time bound and responsibility defined action plan to be monitored by
the management. Such action plans should not spill over beyond the month to
take advantage of the momentum created by the workshop event and to maintain
continued focus on the project.
96 6 Designing the Lean Intervention

6.4.1.7 Celebrate Success


Like all special occasions, the workshop should also culminate in a celebration.
The outcomes of the ups and downs, debates and arguments, late nights and physi-
cal effort that has gone into the improvement activity will need to be validated and
achievement toasted by the entire team. The workshop should always conclude
with the team on a high! The results of a well-run focused improvement workshop
can be phenomenal as this caselet shows.

Pump up the Volume


Our client, a medium-sized pump manufacturer was looking to increase
output from their existing facility. While the downstream pump assem-
bly process had a potential to produce 50% more, it was essential for the
upstream operations to deliver equivalent enhanced quantities of all the
required parts. The machine shop was organized in a typical functional lay-
out with CNC milling machines, presses, turning centres, drilling machines,
heat treatment furnaces, etc., all grouped in their respective locations.
Lean flow required the layout to be changed to part specific cells, and a
focused improvement workshop was planned to complete this project. In a
five-day workshop, the entire layout of the machine shop spread over two
large sheds was realigned. On the first day, the existing product–process
matrix was analysed to form new cells on paper. These cells were laid out
on an AutoCAD drawing with three different layout alternatives generated.
By the end of day 2, the team had brainstormed and finalized the new layout,
and in the closing session, it was approved by the top management.
Days 3 and 4 were devoted to shifting over a hundred machines, at times
even across the two sheds, using two hired cranes. Thousands of parts in
work in progress inventories were shifted manually to the new cell loca-
tions; each and every team member right from the production manager to
the operator were personally involved in shifting this material with their own
hands. On the last day, the newly formed cells were run and the processes
fine-tuned. The second half of the day was devoted to validating the output
over half a shift. The result—over 50% increase in part output in the shortest
throughput time!! The workshop concluded with an evening celebration as
team members basked in the success.

6.4.2 Post-workshop Reviews and Handholding

Once the euphoria of the workshop dies down and employees get back to routine,
it is difficult to keep their focus on completing the pending action points and taking
the project to its final conclusion. The role of the Lean champion (see Sect. 6.5) is
key to ensuring action plan status is tracked and updated, and the top management
kept abreast of the progress. The top management in turn is expected to extend
their support to the teams and enable them to complete the actions. This can be
best done through the following review mechanisms.
6.4 Preparing for Implementation 97

6.4.2.1 Daily Gemba Walk


The SME owner/factory manager should necessarily spend about half an hour each
day walking through the Gemba (Japanese word for shop floor). During this round,
they should pay specific attention to the processes under improvement and check-
up whether the improvements are being sustained. Team leaders can also showcase
new actions and improvements they have implemented post-workshop.

6.4.2.2 Weekly Reviews


A fixed day and time scheduled for weekly review of Lean implementation gives
the team a clear purpose to orient their week’s activities. All the workshop teams
are expected to attend review meetings together to ensure continuity in team work-
ing. Team leaders are encouraged to present their own action plan status supported
by visual evidence (photo or video of changes done). The review meeting is chaired
by the SME management as they continue to support the teams through decisions,
resources, and motivation.
The importance of the review mechanism cannot be overstated. It is quite
normal to see almost zero progress on action points in the first review post the
workshop. At this juncture, the continuity of Lean implementation is threatened,
and management plays a key role in bringing the momentum back. Completion
of 80% of the action points by the end of the month or timeline agreed upon is
considered as good progress. This is the point at which the planning for the next
workshop begins.

An organization making components for the defence sector started imple-


menting Lean at their factory as it shifted from a pilot to full production
stage. The years spent on doing research and development for the new
component had oriented the employees to a laboratory style of working.
Full-scale manufacturing that too employing Lean was a huge paradigm shift.
After the first implementation workshop, progress on action points was slow
and tasks that could be completed in days were being extended to weeks.
After observing this for a month, the factory head fixed a Saturday
evening deadline for each team to update their progress on the action points
and submit the same with photographic evidence. They were also to submit
data on the performance of processes already improved. This report was also
initially marked to CEO and the Sensei guiding the Lean implementation.
Teams reporting minimal progress for two consecutive weeks were called
for a separate review with the factory head to speed up their work.
By the second month, the pace started picking up and the factory head
was in a position to plan for the second focused improvement workshop.

All in all, it typically takes SMEs about six to nine months to complete the first
cycle of improvements. This entails creating and enabling basic flow (see Chap. 7
for details) and is normally achieved through three or four focused improvement
workshops with about six weeks gap between each of them.
98 6 Designing the Lean Intervention

6.5 Organization Structure for Lean

Sustainability of a Lean initiative depends on the people driving it, and it is impor-
tant to identify champions who would take on the mantle. Most large organizations
have an internal operational/manufacturing excellence team with people trained in
Lean, Six Sigma, and other such excellence philosophies. They are expected to
drive the Lean initiative in all the line functions. SMEs lack the resources to have
such people on their rolls and most often depend on working with an external Lean
consultant or Sensei.
It is important to remember that such external resources are temporary and will
never have the ownership for long-term sustenance in the way an employee would.
The best way is to identify at least two internal resources from line functions like
production, quality and maintenance, materials, etc., who can work closely with
the Lean Sensei and learn the entire gamut of concepts and tools. This core team
of Lean champions can then take over from the Sensei over a period of time and
ensure the sustenance.
Having such Lean champions is a conundrum for most SMEs as a valuable
human resource has to be spared for this new initiative. While the champion drives
Lean and ensures its success, he or she gets to learn a lot and grow as a profes-
sional. This knowledge is again a double-edged sword for the SME as such a
person while proving valuable to the organization will also get many more oppor-
tunities outside and is likely to leave the organization for better prospects sooner
than later. More on this is in our chapter on Lean sustenance.
At the roadmap stage, we may not formally define any roles for the identified
team members, but we should at least have them on board so that they are with the
Lean Sensei from the outset. Individual roles and responsibilities can be defined
at a later stage.

6.6 Summary

This chapter presents practical tips in arriving at process-level measures, given


organizations business objectives, identify the gaps, and the bridging improvement
projects. Once this is done, the next important part is sequencing the projects. Best
beginnings are those projects that provide quick wins and also improve operators’
day-to-day work. The later sequencing is best done on a theme-based mode, with
which every organizational level can get associated with. The chapter presents a
few examples of such themes, and the readers may come up with any other such
themes that make sense in their organizational setting.
Once the roadmap is ready, the next step is to implement it. The success or fail-
ure of Lean depends a lot on the approach to implementation. The authors’ decades
of experience in implementing Lean have helped refine the focused improvement
workshop methodology for rapid improvements that ideally suit SME organiza-
tions. The elements that go into such workshops and the post-workshop review
mechanism are discussed in detail. Finally, the importance of having a Lean
champion to drive Lean in the long term is brought out.
Implementing Lean
7

7.1 Introduction

“A plan is only as good as its execution”. This management adage aptly sums
up this module on implementing Lean. In the previous chapters, we discussed the
diagnosis of the current state of operations from a Lean perspective and approaches
to developing an actionable and sustainable roadmap. It is now time to act on the
roadmap. Implementing a Lean roadmap typically follows these three main phases:

• Process Improvement by applying Lean concepts, techniques and tools,


• Standardizing the improved processes and monitoring them,
• Sustaining Lean through repeated cycles of improvement and standardization.

The three chapters of implementation module, starting with this one, deal with each
of these three stages. Of these, the improvement phase has been the most widely
written about and discussed both in literature and in practice. Most well-known
Lean tools have been developed for making process improvements. The initial
improvement phase initiates “Change” and sets the ground for the implementation
of Lean. A profound Zen saying that forms the basis for Lean—goes:

If you do not understand something, it does not change anything


If you understand something, it does not change anything

We have understood the current state through diagnosis, identified what to improve
and laid it out in the roadmap. But nothing has changed as yet. Change at the
gemba (production floor) only occurs by taking appropriate action and this chapter
focuses on how to go about it.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 99
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_7
100 7 Implementing Lean

7.2 Begin with Flow

The universal natural principle of Flow forms the core of Lean. In our experience,
improving Flow of goods/services through the value stream is the best place to
start, because it directly translates to higher revenue and profits for the SME. A
good part of the improvement phase thus focuses on initially creating flow and
then enabling and supporting this flow on a continuous basis. But what should
Flow? It is the value we deliver to our end customer and for which the customer
is paying us. In a manufacturing organization, the customer pays for a specified
product. Hence, the raw material we procure needs to flow through the processes
adding value at each step all the way till it becomes a finished product and gets
delivered to the customer. Similarly, the service or information needs to flow in a
service industry, where often, the customer or guest is herself a part of the process.
For example, in service industries like health care, education, or hospitality the
guest should themselves flow through the delivery process.
Taiichi Ohno, placed a lot of emphasis on flow when he stated his view
on Toyota’s Production System:

All we are doing is looking at the timeline from the moment the customer gives us an order
to the point when we collect the cash. And we are reducing that timeline by reducing the
non-value-added wastes

So, by reducing the non-value-added activities, i.e. wastes, we are in turn speeding
up the flow of value-added activities and reducing the time for completing the
order to cash cycle. The objective is to “reduce the time-to-serve the customer”,
the means is “to increase flow” and the method to achieve this is “to reduce waste
or Muda”.
But where does this emphasis on flow come from in Lean thinking? As we
mentioned in Chap. 2, flow is the natural order of things. All great natural systems
are based on the principles of Flow. For example, the human body is designed with
multiple flows governed by systems such as thought (nervous), blood (circulatory),
breath (respiratory), food and water (digestive). Each system is designed to let the
body absorb what it needs and reject and eliminate from the system what it does
not need. We breathe in air (raw material), absorb the oxygen during processing,
and remove/exhale the carbon dioxide (waste) all in one seamless flow. It is the
same with food and water. The blood is the carrier of all these useful things and is
a self-contained insulated system transporting what is required, where needed, and
when needed in the quantity required (this forms the very definition of Toyota’s Just
in Time system).
Let us stop and ponder for a moment on what would happen if any one of these
flows is disrupted. If we eat too much and overload our digestive system, we are
left with an uncomfortable feeling and may even vomit or regurgitate our food—
retracted flow. Constriction of blood flow such as a block could lead to a severe
breakdown of the body in the form of a heart failure. A cold constricts the nasal
passage leading to shortness of breath and affecting our stamina. However, when
everything is running smoothly as designed, then we find ourselves in that perfect
state of physical and mental fitness and are able to do almost anything we set our
minds to achieve.
7.3 Understanding Basic Flow 101

Similarly, other natural systems such as rivers, trees, the sea, or planetary
motion all exhibit their own forms of flow. Even man-made music sounds best
when all the notes are in harmony and music flows. Replicating these concepts of
flow in the man-made process of manufacturing or service therefore forms the
core of Lean thinking.

7.3 Understanding Basic Flow

Understanding a typical river system is a very good starting point to design busi-
ness processes for flow. In the natural river systems of the past, water found its
own path as there was little or no man-made intervention. Over millions of years,
the river system developed into what we now see and know. Natural constrictions
that came up from time to time resulted in the river choosing an alternate path but
ensuring that it continued to flow towards and reach its ultimate customer—the
sea. At times, when there is more water than the system can process, such as dur-
ing very heavy rains, the river will reject this excess by the simple act of flooding
surrounding areas resulting in devastating losses of life and property.
Over time, humans started harnessing this flowing river water for various pur-
poses; drinking, year-long food cultivation and generating hydel power. To ensure
water is available when and where needed, they developed a series of check dams
and reservoirs at strategic points along the flow. The aim was to retain enough
water for our needs without hampering the basic flow and also control flooding.
When reservoirs are full, dam gates open and the excess water is released to flow
into the sea. Now, the end customers for a river include human beings and not just
the sea.
In the ideal scenario, the value stream of an organization should be akin to a nat-
ural river system, with material, information, or service flowing unhindered to the
customer. Value should flow seamlessly without blockage, constriction, retraction,
or zig zag movements. But in most shop floors, one comes across the “flooding”
of work in process inventories that occupy a majority of the available space. This
leads to significant and often incalculable losses for the organization in terms of
expenditure on handling, storage, damages, rejections, expiry (of shelf-life items),
and housekeeping activities. Hence, there is a clear case for moving towards the
ideal state of flow.
Typical process flows designed on Lean principles are like the current man-
modified river systems with continuous flow in pockets and strategically placed
“reservoirs” of inventory to balance and maintain optimum flow. The system also
incorporates checkpoints or quality gates that ensure only the right quality material
flows towards the customer process.
The strategic inventory points are designed to have standard WIP which is in
turn governed by pull systems operated through tools like Kanban. Here, the cus-
tomer process “pulls” what it needs, when needed, from the standard inventory
which in turn triggers the supplier process to produce and refill or replenish the
inventory consumed. Kanban, a Japanese word for signal, is the primary tool used
102 7 Implementing Lean

2 hours
WIP
Finished
Making Profile Goods
Supermarket

Fabrication Outer Body


Assembly,
Body PUF Testing &
Assembly
Packing
Fabrication Inner Tank

Pull based
Fabrication Tubes schedule

Pull based
replenishment

Fig. 7.1 Flow for a deep freezer manufacturing unit

to maintain this supplier–customer relationship. The origins for Kanban can be


traced to a typical supermarket system, where shelves are loaded with fixed quan-
tities of each product based on typical sales. As customers buy through the day,
the shelves are depleted of inventory and are replenished at specific times based on
certain rules. In Lean factories too, the standard WIP points are generally called
Supermarkets. More about pull-systems are covered in subsequent sections of this
chapter. At this stage, it is important for us to know that a well-designed Flow
actually combines flow and pull.
We see below in Fig. 7.1, basic flow for a deep freezer manufacturing unit.
Each section has been designed to have its own flows, and the sections are inter-
connected by small supermarkets of inventories to maintain overall flow from Raw
Material to finished good.
The Polyurethane Filling (PUF) process is depicted in a heart-shaped box
to indicate that it is the pacemaker of the whole value stream. Its production
rate determines the demand and therefore pull from upstream processes while
after the pacemaker the process is characterized by continuous product flow. The
state depicted here was achieved after two years of focused Lean implementation
and resulted in the reduction of lead time from 10 days to 2 days. We present a
structured approach to improve the flow in the following sections.

7.4 Key Consideration for Flow

The single most important aspect that determines the flow of material or opera-
tions in a facility is the Layout. Maintaining a close relationship between internal
supplier and customer processes is essential for smooth flow. If the supplier and
7.4 Key Consideration for Flow 103

customer are not located within touching distance, it becomes difficult to maintain
flow; handling and transport (Muda) of material between the processes becomes
the norm and inventories start piling up.
Most Lean interventions are done at facilities with operations that have been
built up over a period of time having added product lines, technology, and people
as dictated by the business goals and constraints at different junctures. The infras-
tructure has grown over time and machines added in piecemeal fashion by using
available free space. Hence, the layout very often does not support flow. Many
industries also design their layout on functional basis grouping all machines of a
similar type in a location. This may be done for administrative or technical rea-
sons but lead to increased handling and transport of materials and movement of
people. We share a few typical examples of the impact of layout on the flow of
operations.

General Hospital—a patient moves through several floors and hundreds of


feet to complete consultation and tests. In spite of all the IT-based Hospital
management systems, a discharge process even today takes up most of the
day. Process observation at a mid-sized corporate hospital showed that the
ward nurses had to walk over 400 feet and down two levels to hand over the
case sheet for discharge summary. After this, the file moves another 200 feet
to the In-Patient Billing counter after which the patient is called in for final
settlement. Little wonder then that the documents were waiting to be moved
to the next process for 80% of the time as no nurse was going to move one
document at a time over such distances; she would rather wait till a set of
case sheets were ready and take them down to the next process in one shot.
Food Processing—in a leading pickles brand manufacturing unit, a mate-
rial flow diagram showed the total movement to be 200 m spread over 3
floors within one plant. In addition, the low-cost pouch packing took place
in another shed about 500 m away—the expenditure on transport, people
involved in loading and unloading, material movement was approx. 10% of
the total cost of manufacture.
Battery manufacturing—a global player in this sector was reviewing the
layout for an expansion project. It was observed that each product traversed
a distance of 1 km inside the single shed factory, each battery had 75 touches
for the 20 value-adding operations and 60% of the people employed there
were not directly performing value-added work but involved in some form
of material handling and transport.
Textile Unit—in a newly set up line from which products are supplied to
one of the world’s most admired home products company, a single garment
was found to move about 1000 feet horizontally and between 3 different
floors via lifts to undergo 10 value-adding operations during which it was
touched about 39 times and stored in 13 intermediate points.

These situations, clearly illustrate that to create flow, we need to first address the
layout.
104 7 Implementing Lean

7.5 Creating Flow

Not having an appropriate layout is the biggest impediment to flow. Once a suitable
layout is put in place, organizations can shift their focus on to other impediments
to flow.
The Layout is a function of the process flow; hence, to redesign the layout we
would first need to define clearly the process flow. So, our first task is to freeze
the process flow design. We will see that this task varies widely from industry to
industry. Typically, the entire exercise of creating initial flow can be broken up
into four distinct stages:

Stage 1: Refine Process Flow Design


Stage 2: Evaluate and redesign operational layout
Stage 3: Implement New Layout
Stage 4: Run and validate Flow

Information pertinent to the process flow design can be obtained by revisiting


some of the key questions (Rother & Shook, 2003) asked as a pre-requisite for
making a future state Value Stream Map (VSM). Answers to these questions pro-
vide useful insights into designing ideal process flows and layouts.Kindly note
that, the Rother and Shook (1999) has been changed to Rother and Shook (2003)
to match with list.

Q1. What is Our Takt Time?


As we know, the takt time is the rate at which the factory needs to produce to
meet customer demand. Comparing the current operation(s) cycle times with takt
time helps us calculate the number of equipment/work stations needed to meet the
target. Once this is clear, we can work on designing a layout to accommodate the
required equipment and stations.

For years, McDonald’s, the global fast-food giant worked on the model of pull
from supermarket lanes which maintained a standard stock of packed burgers.
But recently, the focus has shifted to making against customer order and to do
this McDonald’s have changed their back-end process flows and introduced
small automations to reduce cycle time of burger preparation.
At Subway, customers wait in front of the counter as their custom-assembled
sandwich is made, and hence, the sandwich maker needs to be both fast and
flexible. The process design requires maintaining inventories of the compo-
nents—cut vegetables, sauces, breads, etc., which are then assembled quickly
as per customer requirement. The counter layout is also designed to minimize
non-value-added time of movement, searching and picking/placing to ensure
fast sandwich preparation.
7.5 Creating Flow 105

Q2. Will We Build to a Finished Goods (FG) Supermarket from Which Customer
Pulls or Produce Direct to Shipping?
The answer to this question directly impacts the entire process flow design. Building
to a FG supermarket gives us the leeway to have standard WIP buffers while working
to direct customer pull means the process has to be both fast and flexible—here
inventories may be kept at component level.

Q3. Where Can We Use Continuous Flow Processing?


Q4. Where Will We Need to Use Supermarket Pull Systems?
Continuous flow is the ideal state best exemplified by an automobile assembly
line—the vehicle moves from one station to the next without stopping in between
and work is balanced between all the stations. Operations with similar cycle times
pave the way for designing such continuous flow streams. But it may not be always
possible to design for perfect continuous flow and we may need to provide strategic
inventory points in between operations to ensure overall flow.
The answers to these two questions depend on the following:

1. Cycle time imbalance: In case of significant difference between cycle times of


supplier and customer process, a decision on flow vs pull is needed. Should
we balance cycle times through additional capacity and flow or should we use
supermarkets to balance the workflow through pull. For example, in engineering
industries, the assembly line works for one shift while the supplier processes
(injection moulding/ CNC) work all three shifts and supply to a supermarket.
2. Technical Process requirements: A specific process may require a decision on
pull vs flow. For example, after varnishing of the stator used in DC motors, we
need to dry the varnish—this can be done overnight in which case a supermarket
is called for or we can design and invest in forced drying to facilitate continuous
flow. Such changes in process method/technology are therefore to be discussed
where required and incorporated at the outset.
3. Equipment Capacity Considerations: Technology dictates that certain equipment
like furnaces, drying ovens, reaction vessels, etc., are made to a minimum capacity
to be economically viable. Hence, the process design has to incorporate batching
before and after such processes which means designing standard inventory points
at these places.

The decision of flow versus pull has a major impact on the layout design. For pull
systems, the size of supermarket and quantity to be held has to be decided, and based
on this the space to be allocated in the layout calculated and its location fixed. In the
case of flow, number of machines/work stations may increase and their space and
physical location with respect to supplier and customer processes need to be fixed.
106 7 Implementing Lean

In the refrigeration example in Sect. 7.3, we saw a combination of flow and pull
from supermarkets. However, while designing the process for their new factory, the
team questioned some of the traditional manufacturing methods. They then decided
to connect all processes through a continuous single-piece flow and do away with
supermarkets. This change in process design was a result of a decision to change
technology and incorporate flexible sheet metal working equipment and innovative
conveyors for in-line body assembly and PUF processes.
Hence, we should appreciate that the questions asked while making the Future
State Map have a big say on the design of Lean process flow and the layout of
both existing and new facilities. The answers to these questions form the basis for
the systematic step-by-step method of Process and Layout Design detailed in the
following sections.

7.5.1 Stage 1—Refining Process Flow Design

In this section, we look at the first stage and the step-by-step method that can be
used to achieve the optimum process flow.

1. Define product/service groups through Product–Quantity (P–Q) analysis


based on projected business
“We have over 200 SKUs in out plant, with this complexity it is impossible to imple-
ment Lean flow here” or “Our industry cannot be compared to Toyota; they make
fixed car models but we need to make a variety of products as per customer designs”.
How often have Lean practitioners and consultants come across such statements by
factory managers and industry owners. These doubts can easily be alleviated through
a simple P–Q analysis.
P–Q analysis relies on the familiar Pareto principle—80% of the turnover/output
is likely to come from 20% of the product variants less than half of which would be
major contributors. The objective of this exercise is to categorize products as runners,
repeaters, and strangers which would in turn have a major impact on process and
layout design as shown in Table 7.1.
The P–Q analysis is equally applicable in service industries and helps in designing
improved process flows as illustrated by the following the examples.

Table 7.1 Product–Quantity (P–Q) analysis and process design


Product category Proportion of total Frequency of Recommended process
required quantity (%) production design
Runners >50 Daily Dedicated flow line
Repeater >30 Weekly/Biweekly Cellular/Group
technology
Stranger <20 Once in a while Filler in cells or
standalone facility used
as and when needed
7.5 Creating Flow 107

Service Offerings and Designing Their Flows


In the Hyderabad Airport, a study found that a significant number of people
travel with only hand baggage—this is therefore a runner category for the
airport services. The process flow was therefore redesigned to have a dedi-
cated airport entry and security line for passengers with hand baggage only.
This entailed changing the layout and this new line was given access through
an entry point located right next to the main airport entrance itself. The
change resulted in reduced waiting time and higher throughput of passengers
handled.
While designing process flow for an automobile service centre, analysis
showed that 40% of the vehicles come in for the routine warranty service.
Another 30% is for planned service like engine overhaul, etc., while the
rest come in for breakdown repair work. So, processes were redesigned and
layouts changed to have the “quick” service bays at the entrance of the
workshop. This meant 40% of the vehicles did not even enter the workshop
and avoided unnecessary documentation processes and waiting time for this
purpose.

2. Runner Products—Calculate takt time and check feasibility of dedicated flow


lines
Runners are high volume products / services that are regularly produced and shipped.
The following flow chart in Fig. 7.2 depicts the steps to be followed to complete
process design for a runner product.
For example, in the case of routine automobile service, the peak inflow is in the
morning hours. Takt time per vehicle is calculated based on the data. The current
service cycle times are known and are also part of the cost sheet—using these we

Measure cycle time of


Calculate takt time
each operation

Compare

Work out number of machines/ workstations


for each operation

Dedicated line will layout these in


operational sequence

Fig. 7.2 Process design for a “runner” product


108 7 Implementing Lean

can calculate the number of bays and technicians to be allotted to have a dedicated
process flow.

3. Repeaters—Prepare product process (P-P) matrix and group products


with similar flows
Repeaters are mid volume products / services that are intermittently but regularly
required.
The P-P matrix helps group the set of machines/work stations/processes that
can service a group of product families. For repeaters, a dedicated line would be
underutilized. Using the matrix, one can form operating cells that can process a set
of repeater products.
While working on designing a layout for toy manufacturing, we found that the var-
ious components required for the final assembly of each toy variant underwent
multiple and varied operations. A P-P matrix was developed to identify the products
that undergo a similar set of processes Table 7.2; components in the same group are
shaded in the same colour.

4. Calculate common takt time for grouped repeater products. input cycle
times in P-P matrix and arrive at process-wise capacities and equipment
requirements
Let us work on the product group comprising Component 2 (of Toy A) and Com-
ponent B (of Toy B). Both components require the painting and printing processes.
Table 7.3 shows the daily demand and operation cycle times for each item in this
group.
Based on this information, the work content per day is computed to be 46,500 s
for painting and 23,000 s for printing or a ratio of 1.9:1—so a cell of 2 painting and 1
printing machines can be planned to ensure process flow. This cell can complete the
day’s requirement for both products in about 13 h. Duly considering the changeover
time between the components, regular break timings for lunch and tea, this translates
to a 2-shift operation.

Table 7.2 P-P matrix for “Repeater” products

Table 7.3 Daily demand and operation cycle times for one group
7.5 Creating Flow 109

In a similar manner, we calculate the machine/work station requirement for all


the other groups and form cells on paper. The cells/groups so formed will then be
fitted into a layout in the next stage.
With the completion of Steps 1–4, the process flow is now defined for runner and
repeater products. Any leftover machines and facilities are now grouped separately
into a standby cell which can be used for strangers and for new product development
purpose while also serving as an emergency backup for the existing lines and cells.
The process flow design can now be frozen, and we can move onto the next stage of
finalizing layout.

7.5.2 Stage 2—Redesigning the Layout

With the process flow designed, we are now ready to work on redesigning the
layout. Layout design for a new facility is much easier than changing layout in
existing facility since we start literally with a “blank slate” in the former while the
latter has to be done keeping in mind existing infrastructural constraints. The basic
steps however remain more or less the same and are described in the following
subsections.

7.5.2.1 Preparation—Inspect Physical Location (Built up Area/Land)


on Site
The following are the key considerations to be understood during physical factory
site observation.

1. Vaastu/Fengshui: Many entrepreneurs closely follow belief systems such as


Vaastu or Fengshui. Vaastu is an ancient Indian convention of planning space
within uilding structure, likewise Fengshui in China. Both of them have their
own guidelines for locations and directions of specific aspects of layout such as
Raw Material receipt, FG Storage, and Dispatch points. They also recommend
where and how to place heavy equipment and equipment generating heat such
as furnaces, ovens, and boilers. Ancillary services such as canteen, offices, and
restrooms are also Vaastu dependent. The layout design may therefore need to
be worked out keeping in mind at least the broad Vaastu guidelines.
2. Legal: Various aspects of manufacturing, health care, and hospitality are gov-
erned by legal compliances. For example, in a food processing plant, a separate
washing area for cleaning trays/trolleys/containers used for transporting the
product in between the processes is mandatory and this should not share com-
mon space with processing areas. Use of ante rooms to quarantine incoming
material is another such regulatory requirement. Locating boilers, ETPs, etc.,
are also subject to regulations of Pollution Control Boards.
3. Access: The receipt and dispatch of material is mainly a function of access. For
example, in a primarily export unit, the biggest (40 feet) container vehicles need
110 7 Implementing Lean

to be able to reach the docking point easily. Sufficient space has to be avail-
able/provided for manoeuvring such vehicles to enable quick TAT (Turnaround
time) which is a key metric for effectiveness of material handling.

7.5.2.2 Designing the Layout


Traditional Lean experts recommend putting up a scaled printout of the layout
on a board and using small cut-outs (to scale) of the machine/work station to
work out alternative layout designs. Nowadays, tools such as Auto-CAD are well
suited to make layout drawings as they are accurate and flexible enough to make
multiple on the spot changes during the course of discussions between the stake-
holders. Alternate layouts can also be made and compared to facilitate quick
decision-making.
Invariably when we make the first layout design, constraints in terms of space or
handling related difficulties will be observed. At this stage, the team is encouraged
to explore alternate solutions and generate different layout options for discussion
and finalization. Some typical situations that come up and are resolved at this stage
are:

1. Lack of space—alternative storage methods like movable racking systems


or vertical storage (Cubic Feet utilized) solutions may help overcome this
constraint.
2. Safety and environmental considerations—in some cases supplier and cus-
tomer processes are kept apart due to factors like dust/heat/noise. We work
out on how to bring them closer say by enclosing the offending pro-
cesses (dust/noise) or using better insulation to reduce heat losses into the
surrounding areas, etc.
3. Quality Gates—Checkpoints are to be created keeping in mind the basic phi-
losophy that a defect should not pass on to the next process. If a defect is
generated, we need to consider how it would be handled in terms of rework and
when, where, and how would the reworked item join back into the main process
flow.

We summarize the learnings from our numerous experiences of (re) designing


layouts for a diverse array of industries in this quick guide shown in the box
below.

Quick Guide to Designing Layout

1. First fix the RM and FG entry and exit points as decided in Step 1
(Sect. 7.5.2.1) and other key aspects keeping in mind Vaastu, Legal, and
Access criteria.
2. Arrange machines/lines as per the defined product groups (cells) and in
order of process sequence. While doing this:
• Maintain close internal supplier and customer process relationship—
the output side of supplier process should match with the input side of
customer process,
7.5 Creating Flow 111

• Fit in sub-assembly/offline processes at locations close to the main


line,
• Ear mark material storage (standard WIP) locations as per process
design calculations.
3. After completing the main process layout, plot the material/service flow
on to it and check if the layout design adheres to the following core
principles of material/service flow:
• No retraction,
• No haphazard or zig zag movements,
• No stopping except where defined (standard WIP/supermarket),
• Minimum possible movement distance.
4. Now add the supporting facilities such as engineering section, office
space, utilities, and gangways in a manner so as to align with the main
process flow.

Tip: Always keep human safety in mind while designing core material flow;
remember that customer pays us for the material—hence, smooth mate-
rial flow is a MUST and gangways, office areas and ancillary facilities are
designed to support this flow.

7.5.2.3 Freezing the Layout Design


The core team may generate several options by end of previous step. These are
then put up for discussion among the stakeholders through structured meetings.
Normally, the stakeholders include process owners, functional heads of quality,
engineering, materials management, Delivery, and HR. In the case of a new facil-
ity, it is advisable to involve the major plant equipment suppliers at this stage
as they would be in a position to elaborate on equipment constraints or specific
requirements which would impact the proposed layout. Also, any customization
needed to implement the decided layout can be discussed and taken up by the
supplier.

The alternative layouts can then be compared on the following Lean metrics:

1. Material handling—distance moved, number of touches,


2. Inventory—number of additional storage points beyond defined standard WIP,
3. Value-Added Space utilization = Value-Added space (core process/machine
footprint)/Total available FSA.

Post-comparison, the team may select the preferred layout option and work a
little more on fine-tuning this before consensus is reached. This final approved
layout is signed off by the stakeholders and then shared with all concerned for
implementation.
112 7 Implementing Lean

7.5.3 Stage 3—Implement Redesigned Layout

However, good the layout and flow look on paper, there are always several tricky
issues when translating the same on to the physical space. The Pareto principle
applies here as well—80% would adhere to the design while 20% may need to
be fine-tuned and adjusted during the physical setting up of the process. A simple
and systematic approach to implement the physical layout involves the following
steps:

1. Start from the end—could be the FG storage / loading dock/ billing counter
and work backwards,
2. Mark each process/machine footprint area by chalk on the ground, taking care
to orient the In and Out of Customer and Supplier processes,
3. Now start placing the equipment in the areas marked starting again from the
end process,
4. Fine-tune/adjust machine orientation by physically positioning operators and
checking ease of work.

Tips for Machine/Work Station Placement

• No value-adder should move more than one step to pick up or place the
material,
• No material to be kept in a position that requires operator to turn around,
• Customer orientation—material should be placed such that the customer
process can directly work on it,
• Where possible, use gravity for moving materials in case the distance
between processes is beyond the one-step rule,
• Mark location of Standard WIP as per process flow design.

7.5.4 Stage 4—Run and Validate Flow

Running the process in the new layout gives us a clear picture on how effectively it
is working and what else needs to be fine-tuned or worked upon to establish flow.
In Lean, we focus on value addition and therefore observe the flow of value-added
activities. From this perspective, direct process observation of running process
should help us identify:

1. Points of inventory build-up between processes,


2. Operators or machines that are found waiting,
3. Extra motion of people to fetch/place things,
4. Any movement of material beyond designated place.
7.6 Pull Systems 113

Items 1 and 2 are a direct indicator of cycle time imbalance between the operations.
We need to work on how to balance the entire line and achieve smooth flow. Items
3 and 4 indicate that we need to fine-tune the material placement and work out how
to make it strain free for the operators. A common-sense approach using ECSR
(Eliminate, Combine, Simplify, Rearrange) works best and fastest in streamlining
this new layout.

Balancing Cycle Times … Streamlining Flow


ECSR is a popular acronym for describing ways to improve processes. Let us
understand it’s relevance to streamlining the flow in a newly changed layout.

• ELIMINATE Abnormal activities and Waste (Muda)—this is useful in


reducing cycle time of bottleneck operations to bring them in sync with
the rest of the line.
• COMBINE Short cycle activities—if two or more operations are much
faster than the takt time, we can combine them into one station and
increase productivity. Similarly, a faster supplier process can help a slower
customer process by “passing the baton” or delivering the material at point
of use thereby reducing non-value-added picking time of the customer
process.
• SIMPLIFY—Operations to eliminate Muda and Muri—again helps to
reduce cycle time.
• REARRANGE Line Operations—for balancing the flow. A good option
here is introducing the material feeder—instead of many operators spend-
ing part of their time in fetching required components, one feeder will
do the running around and supply all the stations at their designated
locations.

In most plant processes, in spite of our best efforts, there will always be some
processes or operations that are highly imbalanced and cannot be brought into line
through the process improvements described above. These are the points where
we need to plan for strategic inventory utilizing appropriately designed PULL
systems. More on this is explained in the next section. From the layout perspective,
once the appropriate inventory points and quantities are fixed, space would need
to be provided for storing this inventory whether on the floor or in racks. Also,
material handling to and from these inventory points needs to be facilitated through
walkways or gangways.

7.6 Pull Systems

Pull systems link supplier–customer processes through defined standard inventories


that operate under a set of rules. Let us relate to pull from this example of a daily
114 7 Implementing Lean

Table 7.4 JIT rules for


Question Answer
standard inventory points
What Specific items/parts to be kept as WIP
When Frequency or time when inventory should be
replenished
How much Quantity to be maintained—minimum and
maximum levels
Where Location for each specific item

routine. Following a long meeting that has cut into the lunch hour, you are hungry
and walk down to the Subway in an adjoining block to order a sandwich. It is
assembled to your specification, wrapped, and handed over for consumption. In
this case, you, “The Customer” are pulling what you want (“Specific Sandwich”),
when you want (when hungry), in the quantity you can consume, from Subway,
“The Supplier”. Subway does not prepare and keep a bunch of sandwiches ready
and displayed at the counter and call customers to come and buy them, which
would be a Push system.
Similarly, the standard inventory points that connect processes in order to main-
tain flow are governed by a set of rules based on the concept of Just in Time (JIT)
management—one of the two pillars of Toyota Production System (TPS). These
JIT-based rules are derived from the answers to four basic questions shown in
Table 7.4.
Kanban is a powerful tool that can answer these questions and manage the
inventory points. Literature on Kanban types, calculation, and method of imple-
menting has been widely published. Studies on Toyota’ Kanban cards and their
rules are also widely available in the public domain. Here, we focus our attention
on visual Kanbans that are simpler, more effective, and sustainable on a typical
shop floor. A good visual Kanban is one which is integrated into the existing
process and can take one or more of the several forms illustrated here.

7.6.1 Space

A marked space on the floor or in a rack when empty or filled can signal the
supplier process to either produce and replenish or stop producing as these two
cases illustrate.

In a factory assembling Air Handling Units, the throughput depends on avail-


ability of the entire kit of assembly components. While implementing Lean,
it was found that kit shortages were leading to delays in assembly.
The factory had a daily target of 6 product units which could be of differ-
ent variants. The kit of fabricated sheet metal parts was delivered on wooden
pallets. A simple pull system was developed to link the assembly section with
the adjoining machine shop. Six pallet-size boxes were painted, on the floor.
7.6 Pull Systems 115

Once a kit is ready at the machine shop, they were asked to place the pallet
in a vacant slot.
At the end of the day, the number of vacant slots is visible and schedule
is made for the machine shop to produce the kits to fill all the empty slots.
If the slots are full, the machine shop stops producing as there is no pull
from the assembly process.
In the deep freezer manufacturing plant we saw earlier in this chapter,
the main transition point is from body assembly to the PUF jig. Once PUF
operation is completed, the body gets onto the assembly conveyor and flows
continuously through assembly, testing, and packing. The freezer bodies are
placed in individual jigs for the PUF operation; at any time, anywhere from
5 to 7 different products are run in parallel.
At this transition point, a simple visual pull system was put in place to
avoid overproduction from the body assembly section. Low floor belt con-
veyors with capacity to hold a maximum of 10 freezer bodies connected the
two processes. Each conveyor catered to one product variant. The PUF cycle
time was 12 min, which means 5 bodies can be completed every hour.
A mark was painted onto the side of the conveyor at the 5-body mark.
This is the signal for assembly section to start assembling more bodies of
the product variant that is moved on that conveyor. The minute the conveyor
is full, assembly stops production.

7.6.2 Storage or Material Handling Containers

Storage containers can also perform the role of Kanban signals. An empty con-
tainer returned to the supplier process can trigger the supplier to produce and fill it.
By defining specific container types or colours for specific items, the signal will
indicate what to make and the container size (capacity) indicates how much to
make. These could be trolleys, trays, plastic bins, or any other types of containers
that are normally used in the process under consideration.

Biscuit Manufacturing—Trolley Kanban


For hard dough variety of biscuits, the process requires that the mixed dough
“stand” for 30–45 minutes so as to develop the specified properties before it
can be charged into the moulding hopper for biscuit forming. Hence, there
was a need to avoid either overproduction or a hand to mouth situation as
this would either compromise the quality or create a gap in the production
line. The total mixing cycle time including charging, mixing and unloading
was studied and found to be 15 minutes.
116 7 Implementing Lean

A pull system was set up earmarking a separate space in the existing lay-
out for a three-trolley lane with each trolley location marked on the ground.
Each trolley has the capacity to hold one batch of mixed dough. Initially the
forming line commences production by pulling in trolley 1 only after the
third trolley comes into its location. By this time, the dough in the first trolley
dough has completed 30 min of standing time. The emptied trolley goes back
to the mixer for refilling. This system was therefore implemented with four
trolleys in circulation at any point of time. If an empty trolley did not come
back to the mixer, the mixer would stop production after the batch already
in process. At the same time, if the mixing operator observes that the trolley
lane is full, he will not commence mixing of the next batch (Fig. 7.3).

Moulder 1 2 3 Mixer
(Forming)

Fig. 7.3 Biscuit manufacturing—Trolley Kanban

This system ensured that the process quality requirements and the flow
were both maintained.

Pull-Based Replenishment Through Trays


At a restaurant buffet, one of the pain points observed from the customer
perspective was a lack of soup bowls during the peak lunch hour. The bowls
are kept in the shelves at the buffet counter. At the counter, guests pick up
and fill the bowls which after use have to be collected, washed, dried, and
put back onto these shelves. Guests were often seen waiting at the buffet
counter for fresh bowls and following up with the staff. These bowls then
be brought from the washing area in a hurry with the staff making multiple
trips to bring out bowls as per need. Even the planned restocking activity
often interfered with the guest picking up food at the buffet counter as the
waiter would take time to pick out the fresh bowls from his holding tray and
place them carefully on the stack on the counter. A maximum of 64 bowls
could be stacked in this manner.
This process was replaced by a “two bin” tray-based pull system. New
trays were procured, each able to hold 24 bowls in a single layer. The counter
could accommodate four such trays in two slots of two trays (one above the
other) each. As a tray became empty, the waiter would take it to the washing
7.6 Pull Systems 117

areas and this empty tray signalled the back-room staff to refill the tray with
clean bowls. As soon as one slot became empty, the waiter would bring in
filled trays and place them in the slot. Meanwhile the guests would continue
to take bowls from the other slot.
This pull based replenishment also saved time as tray replenishment took
only 30 s in this new method as against almost 5 min being spent earlier in
stacking the bowls on the counter.

7.6.3 Electronic Signals

Electronic signals can also be used as Kanban. The most common example is the
sensor-based refilling of containers, tanks and hoppers in various industries. With
the latest developments in Internet technologies collectively known as Internet of
Things (IoT), things like barcodes, QR codes, and RFID tags are being increasingly
used to play the role of Kanban cards.

Palm Oil Packing Machine


The high-speed packing machine line that fills and packs a typical 1-L pouch
is fed by a service tank containing the refined oil. It is not advisable to keep
the oil in stationary condition for long duration and hence the service tank
has a small capacity. The main tank has a stirring system and holds the bulk
refined oil. As and when the oil level in the service tank reaches a defined
minimum, the sensor is activated and signals the pump in the main tank to
fill up. As the level reaches maximum, the signal again goes to the main tank
pump to Stop.

7.6.4 FIFO Lanes

FIFO lanes were first made famous by McDonald’s standard WIP chutes where
a fixed number of pre-packed burgers of each type would occupy chutes slop-
ing down from the kitchen side to the delivery counter. As a pack is taken out
for delivery, the rest of the packs slide down and after a point a signal goes to
the kitchen to replenish the chute. A customized FIFO lane designed for a pizza
preparation and delivery process at a restaurant buffet is illustrated below.
118 7 Implementing Lean

Pizza Preparation
Pizza slices are a runner item at the A’la Liberty restaurant buffet; pizza is
on the menu on all seven days of the week. A cold pizza slice is a big “No!
No!!” for customers, and hence, the timing of preparation is essential. Too
early and the pizza gets stale. Not in time means customers have to wait and
are unhappy. To get over these issues, a FIFO lane-based pull mechanism
was devised to control the flow of Pizza preparation and delivery process
(Fig. 7.4).

Oven 1 Oven 2 Slice and


Ingredients place

Grated cheese
FIFO
Pizza Base 7 5 3 3 1
Pizza Prep
Cut Veggies
8 6 4 4 2

Fig. 7.4 FIFO-based pull of pizzas

The oven is divided into two plates—first is used for pre-baking and the
second for finishing. Each plate can accommodate 2 pizzas side by side and
give a consistent quality. Hence, a simple pull system was put in place with
a fixed quantity of 2 pizzas to ensure no unnecessary WIP or finished pizzas
are made, thereby reducing quality variations/wastage.
As the sliced pizzas are picked up, the two fresh-baked pizzas from plate
2 are removed, sliced, and placed on the counter. The empty trays of plate
2 are now filled by two pizzas from plate 1 (prebake) which in turn pull
pizzas from the preparation section and signal the assistant chef to prepare
two more pizza (base + toppings). If there is any delay in pick up of sliced
pizzas, the ovens are switched to keep warm mode.
7.7 Value Addition Must Flow 119

7.7 Value Addition Must Flow

Our excitement at seeing a perfectly balanced one-piece flow in an assembly line


set-up sometimes makes us forget that what should flow is Value-Added Work and
not necessarily the material. Even in the assembly line, material does move from
one station to the other, so there is an element of waste or Muda of transport. Of
primary importance therefore is the flow of value-added activities in a manner that
the product / service or information being processed spends a bare minimum of
time waiting. We examine two diametrically opposite scenarios illustrating flow
of value added activities is illustrated, thereby covering the entire continuum in
between.

7.7.1 Product is Fixed

There are industries, where by design the material cannot even move, for example,
a building under construction is a fixed object. In such a case, what should flow and
how do we create it? The case below shows us one way to deploy Lean thinking
and create flow in such a scenario.

Towering Heights
Our client was awarded a contract to construct the then tallest building in
Mumbai, India, comprising six basement levels and fifty floors above ground.
The initial part of the construction activity involved laying the slab and
columns for each level to complete the skeleton of the building. When we
first walked into the site, the first two basement levels had been completed by
the team. Each level with about 50,000 square feet of slab and 112 columns
had taken over 30 days to complete, and at this rate, a significant delay was
anticipated leading to the triggering of the late penalty clause. The client
management was hoping that Lean would help turn the situation around.
We had to first come to terms with the unique nature of building construc-
tion. Materials are transported to point of use, but the product is assembled
at the spot and remains there. Hence, we needed to relook at flow from
the perspective of activities rather than the material in order to reduce lead
time. This meant that the people, machines, materials, and supports all had
to move (aMuda) where required in order to sustain the flow of activities.
Once the site team got this clarity, the first thing they did was to observe,
analyse, and identify the “waiting” periods or time during which no work
was happening at the site. Table 7.5 shows the observations made by the
team for the sequence of activities to complete one column.
120 7 Implementing Lean

Table 7.5 Towering heights


No Activity Standard time (h) Actual time (h) Observations
1 Marking 1 7 Waiting for inspection
2 Staging 4 7 Material transportation,
missing parts
3 Cage fixing 2 6 Delay in supply, rework
in positioning as cages
do not match
4 Formwork 6 8 Delay due to error in
order of fitting plates
5 Pouring 1 1.3
6 Setting 12 12
7 De-shuttering 1.5 6 Waiting for gang, some
plates bent
8 Green cutting 0.5 0.5
9 Curing 0.5 0.5
10 Remove staging 1 1.5

The team then worked on improving processes identified through these


observations. A key change was to sequence the work gangs as Cage fixing,
shuttering (formwork), and pouring of concrete were all done by different
gangs. By synchronizing the working of these gangs and ensuring availability
of all materials at site, the flow of activities improved to the extent that the
next level was finished in 20 days and the one after that in just 12 days!

7.7.2 Person is Fixed

The biggest successes achieved by implementing flow manufacturing have hap-


pened in assembled products. Assembly and packing operations are known to be
the most amenable to single-piece flow, and typically, many of these are already
manufactured in such lines. The identification of bottleneck operations through
process observation and reducing the cycle time of such operations through tools
such as Operations Analysis and Line Balancing have given significant results in
terms of productivity and throughput increase.
But is One-piece Flow the only and best way for an assembled product to
be manufactured? Does it always give the highest Value-Adding Ratio leading
to most effective resource utilization? We find that there is an alternative—the
single-person work station; where the entire set of processes is carried out by one
person from start to finish. This actually harks back to the pre-mass manufacturing
days of craft production. These two cases show us that this concept can at times
and situations be actually more effective than one-piece flow. At the same time,
7.7 Value Addition Must Flow 121

deployment of the single-person work station comes with its own set of challenges
and its success depends on the ability and tenacity of the implementing team.

Moulded Plastic Toys


The global toy industry is seasonal in nature with peak sales just before
Christmas, and the manufacturing level fluctuates accordingly. At FS Toys,
the manufacturing team was working to ramp up the output for a blockbuster
toy being produced on a conveyorized assembly line manned by 11 people.
The highest production achieved till date was 800 units per day. The Lean
project team recorded a maximum cycle time of 30 s per toy. They also
observed that at each work station the operator picked up the toy from
conveyor, completed their work, and placed it back again. This handling
itself took around 5–6 s and also interfered with their concentration as the
operator had to keep diverting her attention from the value-added operation
she was doing to pick up the next piece coming on the conveyor. The team
brainstormed and came out with two improvement options to try out.
Option A—Single-piece flow
The belt conveyor was removed, and work tables were reoriented to enable
operators to stand side by side in order of operation sequence. Each operator
could pick up the piece from her right side,work on it and place it to her
left all with a single touch. One person was redeployed as a material feeder
to supply all the parts in front of the respective work stations. During the
production run, the measured output shot up to 1000 units per day with the
same team of 11 people.
Option B—Single-Person Work station
Two single-person work stations were set up next to each other—all six
assembly operations were carried out by the first person and all four packing
operations by the second person. Each station had all required tools—for
example, the first station had two different fixtures, two pneumatic screw
guns, and all the assembly parts arranged in order of assembly sequence.
The cycle time achieved was 75 s per toy in each station which translated
to an output of 260 toys in a whole day for the 2-person cell. 5 such cells
employing 10 people with the 11th person as a material feeder could make
up to 1,300 toys per day!
Here, it was clear that the single-piece work station scored over single-
piece flow as value-added activities flow better. However, the team decided
to go with Option A as they felt that it was difficult to train people to work
on the multifunction mode required for Option B.

Electronic Product Assembly


Linkwell manufactures electronic energy meters against orders mainly
obtained through government contracts. Hence, the business is of a fluc-
tuating nature with peaks and troughs depending on when governments
float tenders for contracts. Further, the margins are low and getting further
122 7 Implementing Lean

eroded due to higher conversion costs. The team started working on how to
increase the productivity and lower cost so as to remain competitive and prof-
itable. In keeping with Lean paradigms, they started with the final assembly
and packing line.
In one line,12 operators worked to deliver 2000 energy meters per day and
Linkwell operated eight such lines. The team observed the highest cycle time
to be 12 s, while the rest of the operation cycle times varied from 6–10 s.
The team brainstormed and tried out two alternative methods to improve
performance.
Option A—Line balancing
Operators sat side by side in a long line in the existing process. The team
modified this to a U-shaped line so that people could stand on both sides.
After line balancing, a smooth single-piece flow was established delivering
an output of 2,000 units per day utilizing only 8 people.
Option B—Two-person cell
A small table was cut out from the existing long table and two work stations
were created one on each side. The first operator assembled the product and
slid the assembled unit to the second operator who then packed the product,
accessories, and instruction cards. This cell was able to deliver 700 units per
day.
The original productivity of 170 units/person/day went up to 250 in
Option A and 350 in Option B. The team decided to run one line using
Option A, one cell using Option B and continued running other lines in
existing process. After a month, the teams met, analyzed the data and con-
cluded that Option B was the best. Within the next month, they dismantled
all existing assembly lines and replaced them with the two-person work sta-
tions. Linkwell found that this two person cell enabled easy scaling up or
down of production based on market conditions. All that had to be done was
to add or reduce the number of such cells. This ensured that a lot of fixed
overheads became variable thereby reducing the cost of manufacturing.

7.8 Flow in Services

Adapting flow to service environments needs a bit of creative thinking as the cus-
tomer is often a part of the process. Both the single person and line concepts
can be applied to such customer facing processes as well as we see from these
examples.
The well-worn Subway example is a case in point. The customer is standing in
front of the counter and watching her sandwich being prepared. All through this
7.9 Enabling Flow 123

preparation process, there is continuous interaction between the supplier and cus-
tomer as the preparation is being customised to the customer’s needs. Obviously
the customer cannot be kept standing for too long as other customers start queuing
up and waiting their turn. The flow of activities consisting of preparing the sand-
wich, wrapping it, add-ons, and billing has been designed as per a single-person
work station framework. Even the flow of interaction that happens while making
the sandwich is in sync with the activity taking place.
A typical South Indian wedding lunch requires hundreds of guests to be served
within a limited time of about an hour. The wedding ceremony takes place in the
morning and tradition dictates that guests eat lunch before they leave, most often
to their workplace. Hence, everyone is in a hurry to eat but the dining hall capacity
is limited. Most often, three or four batches of guests may be served in succession
on the same table. Here, the layout and workflow are adapted to increase speed
of service—guests remain seated while the servers move in tandem serving more
than 20 items during the course of the meal. As each item is delivered directly on
his or her plate (often a disposable banana leaf), the guest spends time only in the
value-adding activity of eating while socializing with companions. The layout has
long narrow tables in parallel with guests sitting on one side of each table facing
each other across an aisle. Servers move in the aisle linearly from one end to the
other serving first one table and then returning via the other table.

7.9 Enabling Flow

Creating the initial flow is the essential first step of implementing Lean. But like
any natural system, this flow has to be nurtured and maintained throughout the
operating period day after day for years. This is never easy as some disruptions
keep occurrig from time to time which affect the output and delivery. It just needs
one or two such disruptions to flow for the management to decide on repositioning
inventory buffers at those places. Strategic inventories may help in maintaining
overall flow, but they bring along their own set of problems.

Inventory is analogous to the water in a sea. The ship appears to be sailing


smoothly when the water level is high and the underlying rocks are hidden
deep underwater. But as the level goes down, these jagged rocks may even
sink the ship if not navigated through. Inventory buffers also have a tendency
to hide inherent problems and thereby become a barrier to improvement
(Fig. 7.5).
124 7 Implementing Lean
Inventory

Transport Flow
Quality Lead time
Layout
Machines ???
Availability Set up time Organisation Absentism

Fig. 7.5 Inventory is analogous to water in the sea

Where no problem is perceived, there can be no improvement

One important purpose of creating basic flow with minimum possible inventory
points is to reveal the underlying conditions or obstacles to smooth flow. Toyota’s
success has been strongly attributed to the urgency in continuously surfacing and
then solving such problems to smoothen the flow and keep improving towards
the perfect state of single-piece flow. Every time the root cause of a problem is
identified and addressed, the flow becomes smoother and remains undisturbed for
longer intervals of time. As the famous philosopher Swami Vivekananda once said:

In a day when you do not come across any problems, you can be sure that you are travelling
in a wrong path

In fact, the largest set of Lean tools which are also the most widely used and
documented deal with ways to solve the typical obstacles to flow. Table 7.6
summarizes typically faced impediments to flow and the tools used to address
them.
Several publications on TPS, Lean, and Kaizen have described these con-
cepts, tools, and techniques in detail. In the following pages, we look at how
organizations have solved specific issues using some of these Lean tools and
techniques.
The very first thing which comes to sight after basic flow is created is the
imbalance in cycle times between operations. As we run the process, pockets of
inventory alternating with idle machines or operators quickly come to light. A
word of caution here, it is not essential that we immediately address each and every
such imbalance but rather identify and work first on the one (or more) bottleneck
operations that keep the process from meeting its customer requirement targets.
7.9 Enabling Flow 125

Table 7.6 List of lean tools and the flow impediments they address
Resource Typical flow Lean concepts and tools Outcomes
impediments used
Methods Cycle time imbalances Line balancing Reduced cycle time,
Work station improvement Increased throughput, and
waste elimination output
Inconsistencies in Standards, Visual Reduce variations and
process Management improve reliability
People Absenteeism/Turnover Single-person Operational flexibility
workstations Reduced fatigue
Multi-skill operations and
cells
Strain-free workstations
Skill dependency De-skilling of process Anyone can do the job with
minimum training
Machines High changeover time SMED (Single minute Reduced changeover time
exchange of Dies) and cost
Short stops and Root cause analysis Minimize stoppages
breakdowns Autonomous maintenance Maintain machine
condition
Material Defects—rejections Problem solving Reduce defects and cost of
techniques (5W1H, 7 QC poor quality
Tools, etc.) Defect does not pass on
Jidoka “Autonomation” Defect is not generated
Poka Yoke “ Mistake
Proofing”
Availability/shortages Pull systems—KANBAN Avoid stockouts as well as
excess inventories

Such operations can be improved by observing and minimizing the three wastes
of Muda, Muri, and Mura as highlighted by this case of Rockwell.

The Heart of the Refrigerator


At Rockwell, the flow of the deep freezer assembly line and its sub-assembly
branches were smoothened through line balancing. The line showed an
improvement in output from about 90 units to 110 units in an eight-hour shift
operation. However, the market demand in peak season required the line to
deliver 200 units per day. Observing the entire line, the team saw that the
freezer bodies coming in from the PUF process onto the final assembly con-
veyor were piling up before the first stage itself. This operation involved
fitting of the base plate sub-assembly consistsing of the compressor, con-
denser, and fan motor (core refrigeration system) on to the freezer bodies
waiting for this operation.
126 7 Implementing Lean

The cycle time of this operation was already in sync with the rest of the
assembly operations for a target in excess of 200 units per day. So why were
bodies held up here? The root cause was found to be the brazing operation
in the base plate sub-assembly line. This critical operation connects various
system tubing components and is followed by a leak test before the sub-
assembly reaches the main assembly line for base plate fitting. Delay in
brazing kept both the base plate fitting operator and the freezer bodies "both
"waiting (Fig. 7.6).

Compressor
Leak Test Brazing
fitting

From PUF Main Assembly Line

Base Plate
Fitting

Pile Up

Fig. 7.6 The heart of the refrigerator

The team focussed on a detailed observation and video analysis of the


brazing operation to identify the Muda, Muri, and Mura and its impact on
the cycle time. The overall cycle time was 240 s which matched the current
capacity of 110 units in eight hours. The value-added part of brazing actually
occurs when two tubes are joined by melting the brazing rod. The team
observed that this was being done at six different points, involved the use of
two types of brazing rods (copper and silver), and took 60–75 s. The rest of
the time, about 165 s, was for various non-value-adding activities including
pick up or placing tools, brazing rods, adjustments, operator motion, and
rework.
Within a couple of days, the team implemented Kaizen-based solutions
such as improvement of the work station to reduce motion and strain, rear-
ranging work sequence for better flow of activities and transferring a couple
of preparation activities to the earlier supplier process to avoid readjustments
by the brazing operator. The cycle time dropped to 140 s, and output of the
line shot up to 190 units per day! Target in sight!!

The workmen on the shop floor are the real value-adders of any organization
and therefore the key resource from any perspective. So often, the absence of one
skilled or specialist operator brings the entire production to a grinding halt. This
“master” worker may start taking advantage of the factory’s dependence on him
7.9 Enabling Flow 127

and starts dictating things leading to further disruption. The recommended way to
overcome this skill dependence is to de-skill the process. This is not as impossible
as many managers think and one of the world’s great Lean organizations IKEA
shows us how it can be done.

IKEA’s DIY Furniture—de-Skilling Perfection


Anyone who has bought a piece of furniture from IKEA anywhere in the
world can relate to the concept of de-skilling. The customer can easily take
home the purchased items in their own vehicle as they are always packed a
flat pack. Each pack contains the kit of components to be assembled, the join-
ery (mostly screws), tools (allen keys), and an assembly instruction booklet.
By following these instructions, you are expected to assemble the furniture
be it a bookshelf, table, cot or chair without having to possess any carpentry
skills, technical knowledge, or tool skills.
This is the ultimate example of de-skilling where any person should be
able to assemble the same item in his or her house anywhere in the world
in a reasonable time without having to possess any special knowledge or
skills. How has IKEA made this possible? Through visual standards that
are simple, unambiguous, and reasonably easy to follow. The entire assem-
bly instruction is purely visual without a single word being printed in any
language. Dos and don’ts are clearly depicted with visual cues. Tools and
hardware are also standardized to a large extent and many designs incorpo-
rate "tool-free" assembly. All this means that even an illiterate person can
follow the instructions perfectly.
If you want to learn more, just go to an IKEA website, click on any
product, and look at the product details to see for yourself how this works!

The Tricolour Problem


A leading manufacturer of electrical cables in the UAE was implementing
Lean to increase output from their existing facility in order to meet the
demands of the construction boom. Flow of material through the various
stages had been streamlined leading to an increase in total output measured
in terms of kilometres of cable. But customer order deliveries were still get-
ting delayed. Further diagnosis showed that the goods were actually held
up in the finished goods warehouse because the entire kit or set of three
coloured cables were not ready for dispatch. Since all three types of cables
(insulation coloured black, yellow, and red) are routed together in a cable
tray at the building site, they need to be available in equal quantities at the
same time.
On analyzing the problem, the root cause was identified as the changeover
time at the extruder, the penultimate process in the value stream. The
128 7 Implementing Lean

extruder sheaths the copper cable with the coloured insulation, and it was
observed that a colour changeover took 35–40 min on an average. To avoid
frequent disruptions and increase the output from each extruder, long runs
of each colour were being planned.
The solution was pretty straightforward—use SMED to reduce the
changeover time. The team conducted a three-day Focussed Improve-
ment activity starting with observation of the current changeover process,
analysing the reasons for it taking time, identifying several improvements
in methods and tools, and enhancing involvement of the available operators.
By the third day, the changeover time was brought down to 10 min which
meant that three changeovers could be completed in the same time as earlier
without impacting the overall output.
Now, all three colours are run even within the same shift producing equal
quantities as per customer requirement and dispatches are effected within the
same or at the most next day.
In recent years, the shift towards employment of contract, migrant work-
ers in emerging fast-growing economies like India has given rise to other
peculiar problems such as mass absenteeism. An entire group of workers
from the same village or town take leave en masse to attend a festival or
a wedding function. The hapless SME is left to firefight to meet customer
delivery schedules. As this problem recurs, the SME looks to automate to
replace labour but is hesitant to take the risk of the high investment. One
way out is to reduce the need for workers by increasing productivity as this
example of seed industry demonstrates.
One of the biggest contributions made by Toyota to Lean is the con-
cept of reducing changeover time. They realized at the outset that the high
changeover time of the sheet metal presses led to long production runs that
in turn resulted in large inventories of parts which disrupt flow. With Shi-
geo Shingo’s help, this part changeover process was relooked at, leading to
dramatic reduction in the time taken. What took hours was brought down to
minutes and gave birth to the concept of Single Minute Exchange of Dies
or SMED. Product changeovers are a source of disruption in most industries
and SMED is the best way to deal with it.

Seeds Packing Productivity


One of the major constraints for a seasonal industry such as PAN Seeds is
the inconsistent availability of workforce. The plant operates with contracted
labour gangs, each gang being allotted one or more godowns as the process-
ing, packing, and storage facilities are called. The workers are involved in
all the activities of the godown right from unloading incoming seed, yard
7.9 Enabling Flow 129

drying of seed (sunning), operating packing line, and loading packed prod-
uct. In the peak season months of November and April–May, gangs are often
found shifting from one activity to another leading to a loss of packing line
output. Data of the past season showed multiple line stoppages as the entire
gang intermittently stopped the line and to help out in other activities such
as Sunning or loading for dispatch.
The Lean project team took up the goal of establishing packing line oper-
ations free of Muda and Muri, such that the line could meet the target output
with a minimum “fixed” workforce. Then, these workers would continue to
run the line through the day while the others would work on the other activ-
ities in parallel. The workers were free to rotate jobs among themselves.
Through a combination of line balancing, improving work stations, and com-
bining operations, the worker requirement for a semi-automatic packing line
was brought down from seven to just three people.
This packing line improvement project was completed in the first week
of the season and all the lines were operated using the new methods for the
rest of the season. Output was about 10% higher than the previous season
even though a lesser number of people were needed to run the line.

Jidoka is the second of the two pillars of TPS; it loosely translates to autono-
mation or applying human intelligence to machines. The use of Jidoka in stopping
the production line when a problem is detected and bringing people together on
the floor to solve it with a sense of urgency has paved the way for the contin-
uous improvements seen at Toyota. Jidoka, one of the two pillars of TPS, has
also proved to be the most difficult to replicate and practise. If the process is
unstable, the line may be stopping several times a day leading to complete dis-
ruption of schedules—no line manager or SME owner is going to accept this. A
way to partially implement this is that whenever the line stops, record the prob-
lem and restart the line immediately. Through this, data on recurring problems can
be collated and the team can prioritize and commence problem solving using the
structured techniques available.
In general, problem-solving techniques are classified into two types—forward
thinking and backward thinking. The first approach is by far the more popular and
uses relatively simple techniques that involve moving from the cause to the effect.
Likely causes are brainstormed and then verified through field observation, data
collection and analysis to arrive at the specific cause(s) actually bringing about the
effect (problem). The 7 QC tools are an example of forward-thinking technique
as is the popular Why-Why analysis. The backward thinking approach is used
mainly to solve chronic and complex problems that are not solved by forward
thinking. An advantage of backward thinking approach is that one need not be an
expert in the process or area where the problem lies. In this approach, facts and
logical reasoning are valued over opinions. The 5 W-1H and Kepner-Tregoe (KT)
techniques and their offshoots are examples of this “detective” method, which
employs a process of deductive reasoning to eliminate various possibilities and
130 7 Implementing Lean

arrive at the actual root cause. The use of simple Why-Why analysis to arrive at
the root cause of a management problem is highlighted in this example of material
shortages for a mechanical assembly process.

The Incomplete Kit


Even after the final assembly and packing processes were streamlined at
Linkwell and hourly production rates improved, the daily actual output
achieved against the plan was found to be varying. The single biggest rea-
son for production shortfalls and firefighting to achieve targets was found to
be the non-availability of the full kit of materials required for mechanical
assembly. A Why-Why analysis was done to identify root causes and the
team then developed actions to mitigate them. As seen in Table 7.7 multi-
ple branches of whys can emanate from the initial problem giving rise to
multiple root causes each needing their specific solutions.

Table 7.7 The Incomplete Kit


Why Why Why Why (Root Solution
cause) Proposed
Full kit not Specific Delay in Wrong data Manual excel ERP changes
available item getting working
shortage stock
Sub Contractor SC capacity Improve SC
(SC) item delay process
Issued to Issued quantity Defined in Change SOP
other SC as per month current SOP
plan

Initially, the sub-contractor (SC) process flows were also observed and
one-piece flow was established to enhance their capacity. But they were also
found struggling because of shortages in components. Based on monthly
plan, the Planner would split the component quantities to each sub-contractor
and instruct stores to issue them kits accordingly. The stores would issue kits
based on this plan which could be for anywhere from a week to a month’s
production. Meanwhile, a SC who came to collect material at a later date
would find shortage in an item that would have been issued in bulk to another
SC. The planner would then intervene and transfer part quantities from one
SC to the other in order to keep production going. Over a period of time, this
practice had resulted in large mismatches of component stocks at different
SC locations.
To ensure equitable distribution of kits and better controls, a new Stan-
dard Operating Procedure (SOP) was defined. Kits equivalent to 3 days
production capacity of the respective SC unit are issued twice a week, each
kite is collected by the SC he has a day’s safety stock from the previous
draw kit remaining in his premises. The units assembled by the SC are also
Reference 131

routed through stores system. To collect a 5000 Nos kit, the SC has to have
produced and supplied 5,000 Nos from the last kit collected date. This gives
control on the conversion and ensures no excess supply of materials. As both
schedule and quantity are fixed, kits for the day are prepared in the morning
and placed in designated kitting areas. SCs are given time slots post-lunch
to pick up their kits and are able do so with minimum waiting time.
With this changed practice, the problem of kit availability could be
overcome resulting in consistent day-to-day production as per designed flow.

The use of pull systems and tools like Kanban to solve problems pertaining to
material availability has already been discussed earlier in this chapter. Strategic
use of pull mechanism enables flow to be maintained without disruption.

7.10 Summary

The chapter plunges into action from the roadmap creation phase discussed in last
chapter. The most logical way to organize and initiate improvement projects would
be to achieve a basic flow first. The core to implementing Lean in an organization
is creating basic flow and then working on enabling this flow to sustain throughout
the operating time. In any sector be it manufacturing or services, layout of the
facility is the key factor influencing the extent and smoothness of material, service,
or customer flow. (Re)designing the process and then the layout on flow principles
will serve as the base to create flow. Once the layout is implemented, the process is
validated and fine-tuned and this marks the creation of basic flow. The basic flow is
complemented and completed by a judicious use of pull systems that regulate fixed
inventories at strategic points along the material’s flow. The triggering mechanism
to replenish the inventory is established through Kanban techniques.
The chapter cautions the reader that a blind flow of materials is not advocated,
rather one must carefully look and ensure that only value-added flow is designed
into the system. Maintaining this flow all the time and under all circumstances is
never easy, and there are any number of factors that contribute to periodic dis-
ruptions. Many popular Lean concepts, tools, and techniques are used to work on
improving processes and overcoming these disrupting influences. Every such con-
straint that is removed enables flow to smoothen further and remain undisturbed
for longer and longer periods of time resulting in a consistent increase in the over-
all throughput of the system. The chapter also presents scenarios from services
industry where all these methods are implemented.

Reference

Rother, M., & Shook, J. (2003). Learning to see: Value stream mapping to add value and eliminate
muda. Lean Enterprise Institute.
Stabilization
8

8.1 Introduction

Process improvement under Lean paradigms is a relatively simple task; however,


maintaining this improved level over time and delivering a sustained result is much
more difficult. In fact, occasional Kaizen bursts leading to process improvements
are far removed from true Lean philosophy. We can relate the Lean journey to
that of a professional sportsperson . In the early daysthe sportsperson works on
technique and fitness to prepare for competition and may suddenly win a medal or
break a record. She or he is euphoric and celebrate’s the success with one and all.
Then, the weight of expectation bears down as she/he is now expected to regularly
perform and win medals at this achieved level for years together. The careers of
sports icons like Lionel Messi, Rafael Nadal, Michael Phelps or Alison Felix are
a testament to years of diligently following a routine and maintaining discipline
be it of diet, exercise regimen, or practise. What we as the audience get to see
and appreciate is only it’s culmination in the matches and victories (the results).
The longevity of these legends is largely due to the years spent in continuously
evolving, improving, and adapting themselves with the changing conditions be it
environment, competition, or their own ageing mind and body.
It is said that in a competitive environment, to maintain the same level, one
needs to constantly strive to reach a higher level. This is equally true of any orga-
nization or process. The Lean path of continual improvement is sustainable only
through a step-by-step ascent striving constantly to reach that ever elusive perfect
state. Implementing a cycle of process improvement under the Lean paradigms
as discussed in Chap. 7 helps us climb the first step. After improvement, the
process needs time to breathe and stabilize before we can plan the next climb
(Fig. 8.1). The stabilization phase ensures that the improvement levels achieved in
process performance are institutionalized and there is no slipping back to earlier
levels, even after attention is taken off the process.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 133
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_8
134 8 Stabilization

Perfection

P D
Improve
A C
S D
A C Stabilize
P D
Improve
S D A C
A C Stabilize
P D
A C Improve
S D
A C Stabilize
P D
Improve
A C
Current State

Fig. 8.1 The staircase of continuous improvement

In this chapter, we discuss the stabilization phase and explore a few key
concepts and techniques that help develop and implement standards. Organiza-
tions that are able to combine improvement and standardization into regular and
well-thought out iterations become truly sustainable Lean entities.

8.2 Standards

What is a standard? From an operations perspective, a standard is defined as the


“simplest and best way to achieve a defined quality level of product and/or service
at that point in time”. In Lean perspective, the standard is the one which distils the
excitement of improvement into a routine and disciplined way of working. Such
standards enable the process to achieve consistent performance every operating
minute.
A secure standard is one which is simple to follow, reasonable in its scope,
and unambiguous in its definition. Such a standard is more likely to be adhered
to without deviations in practice. It is a fact that human beings receive over 80%
of all information through their eyes. Hence, the best standards are the ones that
are visual in nature. For example, the road rules illustrated through signposts are
universal and largely self-explanatory such as the Red–Orange–Green stoplight
which is a visual traffic management standard. The functions of standards include
the following:

• Reveal problems and deviations,


• Enable error-free and uninterrupted process flow,
• Facilitate standard operations,
• Generate people involvement, responsibility, and accountability.
8.3 Standard Operating Procedures (SOP) and Work Instructions (WI) 135

Set / Update standards and communicate them through visual systems

Continuously assess the process status through visual systems

Highlight problems and deviations through visual systems

Take corrective measures through quick improvement cycles

Validate outcomes as desired

Fig. 8.2 Revising process standards

Standardization is all about ensuring the current running process is always follow-
ing the defined standards. In case a deviation occurs, the process owner needs to
act to correct it immediately. Toyota, one of the origins of Lean thinking, believes
that Kaizen is all about “improving and resetting” standards. Every time we com-
plete a process improvement and validate outcomes are as desired, we need to
revise the process standards as shown in Fig. 8.2.
In any manufacturing process, man and/or machine work on transforming
the material using defined methods in a facility; the interplay between all these
resources impacts process performance. Hence, in order to standardize processes,
we need to develop standards for each of these resources and the methods that bind
them together. Figure 8.3 shows us various tools that are used for standardization.
In the rest of this chapter, we discuss the relevance and impact of the stan-
dardization tools through examples of how organizations across a diverse cross
section of industries have adopted these tools to ensure stabilization of improved
processes.

8.3 Standard Operating Procedures (SOP) and Work


Instructions (WI)

SOP and WI are well-known industry practices that are also embedded into the
requirements of ISO certification. In layman terms, SOP lays out the procedure to
be followed for a process to deliver the expected quality and quantity of output
in a safe manner. The WI provides detailed step-by-step instructions on how the
SOP is to be followed. In short, SOP tells us what needs to be done and the WI
explains how to actually do this.
136 8 Stabilization

Fig. 8.3 Standardization


tools

The trick lies in making the SOP practically implementable through an effective
set of WIs as the following example of biscuit manufacturing shows us.

Work instructions—The Practical Standards


A few years ago, we were working with a leading biscuit brand, helping
them achieve product quality consistency across their multilocation factories.
Through this, the goal was to reduce both customer complaints and internal
rejections. The organization had well-documented quality and process stan-
dards which had been in place for years in spite of which the variations in
the manufacturing process remained significantly high.
The preparation and usage of the milk spray solution was one such pro-
cess observed to have a high number of deviations from the standards. This
“milk” formulation is sprayed onto the biscuit surface as it moves on the
web conveyor from the forming stage into the baking oven. On average
about 5,000 biscuits pass through the spray point every minute. The milk
provides a subtle flavour in taste and gives a shine to the appearance of the
biscuit, both key parameters from the customer viewpoint. The formulation
is first prepared at a separate station and then brought in barrels to the biscuit
making line and poured into a chilled holding/service tank of the spray unit.
The standards specify the parameters to be maintained such as recipe of mix,
temperature, pH, and consistency. The kaizen team recorded repeated devi-
ations in maintaining both the pH and temperature (14–18 °C) of solution,
8.3 Standard Operating Procedures (SOP) and Work Instructions (WI) 137

and went on to observe the process to identify the causes for this variation
in the consistency (Fig. 8.4).

Ingredient
weighment
Mix in water
Transfer to Take to line,
with
barrels fill up spray
continuous
tank
stirring
Fill water in
SS contained

Fig. 8.4 Process for milk solution preparation

The milk solution is refilled into the spray tank every three to four
hours. It was also observed that the temperature reached 28–30 °C by the
completion of refilling and then slowly decrease to the specified range
after an hour . This variation was resulting in surface finish inconsistency
over the course of the day . The team made several improvements in the
preparation work station to reduce operator strain , provided for an arrange-
ment of chilled water, and implemented a simple pull-based system between
spray unit and preparation station to avoid overproduction and stocking of
solution in barrels next to the biscuit line. Having validated these improve-
ments, the next step was to ensure 24 × 7 adherence through detailed Work
Instructions (WI).
The WI converted the four-step SOP into detailed step-by-step instruc-
tions for preparing of formulation and supplying against the spray tank
(customer process) requirement. These were displayed in the local language
to facilitate training of the frequently changing casual workers. Finally, to
ensure that workers were able to easily follow WI, visual standards were
implemented. The refill level in the spray tank was marked as was the loca-
tion of storing the barrel to maintain pull system. The uneducated worker
also knew when to top up the tank, where to place the half full barrel and
when to bring the next barrel from the preparation station.This conversion of
an SOP into practical WI implemented with the aid of visual standards was
a great success and replicated across all the factories making this product.

Figure 8.5 depicts the flow of activities needed to ensure complete adherence
to process standards. The key is having a mechanism to identify deviations from
the standards. If work instructions are in place, one then needs to check if they
are practically simple and easy to follow. WI can be appropriately strengthened
by revisions and by making processes visual. We then validate if these steps are
able to ensure complete adherence to standards. If deviations recur, they need to
be further analysed for root cause(s) and process improvements through Kaizen
implemented to eliminate these causes . The WI are then once again revised
138 8 Stabilization

Deviations
Maintain Prepare/ Revise Make / update Validate
Standard Work Instructions Visual adherence to
Standards standards

Repeated
Do Observe and Deviations
analyze for
Kaizen root cause(s)

100% Adherence

Fig. 8.5 Converting standards to sustainable practices through WIs and visual standards

to incorporate improvements done and the validation step repeated to confirm


adherence to standards.
Standards play a vital role in process industries which operate wholly through
equipment and have minimum manual intervention in production. The main job
of the operator or supervisor here is to ensure all equipments are running as
per the standards. This example from a bulk drug manufacturing unit shows how
supervisory standards can be developed to aid this.

The Supervisor Walk


At KRR Drugs, the solvent recovery process was identified as one of the
bottlenecks to enhancing plant capacity. Having reduced the cycle times of
the main reaction process, it was found that getting the required quantity
of solvent for doing the reaction was taking time and the reactors had to
wait. Adding fresh solvent was an expensive option, it is only used as a
top-up while most of the solvent is recovered through fractional distillation
process in columns. After separation, the recovered pure solvent is pumped
into holding tanks from where it is drawn into the reactor as required.
Detailed observations of a few cycles enabled the team to identify the
wastes in the process and work on implementing solutions to minimize them.
The improvements focussed on timely maintenance of parameters like steam
pressure, temperature, chilling water flow and temperature, transfers from
one column to another, and the management of discharge to ensure no mix-
ing of fractions. Having validated the improvements, it became essential to
maintain these practices 24 x 7 to ensure consistency of all batches. The team
developed a set of standards known collectively as “the supervisory walk”.
This is similar to a typical heritage walk that tourists take during visits to
famous historical sites. There is a route map and an audio guide—the visitor
follows the route in sequence, stopping at each marked point where he or
8.4 Autonomous Maintenance (AM) 139

she can plug in the audio and listen to the stories and history of that section
(Fig. 8.6).

Fig. 8.6 The supervisor walk!

In the plant, every one hour, the column supervisor commences his walk
from point 1 on the route map and follows the route in sequence up to the
finish point. Each point marked on the route is numbered on the floor in a
painted circle. The supervisor has to stand inside the circle and can view
the specified task to be done at that point through a cue of visual standards.
For example, at one point, he needs to check the steam pressure and the
pressure gauge is marked with green and red zones and an arrow points
down at the gauge and the valve to make adjustments next to it. The valve
rotation direction is also marked for increase and decrease of the pressure.
The supervisory walk has ensured that the supervisor covers all the crit-
ical points of the process every hour without fail and without missing any
of them. Visual aids enable the entire route to be covered in just 20 min
leaving him free for other jobs for the remaining 40 min of each hour.
Within days of implementing this standard, the process consistency stabilized
completely enabling the plant to increase its output by 25%.

8.4 Autonomous Maintenance (AM)

In the previous section, we discussed how standards help people focus on value
addition and adhere to the defined process. But a lot of the work in manufactur-
ing is actually done by machines and the utilities that go into supporting these
machines. Where product flow and quality depend on machine operations, any
malfunction or a breakdown is going to disrupt this flow. For example, the team
has been able to reduce the cycle time of a bottleneck machine through reducing
Muda and Muri. Now, the machine is expected to deliver a higher output consis-
tently. But at the end of the month, we find that this increase in output has not
140 8 Stabilization

actually materialized to the extent demonstrated in the improvement phase. On


further analysis, we find out that machine breakdowns, short stops due to mal-
functions, and reduced machine performance have all contributed to a reduction in
actual machine running time. Hence, even with an improved or lower cycle time,
the end output is not significantly higher.
Many a times, the machine does not actually stop, but there are minor niggles
that are likely to impact the performance in terms of speed and process consistency.
For example, we often find injection moulding machines operating at a higher
cycle time per part than the standard. Further analysis might show up the fact
that the product quality is impacted due to overheating of the die which in turn
is linked to the below-par performance of the mould cooling system. In this case,
abnormalities in the cooling system have led to the machine speed being reduced
in order to maintain quality.
When we walk through a plant, the machine condition can easily be verified
at a glance by the number of “bandages” seen on it and the conditions of the
surrounding floor area (spillages, oil leakage, dust and dirt). These bandages are
temporary measures to overcome or bypass abnormalities that have occurred in
machine parts. Just like the human body, for a machine too, what begins as a minor
abnormality today may lead to a major breakdown tomorrow if not addressed at
the earliest. This is where Autonomous Maintenance comes in.
Autonomous means Self, and this philosophy of self-maintenance borrows from
our daily life. For example, we have a daily routine of brushing our teeth on wak-
ing up every morning and before going to bed every night. As young children,
this practice was drilled into us by our parents, and in the initial days, they would
even supervise us as we brushed. Over the years, we follow this routine without
even thinking about it. Why do we brush our teeth? The main purpose is to avoid
build-up of plaque and germs that could lead to bigger problems; of course, we
also want to have fresh breath and presentable appearance to the world outside. As
we brush, we also inspect for any visible defect such as reddishness of gums or
discolouration of tooth. We feel for any shaking tooth and sense any pain which is
a signal of some other underlying issue. In case of pain, depending on its severity,
we may undertake either some home remedy or visit a dentist to get it attended
to immediately. This daily cleaning, inspection, identification, and immediate cor-
rection of abnormalities is the core of Autonomous Maintenance (AM). If we
however choose to ignore the pain or do not bother about brushing regularly and
therefore fail to spot an abnormality we may find ourselves heading to a dentist in
an emergency and end up with a major expensive treatment such as a root canal
or surgery.
In AM, we take ownership of our machine and the responsibility to maintain it
in the condition originally given to us. At the first signs of deterioration, we take
care through prompt corrective actions. In a factory, the person who operates the
machine for eight hours every day is the one who knows it best. But he or she
does not actually own it, just like a driver (chauffer) who is a salaried employee
paid to drive and maintain a car. AM focuses on motivating the operator to take
ownership of his or her machine in the same manner as their own personal vehicle.
8.4 Autonomous Maintenance (AM) 141

AM standards then assist the operator in maintaining the machine condition with
minimum effort and spending the least time. AM is one of the pillars of the house
of Total Productive Maintenance (TPM), another managementphilosophy that has
originated from Japan. The Japanese Institute of Plant Maintenance (JIPM) has
been promoting TPM worldwide and has defined a standard for implementing AM
in a step-by-step manner. More can be read about this in any TPM book. However,
the standard TPM methodology talks about a timeline in years and hence does
not hold attraction to the SMEs. Here, we discuss how AM can be implemented
in SMEs in conjunction with Lean within a reasonable time frame of about six
months. The step-by-step approach is shown (Fig. 8.7).
A factory team which diligently follows the above steps will definitely notice
abnormalities start coming down in a six-to-nine-month time frame and also see
the impact of this in terms of reduction in machine breakdowns and process prob-
lems. AM is a very important standardization tool because keeping the machine
operating in its peak condition is a basic pre-requisite for ensuring the continuity
of the improvements done in the earlier phase of Lean.
It not only brings in ownership for the machine and process by the people who
actually run it but also leads to widespread involvement of the entire workforce
making them an integral part of the Lean journey. Most organizations we have

• This is similar to the spring or annual festival cleaning at home, thoroughly done
• Each part of the machine including hard to access areas is cleaned using appropriate tools
Initial
Cleaning

• Abnormality is anything not seen in the original (as bought) condition of a machine
Identify • While cleaning, note down all the abnormalities including even minor flaws
Abnormalities • Tag the abnormalities on the machine to make them visible and a constant reminder

• Try and resolve as many of the identified abnormalities as quickly as possible


Resolve • This sends a message to the operators that abnormalities are viewed seriously by all
abnormalities

• Develop Cleaning, Lubrication, Retightening and Inspection (CLRI) standards


• Institue daily CLRI by the operators to surface and resolve abnormalities
CLRI • Make CLRI visual on the machines to ease and speed up daily cleaning (<10 minutes)

• Make it a practice to record all abnormalities even if resolved on the spot


Record • Maintain abnormality register on the shop floor for this purpose
Abnormalities

• Every three months analysze these records to identify repetitive abnormalities


• Use simple Why-Why technique to analyze for root cause fo such abnormalities
Root Cause • Implement appropriate permanent solutions to prevent recurrence
analysis

• Use data on abnromalities and breakdwons to refine CLRI standards


• Continuously work on reducing time to do CLRI through visual aids
Improve CLRI

Fig. 8.7 Steps to implement AM


142 8 Stabilization

seen make an earnest beginning at implementing the steps of AM. However, in a


majority of them, AM has ended up as just another colourful checklist displayed
on the machines with few people actually practising it as a daily routine. The first
three steps (Fig. 8.7) are done with a great gusto as they are often carried out in
a focussed training workshop or event mode in which cross-functional teams are
excited to outdo each other in identifying and resolving more abnormalities. Step 4
also takes off due to the interest of the maintenance and production team members
in sitting together to develop CLRI sheets, decorating the machine with symbol
stickers, and training the operators to do CLRI.
But as weeks go by, the interest often starts to wane as day-to-day firefighting
due to production target pressures or client schedule changes start taking prece-
dence over the routine of CLRI. The number of abnormalities recorded comes
down as people start accepting some of them as normal condition. Another major
hurdle is the delay in resolving abnormalities either because maintenance team is
over stretched or there is a delay in procuring needed parts. The machine owner
gets de-motivated after a while and starts feeling that recording abnormalities is
futile as they are anyway not going to be resolved.
Providing the right environment and the motivation for people to practice this
daily routine of AM is the responsibility of the top management. Simple measures
like allowing the machines to be stopped for cleaning, encouraging the reporting
of abnormalities, and supporting their quick resolution will go a long way in sus-
taining this important standardization tool. Let us look at how a couple of SMEs
have got down to making AM a success.

AM to the Fore at PAN Seeds


Even for a seasonal industry, the paddy (rice) seed processing and packing
plants operate in a really narrow window. India has two main crops—Kharif
(summer) and Rabi (winter) with the monsoon in between. As the germina-
tion property of seed is critical, it is essential to process and pack it just in
time for the season. Hence, the plant has to ensure the entire market demand
is met during its short runs of only 30–45 days in the pre-winter and 60–
75 days in summer. Any plant breakdowns during this short window lead to
an immediate and irrecoverable loss of sales in the market.
Realizing the criticality of machine uptime, PAN Seeds dived wholeheart-
edly into implementing Autonomous Maintenance as an essential part of the
Lean journey. Due to the seasonal nature of the industry, there are no full-
time permanent operators in PAN. During the season, labour gangs come in
and take up the job of running the packing lines and helping the supervisor
in running the grading line. Despite the transient nature of workforce, PAN
felt it was essential to have some form of AM in place. Initially, the senior
supervisors and plant supervisors (known as Godown boys) were trained in
the first four steps of AM and they personally took ownership for the daily
8.4 Autonomous Maintenance (AM) 143

CLRI activities, filling up the abnormality registers and taking corrective


actions.
In the next phase, the labour gangs were trained on CLRI on the job. In
this phase (Kharif season), daily cleaning and practice of CLRI was done
by entire workforce which also helped reduce the effort and time needed
for this activity. Though paid on a piece rate (per tonne of seed processed)
basis, these part time workers understood that keeping the machine in good
condition was only going to help them maximizing their earnings and took
to implementing AM without hesitation.
By end of May (Kharif season), the plant teams had adapted CLRI
completely. Every evening, the team members cleaned and inspected their
respective machines, noted down abnormalities, and kept an eye out for
sources of dust. Data was compiled in June to understand the impact of
AM on plant performance over the year. OEE of the main plants had risen
from 70 to 85% as a result of significant reduction in breakdowns and short
stops. As PAN moved into the second year of their Lean journey, the team
moved ahead in AM based on the learning of the previous two seasons. The
focus now shifted to bringing down the time and effort for daily CLRI activ-
ities without compromising on its rigour. This could be achieved by doing
the following:

• Few non-critical activities shifted from Daily to Weekly/Monthly fre-


quency after finding no abnormalities reported earlier,
• Separated the activities that can be done while machine is running, to
reduce the machine stop time,
• Identifying and containing or minimizing sources of product dust and seed
spillages thereby reducing the cleaning time, and
• Marking visual standards at all the CLRI points including gauges, oil
levels, valves, motors, panel fans, etc., to reduce inspection time.

Now in the third season, the CLRI which used to take 30–40 min in the
previous season is now being completed within 15 min providing further
impetus to continue doing this routine!

ICLEAN Says… “I Will Clean”


ICLEAN, a pioneer in the design, manufacture, and erection of clean
rooms for the pharmaceutical industry, has invested a lot on state-of-the-art
machines over the years. In 2015, more than six months into their Lean jour-
ney, AM was introduced to help stabilize the plant performance. Thanks to
implementing flow, cycle time reduction, and SMED in the AMADA bend-
ing machine and the powder coating line, ICLEAN had enhanced its output
144 8 Stabilization

capability. Now, it was time to ensure the machines operated in a trouble-free


manner to sustain the improved output levels.
Within a couple of months, CLRI standards were ready to be imple-
mented as a daily routine. The plant operated in two shifts and this begged
the question—who will do the CLRI and when? To avoid eating into “pro-
duction time”, the plant management decided that the major part of CLRI
that required machine stop condition should be done during the shift change
time. The factory already had a 30-min window from 2.00 to 2.30 pm for
shift handover. During this period, the operators of both the first and second
shifts are available at the machine together.
Now, both sets of operators work together for about 15 min, cleaning
the machine and completing the checks as per CLRI standards displayed on
their machines. Abnormalities found are recorded in a register provided for
each section. The maintenance engineer reviews all these registers on a daily
basis during his plant walk and plans with his team for the resolution of the
abnormalities. As and when an abnormality is resolved, he signs off against
the corresponding entry in the register.
This synergy between the production and maintenance teams has worked
well in sustaining the AM practices over the last three to four years.

The need to implement AM is even more acute for SMEs than larger organi-
zations. Most often the maintenance team is limited to a mechanical fitter and an
electrician who report directly to the production manager, there being no separate
maintenance engineer or manager. It is the responsibility of the production head
to take care of the machines in order to meet the targets. Any machine break-
downs are attended to by the machine manufacturer’s team who are likely to have
been given an Annual Maintenance Contract (AMC) for preventive and breakdown
maintenance. The service engineer may be delayed either because of being busy
at another factory or by virtue of having to travel from an outstation location and
this means the breakdown can take a day or more to resolve. Hence, it is all the
more important that SMEs adopt AM to maintain their machines in good operating
condition and minimize the chances of failures.

8.5 Planned Maintenance

Planned Maintenance (PM), the second important pillar of TPM, is complementary


to AM and once again draws from our daily life routines. Let’s recollect our oral
care example from the previous section.While brushing of teeth, gargling, etc., are
part of daily self-care routine, we may also make half yearly or annual visits to
the dentist for teeth cleaning and check-up. The dentist looks for signs of wear
8.6 5S 145

and tear of our teeth and may suggest preventive treatment to ensure long-term
health. This is Planned Maintenance—it simply means periodic inspection and
taking preventive measures aimed to increase the longevity of the part or machine.
PM can be done through Time-Based Monitoring (TBM) and Condition-
Based Monitoring (CBM). Visiting a dentist every six months for check-up is
a time-based activity while going there when you experience a slight discomfort
(abnormality identified during daily self-care) is what we call CBM. Maintain-
ing our car is another good illustration of this concept. We do (autonomous)
self-maintenance in terms of daily outer body dry cleaning and checks like tyre
pressure and some weekly or monthly activities such as wet cleaning, vacuum
cleaning of seats and upholstery, checking of engine oil, coolant and brake fluid
levels. .But there are also certain PM activities specified in the car user manual.
For example, the manual tells us that we should monitor and replace the tyres after
4 years or 40,000 km of use whichever is earlier (TBM). The tyres come with an
ingrained mark (arrow) indicating the tread depth to be maintained for safe driv-
ing. When the tread wears out and the tyre surface reduces to below this mark, it
is a signal to change the tyres; this is CBM.
Hence, PM, whether time based or condition based, focuses on replacing the
worn-out parts in advance so as to avoid major breakdowns thereby extending the
long-term operating life of the machine. It requires a degree of technical compe-
tence and an understanding of the internal workings ofthe machine. This is why,
very few SMEs have even attempted to implement this concept. Most SMEs con-
tinue to depend on external service technicians to do this job, if at all, or simply
attend to breakdowns as and when they occur. However, not reading the signs of
wear and tear or monitoring conditions can prove disastrous when there is an unex-
pected failure and breakdown at a critical juncture and one finds that the required
spare is also not available in the stores. PM checklists are therefore also a direct
input to spares management—based on the expected replacement schedule, the
organization can procure and keep the parts in stock.
A reasonably competent engineer can scan the equipment manuals and be able
to develop a first-cut TBM sheet. CBM may require procuring certain instruments
such as vibration testers and clamp meters, most of which are relatively inex-
pensive and easy to procure. Once the TBM sheet is put into practice, it will
keep getting refined or updated over a period of time, based on data of wear and
tear, abnormalities, and failures. This updation as we already saw is the inherent
characteristic of any good standard.

8.6 5S

Almost a hundred years ago, Henry Ford initiated “CAN DO”, to streamline the
operations of the Ford factory. Decades later, the Japanese established a similar
concept which became popularly known as 5S. 5S goes a long way in stabilizing
the improvements in any process as we will see in this section. The similarities
between CANDO and 5S are captured in Table 8.1.
Lean practitioners believe that 5S is the foundation on which the House of TPS
(Toyota Production System) is built upon. Most like to start a Lean journey with
146 8 Stabilization

Table 8.1 Ford’s CANDO and Japanese 5S


Ford 5S What it means
Cleaning up Seiri (Sort) Distinguish between what is necessary
and unnecessary. Dispose the latter
Arranging Seiton (Set in Order) Enforce a place for everything and
everything in its place
Neatness Seiso (Shine) Clean up the workplace and look for
ways to keep it clean
Discipline Seiketsu (Standardize) Maintain and monitor adherence to the
first three Ss
Ongoing improvement Shitsuke (Sustain) Follow the rules to keep the workplace
5S-right. “Hold the gain “ through
self-discipline

5S; the rationale being that the initial “Sort out” helps realize quick financial gains
through the disposal of unwanted and unused items. This gain will motivate the
top management to pursue Lean further. While this is not a bad start, we have seen
many organizations take up 5S as a one-off activity, thereby failing to capitalize
on its true benefit. 5S involves building up a routine that is to be followed with
discipline thereby making continuous improvements stick. The best part about the
5S concept is that it is as much relatable to the improvement phase as it is a tool
for standardization. Because all 5S begins with an effort to change the mindset of
the people and orient them towards Lean thinking.
We saw earlier that Lean is a paradigm offering an alternate view of the pro-
cesses and activities in an organization. The ability to use this “Lean lens” calls for
a change in our mindset, and it is 5S that provides us the framework for bringing
about this change. Table 8.2 summarizes how the 5S relate equally to the change
in mindset, improvement thinking, and to standardization.
The highest level, self-discipline, can be understood through another daily life
situation that each of us would have experienced at some point in time. Adherence
to the traffic stop lights. In normal working hours, you would be halt as the light
turns red and move on after it goes green. Everyone is doing the same thing, and
in all probabilitly, a cop is monitoring the junction. Now let us say, you are driving
home one night at 2 a.m. and reach an intersection. It is very quiet with only an
occassional vehicle passing by. The junction is unmanned, and the light turns red
just as you reach it. It will be another three minutes before it turns green again. You
are the only one there. Will you wait or go past? When you automatically come to
a halt and wait, you are following the standard (that may be easy to deviate from)
even when no one is monitoring you. You are doing it on your own accord because
you believe in it. This is individual self-discipline. Now, if everyoneon the roads
behave in a similar fashion, the community can be said to have achieved the 5th
S—Sustain.
Only a handful of organizations have actually reached this level of self-
discipline. There may be individuals or sections within an organization who
8.6 5S 147

Table 8.2 5S related to personal change, lean improvement, and standardization


5S Personal Change Lean Improvement Standardization
Sort Remove unwanted Eliminate unnecessary Discard unused and
thoughts and wastes (Muda, Muri, unwanted items from the
information clutter and Mura) workplace
Set in Order Clarity of Arrange value adding Arrange the required
thought—prioritize processes in flow items to maximize value
what to do when addition
Shine Keep polishing up Make abnormalities Keep workplace clean
your knowledge and and deviations visible and inspect for
skills abnormalities
Standardize Make it a practice to Write and implement Monitor adherence to
follow the first 3S standards for the first 3S’s through
improved process assessments
Sustain When the 4S becomes Follow improved Everyone in the
(Self-discipline) a way of life process without organization practices
deviations under any 3S even when no one is
circumstance monitoring

may be there but most organizations reach the 4S level and depend on periodic
assessments to motivate and drive employee teams to keep practicing 5S.
So how does 5S typically get implemented? The factory is divided into Zones—
a section (say machine shop) or a line or a department (e.g. stores) can be one
Zone. Each Zone has a designated Leader who takes ownership for 5S in his or
her zone. After a formal kick-off session, the Zone leaders and their team members
undergo a few rounds of practical training. The best way to start is to do an initial
assessment of all the zones using standard 5S checklists. Each zone is scored
and improvement opportunities identified from amongst the low score parameters.
The initial score sets the base level from where the zone is expected to improve
in practice and adherence of 5 S and strive towards higher scores. A quarterly
assessment in conjunction with a Reward and Recognition (R&R) program keeps
the momentum going as each zone tries to become the best. The 4th S is the
critical stage of 5S implementation. The organization needs to have patience and
will to stick with these assessment cycles and R&R programs for at least 3 years to
ensure they become a part of routine. As scores start plateauing, the 5S standards
in the checklist themselves need to be reviewed and improved to take the factory
to the next level. Figure 8.8 depicts a typical 5S implementation cycle.
This implementation cycle serves the purpose of not only sustaining improve-
ments but also to bring in a culture of continual improvement. Teams need to
improve their workplaces, methods, and processes further through Kaizen activi-
ties in order to increase their scores. After at least two such cycles, the organization
may reach the Sustain level where Lean and 5S is a way of life.
148 8 Stabilization

Improve 5S
Standard (updated
checklist)

Half Yearly
Assessments

Quarterly
Assessments

Initial Assessment
(Base Level)

3 months 1 year 2nd year 3rd year

Fig. 8.8 5S implementation cycle

Sustaining Lean….How Linkwell is Implementing 5S


At the end of an improvement packed nine-month Lean journey, Linkwell
formally launched 5S across its multiple factories with an initial assessment
done by the Lean consultant. A basic 25-point checklist covering six key
aspects of operational excellence, viz., Safety, Quality, Machines, Work-
place, Material Flow, and Daily Work Management, was used to assess Lean
sustainability.
Each production line or standalone unit was designated as a Separate
Zone while non-production zones included stores, finished goods warehouse,
utilities, laboratory, and administration. The assessment key was tweaked to
have variations for each type of Zone. For example, stores or warehouse was
assessed more on material storage systems and documentation while these
were less emphasis on machines, tools, dies, and fixtures. The reader can
view the typical checklist used in the detailed case study on Linkwell.
Simplicity is the key to making such assessment-based models a suc-
cess. An easy-to-understand and use 4-point (0-2-4) scale was used to score
each parameter. For example, one of the questions focussing on Autonomous
Maintenance, reads “Is the machine maintained in a clean condition?”. If the
observation showed one or two instances of dirt or oil spillage, then a par-
tial score of 2 marks was given. Perfect cleanliness earned a full 4 marks,
while several instances of such “non-conformances” resulted in a Zero-mark
score. Since the scoring was based on direct observation of the process and
8.6 5S 149

assessment done jointly along with the concerned Zone Leader, there was
very little subjectivity involved.
The assessment process itself evolved to ensure a seamless handover of
the Lean initiative from the external consultant to the internal Lean Core
team. The initial assessment was done completely by the consultant with the
core team observing the process. The second assessment was done jointly.
The third assessment was done by two-member internal teams and moderated
by the consultant. To avoid any bias or variation, one team covered all the
production zones and the other team focussed on non-production zones.
Scores from the third assessment were used to prepare the honour rolls. In
a special awards function chaired by the Executive Directors, thetop scorers
in production and non-production categories were presented a rolling trophy
for Best 5S Zone . The Executive Director of the company also personally
handed out individual certificates to the team members of the winning Zones
and reiterated the management’s commitment to Lean in her closing remarks.
Has Linkwell been able to sustain 5S and through this continue the Lean
journey over the next two years? Read more about this in the case study on
Linkwell in the latter part of this book.

5S competitions are a positive way to motivate the teams to keep practicing and
can even be run between multiple organizations. For example, in India, the Con-
federation of Indian Industry (CII), a national body, invites members to participate
in the annual competition. Implementing Lean under the National Manufacturing
Competitiveness Program, cluster program gave us an opportunity to utilize 5S as
a competitive tool to encourage participating industries improve their workplace
and sustain the Lean journey. A snapshot of this is shown in this snippet.

Inter Factory 5S
Under the National Manufacturing Competitiveness Program run by the
Indian Government, SME clusters were formed to share consulting resources
and implement Lean to improve their profitability and long-term sustain-
ability. Of the eighteen-month schedule, the last six months were devoted
mainly to implementing 5S and sustaining the improvements made through
5S assessments.
The Light Engineering cluster included several small organizations, and
we designed a simple 20 parameter checklist to assess them and guide them
on 5S. The same checklist was used for all the 10 participating units which
meant that the scores could be compared and made available to all of them.
Each factory was able to see where they stand with respect to the others and
were therefore motivated to be seen as the best. This healthy competitive
spirit helped in pulling up the scores of nearly all the factories over the six
months (Fig. 8.9).
150 8 Stabilization

100

90
91
80
80
70
68
60
62 60
50 56 55
48 50 47
40 47 45
43 44
30 36 35 35
33
20

10

0
Textek MAS Solar Rangamma GEM Ramdevs CEC Rajinikanth VPF Sundar
Motors Instruments

Fig. 8.9 5S Assessment score card—LE cluster

Chart comparing scores of each factory for Assessment 1 and Assessment


2.
The improvement in scores was a reflection of the shop floor adherence
to Lean practices that had been implemented over the previous nine months.
Unfortunately, the smaller factories have found it difficult to sustain Lean
post-withdrawal of the external Lean consultants. When we surveyed the
practices of one of the clusters a year after we completed our engagement
with them, many of them had stopped the 5S assessment and the continual
improvement journey. The single biggest reason for this was that the desig-
nated Lean champion had left the organization and there was no alternate or
backup person to carry forward the initiative. We discuss more on this in the
next chapter on sustenance of Lean in SMEs.

8.7 Summary

In this chapter, we discuss various steps an organization can take to guarantee the
continuation of improved performance of a given process after a Lean intervention.
A typical Lean intervention is done in an experimental mode, and in order to
stabilize the results, certain resource conditions need to be institutionalized. For
people, this is achieved through documented work instructions/standard operating
procedures. For equipment, autonomous and planned maintenance are a way to
not only extend their operating life, but also operate within specifications. 5S is
an organization-wide initiative to stabilize the process improvements and pave the
way for continual improvement. The methods stipulated in these SOPs, AM, PM,
8.7 Summary 151

and 5S documents are collectively referred to as standards. Standards reflect the


overall organizational knowledge of the current best procedures. Lastly, they are
upgraded every time a continuous improvement project needs to be embedded into
the regular functioning of the organization.
Sustaining Lean
9

9.1 Introduction

Management involves culture setting, people management, organizational routines,


performance management systems, and the way organization interactions with var-
ious external stakeholders such as customers, suppliers, government, and public at
large. It also involves design of end-to-end value delivery process. Lean manage-
ment specifically ensures a smooth flow of value across departments and busts any
batch formation to serve the convenience of departments for e.g. large machines,
huge changeovers, big facility, long lead times, etc. A Lean organization is focused
on creating more value for the customers while using fewer resources. This is made
possible by a heightened awareness of the customer value, product quality, process
capability, among very empowered workforce. A deep understanding of the pro-
cess capability is essential to keep it available when there is a pull signal and resist
from just-in-case pre-production. A Lean sales operation tries to stabilize and pass
true demand to operations, as against quota-driven push of discounted sales. This
results in a stable flow of orders across the entire supply chain, thereby elimi-
nating variance-related inventory expenses across the chain. The cycle stocks are
reduced by reducing the changeover, transportation, and other non-value-adding
components of lead time. At the foundation of all this is a steadfast commitment
to quality at the source to keep a tab on quality-induced uncertainty. No defect is
allowed to enter the system, and a 100% inspection is instituted through process
autonomation. If a defect is found, an immediate cure is performed to eliminate
the root cause and the recurrence of the problem.
Lasting results from Lean management accrue to firms that ingrain the Lean
thinking into the firms’ daily routines. Results ranging from ‘less than best’ to
‘dwindling within a couple of years’ are seen by a large number of firms that
adapt Lean management, either because they picked only few techniques and not
the whole thinking or did not create a series of ever increasing challenging kaizen
projects to maintain the momentum. In the initial years (upto 5 years), it is easy for

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 153
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_9
154 9 Sustaining Lean

firms to slip back to unthinking, batch mode of old ways of doing things, riding on
the success of Lean in getting through the immediate crisis/challenge. So, for Lean
to unleash its full and cascading benefit, the organizational culture must be such
that it relentlessly pushes the processes towards the ideal, never resting in status
quo. However, organizations that do not go whole hog, also do realize benefits to
the extent steam lasts in them or by virtue of the improvements already done.
And yet, unfortunately, a number of organizations (including some very big)
limited their understanding and hence application of Lean to inventory control,
process improvement, lead time reduction, or such operational goals. The result of
such narrow thinking is now evident in the form of public outcry in many forums
that JIT and therefore Lean are the causes of shortages in the pandemic. One can
easily verify the falsity of these claims by looking at the performance of truly Lean
organizations versus those whose Lean adoption is atbest partial.
Toyota has been on the Lean journey for over 70 years now, and the DNA of
continual improvement and Kaizen culture is well entrenched. But of the thousands
of organizations across the world who have been trying to follow “The Toyota
Way”, only a handful have truly sustained continual improvement. When it comes
to SMEs and family-managed businesses, the record is even more abysmal. Our
own studies of SMEs that were part of the Lean cluster scheme have thrown up
certain recurring issues that have primarily contributed to this failure to sustain
Lean. We look at these issues and try to identify ways and means to tackle them.
Based on our numerous implementation experiences, we observed that SMEs
that sustain Lean master a few key maturity stages. Our model highlights various
short-term temptations for the organizations to overcome in their Lean journey,
and through this model, we want to make SMEs vary of potential paths of least
resistance, so that they can challenge themselves with higher goals. Good news is
every organization has to go through these hoops of ever-increasing improvements,
and bad news is only a few cross-over all the hurdles in their journey to perfection.
Nonetheless, the journey is itself very rewarding for every organization, regardless
of the stage. After describing the model, we then proceed to review the factors
that contribute to persistence or dropping of Lean, distilled from our interations
with a number of SME owners and managers. The chapter is concluded with the
summary of empirical findings from a study of Lean sustainability conducted by
the authors in an SME cluster from Southern India.

9.2 Sustenance Stages of Lean Adoption

The following Fig. 9.1 shows various stages of sustenance of Lean adoption. These
stages are briefy described below. All the case examples we have provided in this
book relate to SMEs that are on the Lean path of their own will and not due to
customer or regulatory enforcement. The automobile industry is an example of
enforced Lean where the car manufacturer (OEM) works with and expects Tier 1
and Tier 2 vendors to follow JIT and other Lean concepts to align with the OEM’s
requrements in terms of quality, quantity, or price. Vendors who are lagging may
get downgraded and orders reduced.
9.2 Sustenance Stages of Lean Adoption 155

V. Lean
Thinking
IV. Enhanced
Lean
III.
Standalone
II. Successful Intervention
Pilot
I. Tried and
Failed

Fig. 9.1 Sustenance stages of Lean adoption

9.2.1 Stage I: Tried and Failed

Firms in this stage are those who have tried Lean and percieve no improvement
to their operations. Typical reasons for these initial failures are (a) not having or
creating a compelling narrative for Lean, (b) not creating circumstances suitable
for Lean implementation, (c) trying to implement Lean without the right Sensei,
(d) not providing basic training on Lean to the core team, and (e) half-hearted
Lean implementation without empowered team, etc. When the management does
not see the results and percieves that the effort is “eating away” the productive time
of the team, they would quickly dismantle the team and go back to old ways of
doing things. They would further justify their inaction with some excuses such as
Lean is not for this industry, or we just do not have bandwidth. Following caselet
illustrates one such example.

MGP, a producer of polycarbonated sheets, had set up a spanking new facility


riding on the high market demand. The partners were earlier importing these
products and trading them but spotted the opportunity to manufacture and
sell the products themselves. Their children had recently graduated from
management schools, and one of them had visited Japan as part of the course.
She was enamoured by what she saw there and invited her professor to visit
the MGP factory. After meeting the partners and a half-day training cum
observation exercise on basics of Lean, the partners agreed to do a Lean
pilot to make their plant operations efficient. However, there were no clear
business goals defined for this initiative.
A couple of months into the implementation, notwithstanding small suc-
cesses like productivity increase in a couple of processes and systematic
arrangement in the stores and warehouse, the initiative lost steam. The part-
ners used to sit in the corporate office and visited the factory only once
in a while. There was no consistent communication from them to the plant
employees on Lean and its importance.
156 9 Sustaining Lean

By the third month, one of the partner’s started questioning the fees being
paid to the Lean consultant, and the value MGP was gaining out of the
exercise. At this stage, Lean stopped.

9.2.2 Stage II: Successful Pilot

Typical approach to Lean implementation is through a pilot project, in which a


small area is chosen for testing, training, and showcasing the results to manage-
ment, supervisors, and workers. Firms that reach this stage are the ones that have
got the intended benefits from the pilot implementation. However, if the chosen
area for pilot Lean implementation is the major bottleneck of the firm and if the
performance improvement is enough for the firm to tide over its immediate needs,
then it may so happen that the firm may decide to not pursue/scale Lean to other
areas. Ironically, the success of Lean itself in this case is the reason for the firm
to not extend Lean. Some firms after a while, even slip back to previous non-Lean
methods at the pilot area, perhaps because the crisis has passed on. Following
caselet illustrates one such example.

RDN, a fabricator of structures used in the solar industry, was struggling


to meet the customer requirements which had piled up due to simulatneous
wins of several government contracts. The second-generation entrepreneur
had studied operations management a couple of years ago and reconnected
with his teacher to take advice on sorting out the plant operations. We were
called in to help RDN and suggested implementing Lean to increase output
while reducing inventories and lead times. RDN agreed for a six-month Lean
intervention to be extended to a longer term association.
The first three months were a revelation to RDN. The plant layout was
completely changed to get flow. Bottleneck process of welding was improved
through operations analysis and waste reduction. The internal supplier cus-
tomer links especially to the galvanizing sub-contractors was strengthened,
and finished goods warehouse streamlined to facilitate count-free, search-
free storage, and easy retrieval of items for dispatch. The operations team
was young and energetic and seemed to be enjoying Lean.
RDN was happy as orders started to get fulfilled and material shortages
started coming down. Then, the Lean intervention petered out. RDN man-
agement felt the team’s bandwidth was not adequate to continue the Lean
journey at the fast pace in which it had moved till then. They wanted time
to hold the gains and prepare their team before restarting the next phase.
9.2 Sustenance Stages of Lean Adoption 157

It has been about three years now, and RDN has not moved ahead. They
have made some strategic decisions like starting their own galvanizing facil-
ity and integrating the plant processes. They continue to maintain the flow
and pull systems set-up during the pilot Lean phase. And remain content!

9.2.3 Stage III: Standalone Intervention

Several SMEs who have tasted success in the pilot stage have gone on to imple-
ment Lean across the major part of their value stream. They have seen an improved
operational and business performance during this period of Lean implementation
and have been satisfied with their achievements. However, they have not felt the
need to build a culture of continual improvement or build internal teams to take
Lean forward. One of two things generally happen to them in the long run. They
sustain what has been achieved over time without really trying anything new or
improving further. Or, they start slipping back to some extent over the years. In
both cases, since the organization has tasted the fruits of Lean, they genereally end
up reconnecting with their Sensei either when the business faces new challenges or
when things start slip up badly. For these organizations, Lean is not a conitnuous
journey but a series of intermittent standalone interventions. A bulk of companies
typically belong to this category. A caselet is described below.

We had an opportunity to work with a seafoods pioneer way back in 2010.


The industry was limping back after two years of global disruption, and
the CEO suddenly found that his plants were finding it difficult to meet the
rising demand inspite of having the capacity. This was a continuous pro-
cess for making prawn feed, and initial data showed a low OEE of 50%
due to a variety of factors such as quality problems leading to running the
mill at low speeds and die changeover time being high. A six-month inter-
vention ensued, stabilizing process parameters (eliminating Mura), reducing
changeover times, and streamlining the raw material handling leading OEE
to shoot upto 78%. The CEO was pleased, and we ended the engagement on
a high note. The plant sustained and kept up its performance over the next
few years.
Meanwhile in 2014, the company now growing rapidly sets up a new
plant on the west coast to cater to the market there. Within the first year, the
familiar issues started to surface and the CEO once again got in touch with
us. We worked with this plant for almost nine months with similar results
and left them to sustain.
By 2020, the SME has grown to a large publicly listed company with
global tie ups and has set up two more plants. They have recently roped
158 9 Sustaining Lean

in an international consulting firm to work on improving plant performance


across their plants!!

9.2.4 Stage IV: Enhanced Lean

Firms in this stage truly understand Lean well enough to be able to confidently
apply them in other production/operations areas with their own internal team and
without any support from external Sensei. The Lean team is clearly identified,
trained, and empowered to deploy the concepts to pre-determined improvement
roadmap. Management in these firms in this stage develops, commits to, and
communicates improvement roadmaps.
Reaching this stage is not easy. Over the years, there are so many challenges and
disruptions that can derail Lean such as business cycle downturns, key employee
turnover, pandemic, product life cycle changes. The firms in this zone are the ones
with discipline of execution, firmness of vision, alacrity to adjust to internal and
external changes, take failures on their stride, and move on to keep getting better.
The following caselet is an illustration of such an example.

Our client, a supplier of facility equipment such as clean rooms, Air


Handling Units, clean room accessories, laboratory furniture to the pharma-
ceutical industry, has grown tenfold in the last decade. In 2010, they were
categorized as a small to medium industry, operating in a couple of factory
sheds when the top management happened to attend a Lean seminar at their
customer’s (a global pharma major) training academy. We started a year-long
Lean journey helping them implement basic flows and gain throughput and
reduce wastes throughout their operations across product streams. Five years
later, they reconnected to take their plant operations to the next level and we
strengthened Lean with stabilization tools such as AM and PM in addition
to enhancing the flow and further reducing WIP in between the downstream
processes.
When they had grown big enough to relocate the mutliple business divi-
sions to a new green field factory, they used Lean right from the initial
process flow and layout design stage. A core team of Factory Manager,
Methods Manager, and Director (Ops) who were all well versed in Lean
by this time worked on translating this Lean layout to reality.
In 2020, they decided that Lean should become a regular part of their
developmental activity and formed a core four-member Lean team from
existing staff. Customized 5S assessment checklists were used to moni-
tor zone performance and motivate team members to keep improving their
scores. The past year has seen two rounds of such improvement. The team
9.3 Factors Affecting Lean Sustenance 159

also regularly conducts internal refresher training programs on relevant Lean


concepts. In parallel, the factory at a second location is also on the Lean
path to catch up with the main unit.
As the company is on the threshold of breaking out of the SME category,
the management has acknowledged the part played by Lean in their growth
story.

Meanwhile, the case of Linkwell is another good example of an SME driving Lean
in the long term. Please see the appendix for the detailed case study of their Lean
journey so far.

9.2.5 Lean Thinking

Think of a mini Toyota or an IKEA. Firms that reach this stage have extended Lean
to not only all the support functions within their organization, but also encourage
their customers and suppliers to adopt the same. These firms would institutionalize
Lean, by linking Lean goals to individual performance appraisal. Most employees
adopt Lean as part of their regular work, and in other words, Lean philosophy
is their way of thinking. Lean is part of the organization DNA and runs in the
day-to-day thinking and decision-making of each and every employee working
there.
Unfortunately, very few SMEs have got to this stage, though several of them as
we saw in Stage IV are likely to get here soon. Of course, by the time they reach
this level, they are often no longer SMEs thanks to their sustained growth.

9.3 Factors Affecting Lean Sustenance

Factors contributing to either sustaining or dropping of Lean are categorized into


the following sub-groups.

9.3.1 People Related

The most common cause for dropping Lean is attrition of key employees,
especially the Lean champions who are able to get alternative employment
opportunities thanks to their knowledge of Lean. Another is the shifting/migrant
workforce and their lack of ownership for the process/organization.
160 9 Sustaining Lean

9.3.2 Scaling Across

In most organizations, Lean strategy is first formulated and implemented for the
core operations. Once operational efficiencies have improved, the pressure of busi-
ness growth often shifts to the front-end sales and marketing. Procurement is also
pressed to deliver and follow principles like JIT to align with the core operations.
At this stage, the organization needs to initiate Lean in the other functions includ-
ing purchase, sales, accounting, and HR, among others. If this natural roll out does
not happen, there is imbalance among the functions leading to a gradual drop in
the sustenance of Lean in the core operations.

9.3.3 Creating Headroom

Perhaps, a less spoken reasons for Lean to lose steam especially in its early stages
(within 2 years) are because of the organizations’ inablity to balance the long- and
short-run effects. One of the immediate effects of Lean is increased productiv-
ity, capacity, and lowering of direct costs. If there is no sufficient demand for the
newly available capacity, senior management finds it difficult to justify the fixed
expenses such as wages. They may be forced to downsize, the very anti-thesis for
Lean to sustain (Sterman et al., 1997). Thus for the SME to keep up the momen-
tum, it is important for its senior leadership to find suitable avenues to deploy the
released capacity so that the true long-term outcome of Lean to achieve higher
profits for the organization is ensued rather than a debilitating short-term cost
reduction. Long-term commitment of the top management/owners and their inabil-
ity to align Lean initiatives to the changes in business environment and strategy
also leads to a roll back of Lean over time.

9.3.4 Making Own Lean Operating Model

In many cases, the top management has not really understood Lean philosophy,
but only the few select tools and techniques were deployed which gave them quick
benefits and stopped at that. The few firms we spoke where Lean sustained were
unequivocal about making Lean as their own manufacturing system. By this, we
mean that the day-to-day running of the organization is completely governed by
Lean principles.

9.3.5 Periodic Lean Assessment

Having a periodic assessment of Lean activities and results and incorporating it


into the organizational reporting hierarchy help in sustaining Lean. A direct benefit
of periodic reporting is that it helps keep track on the progress of Lean implemen-
tation and early course correction whenever a deviation from expected pattern is
9.4 Empirical Study of Lean Sustenance 161

observed. In SMEs, regular reviews (fortnightly/monthly) by the owners keep the


pressure on the plant team while also providing them with timely decisions and
resources to complete their activities. In the larger context, this mechanism helps
in establishing a data-driven decision-making culture, which is essential for Lean
to survive and thrive.

9.3.6 Top Management Commitment

Much has been said about the importance of top management involvement to
ensure the success of Lean in the first place and sustain in the long run. Apart
from the pressure to show improvements, this also passes an important signal to
the operating staff that the senior management is serious about the Lean results.
One of the SME owners we spoke to was emphatic about the importance of senior
management’s preparedness to get their hands dirty, willingness to change their
worldview, and willingness to learn the right way. When the senior management
is seen to be prepared to change their perspective and is willing to experiment, the
next levels of management would find it easier to adapt and convince the rest of
the organization.

9.4 Empirical Study of Lean Sustenance

One of the authors of this book was engaged with ten manufacturing SMEs, part
of a Lean cluster, based in the southern part of India. This Government of India
sponsored program had the following five objectives:

i. Reducing waste
ii. Increasing productivity
iii. Introduce innovative practices to improve overall competitiveness
iv. Inculcate good management systems
v. Imbibing a culture of continuous improvement.

The author led a team of Lean consultants that worked on implementing Lean
at these SMEs. The 18-month long Lean intervention ran the whole gamut of
training, identification of improvement opportunities, implementing the same, and
finally setting up processes to ensure sustenance of the Lean management. The
implementation resulted in differing levels of success at individual units which
were recorded and documented in the form of case studies and a summary report
submitted to the monitoring government agency. Out of our own interest, we sur-
veyed these SMEs about a year after the consulting team disengaged with the
cluster to understand the levels of sustenance. This section presents the differences
in sustenance of Lean in different companies.
The cluster of companies measured and undertook improvement targets in the
following four groups of operational metrics.
162 9 Sustaining Lean

1. Productivity improvement (PI-A, PI-L): asset and labour


productivity/utilization-related metrics,
2. Process improvement (PI-CT, PI-WIP, PI-P, CC): cycle time, WIP, and
customer-centric metrics,
3. Culture building (CB): implementation of systems such as Kaizen reporting,
responsibility matrix, and checklists is grouped as culture-building metrics,
4. Quality improvement (QI): defect reduction and implementation of quality
management systems are covered under this head.

The overall improvement (in terms of reported instances of increased value for
each of the metric) soon after implementation is as shown in Fig. 9.2.
Of these group of metrics, process improvement (PI) and culture building con-
tribute to long-term sustainability of Lean. When these companies were surveyed
for the status of the metrics after about a year of implementation, the following
were reported.

• Unfortunately, largest deterioration occurred in culture-building metrics (70%),


• Within culture building, 5S, Kaizen reporting system (KRS), and TPM imple-
mentation are the reported metrics which deteriorated,
• None of the productivity improvement metrics have deteriorated. Further, only
a couple of process improvement measures have deteriorated, indicating that
the concrete benefits attained in terms of productivity and process improvement
last longer than softer measures such as culture building.

30
26
25

20

15
15 14

10
6
5

0
Productivity Process Improvement Culture Building Customer Centric
Improvement Metrics improvement

Fig. 9.2 Post-implementation improvements at the lean cluster


Reference 163

While the other operational measures such as productivity and customer-facing


metrics did improve/stayed at the implementation levels, from a sustainance stand-
point, the deterioration of CB forebodes a bleak future for this cluster. All the firms
in this cluster belong to either Stage II or Stage III of the Lean sustainability matu-
rity model presented earlier. The reasons for deterioration include loss of steam by
top management, lack continuity of workforce, not finding new applications for
Lean, etc. and agree with what we have presented in the previous sections.

9.5 Summary

In this chapter, we discuss the sustenance of Lean implementations at SMEs. We


present a maturity framework based on our direct experience with over a hundred
SME implementations across sectors and geographies. The reasons contributing to
either sustenance or deterioration are discussed. Finally, a study of findings from
Lean implementation at an SME cluster is discussed. We highlight the pitfalls at
different junctures of a Lean journey for an SME to caution them against and
prepare well for next level of improvement.

Reference

Sterman, J. D., Repenning, N. P., & Kofman, F. (1997). Unanticipated side effects of successful
quality programs: Exploring a paradox of organizational improvement. Management Science,
43(4), 503–521.
Beyond Lean
10

10.1 Introduction

Every organization with a long-term vision should aspire to excel in all aspects of
their business. The new millenium has given birth to alternate thought processes
and concepts, some of which complement Lean admirably and can help SMEs
move closer to excellence. Sustainability is the overarching umbrella under which
operational excellence tools such as Six Sigma, Theory of Constraints, Industry
4.0, and Time-driven Activity-Based Costing (TdABC) can be pursued by SMEs
(Fig. 10.1). Each of these tools is in alignment with Lean philosophy and thus can
reinforce the benefits accrued from Lean implementations at SMEs.
Each of these is a powerful, wide-reaching, and interrelated idea that can not
only significantly impact organization’s long-term performance, but also provide
competitive advantages previously unavailable. SME owners and managers might
find it useful to understand and integrate these concepts to turbo charge their Lean
implementation and uplift their business models.
In our experience, we see many businesses shying away from learning and
experimenting with these new concepts. Perhaps, a simplified explanation of these
ideas, along with suggestions on how to integrate into the Lean implementation,
could help in breaking down the mental barriers. Further, not all the ideas cost
a lot of money to implement. The ubiquity of technology, and cloud-based solu-
tions have helped break the economic barrier in adoption of many of these newer
concepts. We will provide a brief overview of these developments and place them
in the context of Lean management from an SME perspective in this chapter. In
the process, we explain how these developments might be integrated with existing
Lean implementation or help start Lean initiatives.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 165
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_10
166 10 Beyond Lean

Fig. 10.1 Schematic of


initiatives complementing
Lean

Sustainability

Theory of Constraints
Industry 4.0
Six Sigma

TdABC
10.2 Sustainability

10.2.1 Description

Sustainability refers to balanced growth that addresses the economic, environmen-


tal, and social goals of an enterprise. Until recently, sustainability was treated as
an additional expense that eats away the profitability of an enterprise, especially by
SMEs. This perspective is now questioned by all and sundry, due to irrefutable evi-
dences in terms of environmental and social disturbances across the globe caused
by unsustainable business practices. Many governments, quasi-government bodies,
and not-for-profit organizations have taken up the cause of sustainability and bring
awareness of the ill effects of unsustainable development to all the stakeholders.
Besides, there is also growing and substantial evidence that points out that the
long-term economic performance actually gets positively impacted for those orga-
nizations which take care of the environment and the society they operate in. For
example, in their study involving close to 250 SMEs, Sajan et al. (2017) have
established that Lean management practices enhance the overall sustainability in
terms of economic, social, and environmental performance of the organizations.
There are several other SME-specific studies covering sustainability, firm perfor-
mance, and other related factors using samples of SMEs from different nations.
Few such studies are summarized in Table 10.1.
10.2 Sustainability 167

Table 10.1 Various studies focusing on sustainability in SMEs


Researcher Source Topic Findings
López-Pérez 1000 Spanish Effect of sustainability • CSR, reputation, and financial
et al. (2018) SMEs on business outcomes value of the firms are higher for
sustainability-oriented SMEs
• This linkage is stronger for
family businesses than
non-family businesses
Masocha (2018) 208 South Effect of • Environmental sustainability
African SMEs environmental practices positively impact firm
sustainability on firm innovation, ecological, and
performance social performances
Battistella et al. 7 European Sustainable business • Economic, societal, and
(2018) tourism SMEs models environmental challenges affect
sustainable business model
(value proposition, value
creation, and value capture)
development in different sectors
of service SMEs
Pigosso et al. 108 Danish Potential for • Potential for eco-innovation
(2018) SMEs eco-innovation among SMEs increases with the
industrial symbiosis
opportunities and usage of green
business models
Schmidt et al. 16 SMEs in Adoption of • There is a need to strengthen the
(2018) Brazil sustainability practices adoption of practices such as
values and transparency, internal
audience, environment, supplier
customer, and community
relationships among SMEs
Jahanshahi and 40 SMEs in Role of top • Top management behaviour
Bre (2018) Iran management integrity increased the team
behaviour innovativeness and sustainability
implementation

In terms of contribution to UN Sustainable Development Goals (SDGs),1 SMEs


contribute to Decent Work and Economic Growth (SDG#8), Industry, Innova-
tion and Infrastructure (SDG#9), and Responsible Consumption and Production
(SDG#12). The UN environment program has charted a roadmap to aid SMEs
adapt resource efficiency practices through national technical assistance packages,
training and network activities, and implementation tools such as methodologies,
toolkits, guidelines, and standards.2 A recent report by World Economic Forum
using inputs from over 300 CEOs and founders of SMEs highlights that SMEs

1https://sdgs.un.org/goals.
2https://www.unep.org/regions/asia-and-pacific/regional-initiatives/supporting-resource-effici
ency/asia-pacific-roadmap-8.
168 10 Beyond Lean

have a potential to achieve sustainable growth, positive societal impact, and robust
adaptive capacity leveraging their size, network, people, and technology.3

10.2.2 Integration with Lean

The roots of Lean lay at frail and short-on-capital business and social condi-
tions prevailed at Japan post-Second World War. Thus, by design, Lean’s foremost
objective is to identify and avoid wastes in any form. Less waste implies more
sustainability.4 This can be achieved only when all the employees are committed
to the goal and work in empowered cross-functional teams. So, in a Lean orga-
nization, the fundamental building blocks required for sustainable development,
viz. eye for spotting wastes, customer centricity, respect for employees (celebrat-
ing team success rather than exploitation mindset), and long-term perspective (and
not getting drawn by short-term benefits that actually harm the long-term sustain-
ability) are already present. Key green enablers relevant to SMEs are life cycle
assessment (LCA), reduce, reuse, and recycle (3R), environment emission control,
and green procurement (Ahmad et al., 2020). Pursuit of these enablers feeds into
the essence of Lean implementation at SMEs and derives long-term sustainability
benefits.

• Life cycle assessment refers to the holistic assessment of the environmental foot-
print of the product right through design, production, usage, and disposal and
undertaking early measures that minimize such footprint. There are a number
of easy-to-use public databases and resources available5 that SMEs can use
to make a standardized assessment of the products and services they procure,
design, produce, or consume.6
• 3R (reduce, reuse, and recycle) is part of larger concept called circular econ-
omy. In linear economy, products are discarded as waste into landfills once the
usefulness of the product is perceived to be diminished. As against, in a cir-
cular economy, efforts are made to increase the economic life of products by
reducing their consumption, reusing by finding secondary uses for the product
reaching their end of life, and finally recycling either as newer products (reman-
ufacturing) or into constituent elements. Institutionalizing 3Rs into a Lean SME
saves energy, effort, material, and investments required for the given production
levels. Further, remanufacturing as a business idea itself has a big potential for
SMEs to undertake.
• Environment emission control refers to the practices that evaluate all the efflu-
ents (wastes) and by-products (solid, liquid, and gas) of a given firm for their

3 https://www3.weforum.org/docs/WEF_Future_Readiness_of_SMEs_2021.pdf.
4 https://www.sme.org/technologies/articles/2017/january/lean-green-boost-sustainability-with-
lean-manufacturing-principles/.
5 https://nexus.openlca.org/databases.
6 https://www.globallcadataaccess.org/.
10.3 Six Sigma 169

suitability to be released into environment, for selling downstream, or treating


them appropriately, if found unsuitable for releasing/selling.
• Green procurement refers to reducing supply chain carbon footprint by sourcing
environment-friendly raw materials and insisting for eco-labels for component
procurement. While SMEs may not be able to afford frequent supplier audits,
eco-labels/green stickers can help them decide which of the suppliers are more
sustainability oriented. Further, they themselves can begin the practice of eco-
labeling their products to draw customers with sustainable mind-set.

10.3 Six Sigma

10.3.1 Description

Six Sigma as a quality improvement methodology was pioneered by Motorola.


General Electric then expanded the scope of Six Sigma to include non-
manufacturing processes and has rolled the methodology across the enterprise.
GE is credited with the current popularity of Six Sigma as a five-phased pro-
cess improvement methodology: design, measure, analyse, improve, and control
(DMAIC). DMAIC is built on Deming’s Plan, Do, Check, and Act (PDCA) frame-
work for process improvement. Six Sigma is a highly rigorous problem-solving
tool grounded firmly in statistics and requires significant amount of data to be col-
lected for analysis. Like Lean, Six Sigma encourages unit experiments to establish
the process relationships. Six Sigma compares Voice of the Customer (specifica-
tions) with process capability for measuring the Defects Per Million Opportunities
(DPMO). A process is said to be performing at Six Sigma level, if there are no
more than 3.4 DPMO. Voluminous, highly repetitive processes are best suited to
benefit from Six Sigma implementations, provided all process data is measured
and documented.

10.3.2 Integration with Lean

Among all the initiatives listed in this chapter, Six Sigma is the one which is most
commonly used in conjunction with Lean. So much so that the phrase “Lean Six
Sigma” is now a fairly common terminology among operations excellence circles.
As noted in previous chapters, variation (Mura) is one of the main sources of
wastes. The main strength of Six Sigma methodology is the robust procedure to
capture and control process/output variations. And thus, the fitment of Six Sigma
into the execution stage (series of Lean projects to reduce/eliminate variation) of
the overall Lean methodology is as snug as a hand in glove. Therefore, Lean plays
the role of overall concept map and a philosophic vision and Six Sigma a powerful
execution tool to translate parts of the vision into action. However as cautioned
before, Six Sigma relies heavily on data and statistics and yields best results when
processes are fast and repetitive. SMEs should evaluate this requirement before
170 10 Beyond Lean

integrating Six Sigma into their Lean implementation. Please see below the caselet
of Avanti Feeds, where we used Lean Six Sigma to improve OEE.

OEE Improvement Through Lean Sigma at Avanti Feeds (AF)


A decade ago, the global seafood industry was just emerging out of a severe
disruption due to ravage by disease. AF, a pioneer in prawns and prawn
feed, was trying to get their feed mills back on track to supply to growing
demand. However, for several months output lagged well below the expected
levels. Lean experts were called in. Being a continuous process plant, our
first diagnostic was to compute of Overall Equipment Effectiveness, which
was low at 50%. The main culprit was the performance ratio, i.e. mill speeds
were only 60% of the rated speed. The reason was quality problems at the
customer ponds for the feed when produced at higher speeds. The complaint
was that the feed was floating on being thrown into the pond instead of
sinking and be available to the prawns at the bottom of the pond.
At this stage, it was clear that the only way to increase OEE significantly
was to solve this “floating” problem. Two key product quality parameters,
size and moisture content, were taken up as they in turn influence prod-
uct stability and floating. Regression analysis was done to relate process
parameters such as mill speed, mix moisture, and % of various raw materi-
als, with mill speed and product moisture. Using this analysis, control limits
were fixed for each of parameter and process control charts established at
the shop floor. Substantial variation was seen in the mix moisture and mill
temperature parameters and the process capability prevailed at 1.55 sigma.
A target of 3.5 sigma was fixed.
Root cause analysis was then carried out to identify reasons for these vari-
ations. A series of kaizen projects such as introducing error proofing, visual
controls on the steam valves, and workstation improvements at the water
addition stage helped in bringing down the variation in moisture content and
maintaining specified mix temperature before the critical die extrusion pro-
cess where the feed is actually formed into pellets. Mill speed was soon
increased back to 90% of rated with no negative impact on product quality.
OEE went up to 70%!!

10.4 Theory of Constraints

10.4.1 Description

Goldratt’s Theory of Constraints (ToC), as the name suggests, is based on a clear


definition, identification, and management of the constraint (also called a bottle-
neck) faced by a firm. A constraint is defined as anything (a machine, a worker, a
policy etc.) that limits the ability of the firm to increase its throughput. ToC offers
10.4 Theory of Constraints 171

an approach to identify bottlenecks that limit the output and, hence, a guideline to
prioritize the improvement initiatives. The ToC-based Drum-Buffer-Rope (DBR)
approach is used to schedule the bottleneck at its capacity, which defines the drum
beat of the factory. The buffer is placed before the constraint (constraint buffer)
and as well at the end of the line (customer buffer) to prevent the possibility of
starving bottleneck or missing customer orders. Lastly, the rope is the informa-
tion link between the level of buffer before the constraint buffer to the beginning
of the line which gets triggered based on the lead time. Thus, a DBR schedules
the bottleneck, the downstream production of bottleneck flows, and the upstream
production is pulled using the rope.
In most ways, ToC is almost identical to Lean value stream concepts. We said
previously that a pacemaker process in the value stream defines the rate of through-
put. A pacemaker is defined as the process in the value stream after which there
exists only continuous flow. The pacemaker pulls from the upstream processes
using standard WIP (akin to constraint buffers) and through the mechanism of
kanbans. In Lean, the pacemaker is the single point of scheduling and replaces the
machine-wise or section-wise schedules of traditional organizations. So the DBR
approach and the Lean value stream are quite similar.
A key difference is that Lean focuses and improves the value addition in each
process step, by eliminating waste, while ToC’s focus is in improving the through-
put especially at the bottleneck.7 Both approaches are wary of the conventional
accounting methods that mask the true costs, and contributions by processes,
and thus mislead managerial decision-making. Throughput accounting and Lean
accounting are proposed as part of these respective methodologies.

10.4.2 Integration with Lean

But for the minor differences in the approach of the two methods, viz. ToC and
Lean, as stated above, they should be considered complementary to one another,
rather than competing. ToC offers an excellent framework for enhancing through-
put by an effective constraint management, which is based on the following five
iterative steps:

• Identify the bottleneck,


• Exploit the bottleneck,
• Subordinate everything else to the bottleneck,
• Elevate the bottleneck, and
• Review the system for the new bottleneck and start the process over again.

7https://www.lean.org/the-lean-post/articles/what-is-the-theory-of-constraints-and-how-does-it-
compare-to-lean-thinking/.
172 10 Beyond Lean

Most importantly, the simplistic financial metric structure of ToC that relates
throughput, operating expenses, and investments could be appealing and easier
to follow by many SMEs as it gives a clear picture on the cost benefit of any
improvement and offers the SME a clear sight of financial gains. The constraint-
based thinking of ToC is also relatively easier to understand. Considering these
advantages, it would be highly beneficial for firms and consultants to judiciously
integrate the concepts and practices from ToC and Lean into their operations
excellence journey.

10.5 Industry 4.0

10.5.1 Description

At the heart of Industry 4.0 is a set of technological developments that are grouped
as cyber-physical systems (CPS). As the name suggests, CPS provides a 360°,
seamless, persistent, and Internet-enabled interface between sensor generated data
acquisition, analytics-based data processing, and robotics-based process control.
Increasing semiconductor-based electronics in various machinery, equipment, and
utilities enhanced the digitization of manufacturing and allied operations such as
transportation and storage. Further, increasing use of information systems such
as ERP, MES, and Web 2.0 to plan, execute, integrate, and monitor internal and
extenal business processes has resulted in digitalization of business and non-
business organizations alike. Both digitization and digitalization have become a
rich source of real-time data, which coupled with other emerging technologies
such as robotics, machine learning, Industrial Internet of Things (IIoT), analyt-
ics, and big data have made Industry 4.0 necessarily a new way of doing business.
Besides, the affordability of most of these technologies has contributed to the rapid
adoption and quicker payback of investments in Industry 4.0 even by SMEs.
According to BCG,8 firms can achieve close to 40% cost reductions when Lean
and Industry 4.0 are jointly implemented, as against approximately 15% from each
of these individual initiatives. One of the direct ways in which Industry 4.0 sup-
ports Lean is by drastically bringing down the gap between the time to sense and
the time to act, reducing the information lead time. Lean firms can enhance flex-
ibility, productivity, speed, quality, and safety by adapting Industry 4.0 powered
tools such as sensors, predictive analytics, and virtual reality. Specifically, a sam-
ple list of Lean tools and supporting Industry 4.0 technologies are mapped in Table
10.2.

8 https://www.bcg.com/publications/2017/lean-meets-industry-4.0.
10.5 Industry 4.0 173

Table 10.2 Lean tools and supporting technology solutions mapping


Lean tool Technology solution
SMED Robotics
Poka-yoke IoT devices, barcode scanners, pick-to-light,
camera
5S audits Visual management systems
Root cause analysis, continuous improvement Big data, analytics
Kanban Sensors at critical components
Standardization, training Virtual reality
TPM Augmented reality
Teamwork Collaboration tools

• Deploy data monitoring solutions at critical points to enable diagnosis


Diagnose

• Make digital technology / Low Cost Automation (LCA) plan within the Lean roadmap
Design

• Data needed for improvement projects can be collated from machines / lines / materials
Prepare • Appropriate devices for computing OEE / line stops /rejections can be introduced at this stage

• LCA for improvements such as Jidoka (line stop), error proofing to avoid defects
• Analysis and alerts for line stops, bottleneck identification, deviations
Improve • Monitoring and deviation alerts for process, machine conditions
• Material flow - kitting, pull based WIP management can all be electronically facilitated and discipline enforced

• Validating the results of improvements made and daily works management requires continuous monitoring of data
Standardize • Digital SOPs both for display as well as guidance, also for deviation monitoring

• Various assessment checklists and scores can be digitized and visually displayed
Sustain • Monitoring can help in continual improvement journey as data is now readily available for next round of diagnosis

Fig. 10.2 IIoT solutions for different stages of Lean implementation

10.5.2 Integration with Lean

Different stages of Lean implementation at various SMEs in our experience are


benefitted by Industry 4.0 (specifically IIoT) solutions as outlined in Fig. 10.2.
Powering the fundamental Lean principle of waste reduction by leveraging the
relevant technology in various maturity stages of SMEs should be the goal of inte-
grating Industry 4.0 with Lean, in other words, digitally enabled Lean.9 There

9 https://www.mckinsey.com/business-functions/operations/our-insights/industry-4-0-demyst
ified-leans-next-level.
174 10 Beyond Lean

are a number of case studies that detail successful integration of Industry 4.0
with Lean (Mrugalska & Wyrwicka, 2017). As with every other new technology,
implementation should always be carefully thought through and backed by a pilot
implementation that clearly captures the lessons learnt. What works for one indus-
try or even one firm does not automatically guarantee that such technology will be
successful for an other firm. Blind imitation without appropriate customizations is
one of the major reasons of failures of technology implementations.

Lean Industry 4.0 for 24 × 7 Process Performance at ICLEAN


We have been working with ICLEAN over a decade implementing Lean
in bits and pieces, extending the concepts in every successive intervention.
In 2017, we helped design a layout for their new integrated and expanded
facility with a 75% jump in output expected to cater to customer demands
from the pharmaceutical sector. The Lean layout was implemented in 2019,
and within a year, streamlined flow and standard practices were established.
At this point in time, two constraints were identified that could hinder the
plant performance.

1. Complete kit of items to be available against each customer project to


ensure delivery and installation at the customer site. It was quite normal
to find a shortage of a couple of panels at the FG stage resulting in delays
in dispatch and necessitating costly crisis-mode operations at the plant to
produce the missing panels.
2. The last process before packing is the “Infill” where insulation foam is
filled into the panels on large jigs that maintain shape during the exother-
mic expansion process of curing. This turned out to be a constraint as
most of the panel preparation and post-curing activities were manually
done. Delays in this process manifested as an ever-present inventory of
panels before the Infill process.

The company decided to go in for IoT-enabled solutions that can ensure


100% adherence to the defined Lean flow and process. A solution comprising
two parts was designed and implemented in 2021. First is a track and trace
system using barcodes on each panel. A barcode is placed on each panel
at the end of first process—sheet cutting, and on scanning—it records the
panel number that has been generated by PPC in the project-wise schedule.
This barcode is scanned at each stage up to dispatch, and information on
completion status of any customer order is always available online. If all the
panels are not complete at any stage, and an attempt is made to start the next
order, an alert is flagged.
The second is an OEE monitoring system at the Infill jigs. Each jig has a
fixed capacity based on panel size. The curing time depended on the thick-
ness of fill. If the jig is not opened after curing time is done or not closed
10.6 Time-Driven Activity-Based Costing (TdABC) 175

within a set time interval, a sensor reads this as a short stop and records the
loss time. The system is configured to send escalations to production and
plant managers as the loss time crosses certain thresholds.
Thus, by using IoT to complement already established Lean layout and
flow, ICLEAN hopes to avoid crisis-mode operations, and also achieve the
targeted output consistently.

10.6 Time-Driven Activity-Based Costing (TdABC)

10.6.1 Description

Traditional accounting accumulates costs department-wise and allocates overheads


for periodic reporting. Due to inherent inconsistencies in the overheads allocation,
the information from traditional accounting systems does not reflect the true cost
of business processes, leading to difficulties in accurately matching revenues with
expenses. This problem is exacerbated due to increasing mechanization of orga-
nizational processes that use shared resources and overheads rather than direct
labour. Process-based approaches such as Time-driven Activity-Based Costing
(TdABC) by Kaplan and Anderson (2007), Lean accounting (Maskell & Bag-
galey, 2006), and throughput accounting (DugDale & Jones, 1998) have been
proposed to overcome these limitations of traditional accounting. Of the three,
TdABC is well established and has more implementations than the latter two.
Given the GAAP reporting requirements, all these alternate accounting systems
typically implemented as add-ons to the existing systems, that draw on the con-
ventional accounting transactional data, and recast costs as shared resources are
consumed in revenue generation. Figure 10.3 provides a conceptual overview of
revenue generators and cost accumulation in a typical organization.
TdABC approach can effectively answer questions such as—Is this prod-
uct/service, order, customer, channel, or business profitable? If not, what does it
take to make it profitable (reduce service, increase price), how does it compare
with other businesses? etc. (Everaert et al., 2008)—given its accurate revenue and
expenditure matching.

10.6.2 Integration with Lean

Lean is built on the core philosophy of separating waste from value addition.
A process comprises multiple activities, and each activity at the unitary level is
either work or waste. Keeping with this, from Lean perspective, any expenditure
is either (legitimate) cost or waste.

Activity = Work + Muda


176 10 Beyond Lean
Revenue Generators

Businesses Channels Customers Orders* Products & Services

Cost Accumulation

Direct Material, Labor &


All other overheads &

Distribution, Channel

process / activities
Order Fulfillment

Plant Overheads
Customer Support,
period expenses

Sales, Marketing,
Promotion etc.,

After Sales etc.,


Logistics &

Fig. 10.3 Revenue generation and cost accumulation in organizations

Expenditure = Cost + Waste

Cost is the amount of expenses (actual/notional) incurred on or attributable to a


specified thing or activity. Cost is not by default “what is incurred”, in its entirety.
It should reflect only purposeful spending on value-added activities. Rest of the
expenditure are all “Waste”.
TdABC proponents recommend that it is better to be approximately right than
being precisely wrong. TdABC requires two parameters: the cost per time unit of
the activity and the time required to perform an activity (Somapa et al., 2011),
which are captured by direct observation or by interacting with process owners.
The VSM exercise by all the Lean organizations is a ready source of much of
the information that is needed for TdABC. Further, once the required interface
(between traditional accounting transaction database and the TdABC database)
programs are developed, organizations will have the advantage of evaluating the
performance of their value streams on a dynamic basis.
First among the difficulties in implementing TdABC is the effort involved in
recasting the general ledger financial data to the supporting and then to oper-
ating departments. Allocations are usually hard coded in the ERP systems, and
rewinding them requires some manual effort (Siguenza-Guzman et al., 2013), at
least the first time. Second challenge is to have the team members agree to be
observed or interviewed with consultants, which may not truly reflect the long-
term performance in terms of time equation coefficients due to behavioural biases
(Cardinaels & Labro, 2008). However, in a Lean intervention, the second challenge
References 177

is overcome as direct observation of the process is mandatory for making the VSM,
and most of the relevant data such as cycle time, manpower, and machine time is
being captured at this juncture. Lastly, some process variants may be so rare that
they may not occur during the observation period of the consultants, leading to
missing them altogether.

10.7 Summary

In this chapter, we presented some of the emerging business philosophies, meth-


ods, and technologies that have the potential to provide the requisite thrust to
Lean organizations to take their performance to the next levels. The functional ini-
tiatives, viz., Six Sigma, Theory of Constraints, Industry 4.0, and TdABC, support
operational performance, whereas an overarching sustainability initiative helps the
firm develop and maintain a long-term perspective. A brief overview of each of
these opportunity areas was provided, followed by some suggestions on how SMEs
can plug these methods into their existing Lean implementations.

References

Ahmad, S., Abdullah, A., & Talib, F., (2020). Lean-green performance management in Indian
SMEs: A novel perspective using the best-worst method approach. Benchmarking: An Inter-
national Journal.
Battistella, C., Cagnina, M. R., Cicero, L., & Preghenella, N. (2018). Sustainable business models
of SMEs: Challenges in yacht tourism sector. Sustainability, 10(10), 3437.
Cardinaels, E., & Labro, E. (2008). On the Determinants of measurement error in time driven
costing. The Accounting Review, 83(3), 735–756.
DugDale, D., & Jones, T. C. (1998). Throughput accounting: Transforming practices? The British
Accounting Review, 30(3), 203–220.
Everaert, P., Bruggeman, W., Sarens, G., Anderson, S. R., & Levant, Y. (2008). Cost modeling in
logistics using time-driven ABC. International Journal of Physical Distribution & Logistics
Management.
Jahanshahi, A. A., & Brem, A. (2017). Sustainability in SMEs: Top management teams behavioral
integration as source of innovativeness. Sustainability, 9(10), 1899.
Kaplan, R. S., Anderson, S. R. (2007). Time-driven activity-based costing: a simpler and more
powerful path to higher profits. Harvard Business Press.
López-Pérez, M. E., Melero-Polo, I., Vázquez-Carrasco, R., Cambra-Fierro, J. (2018). Sustainabil-
ity and business outcomes in the context of SMEs: Comparing family firms versus non-family
firms. Sustainability, 10(11), 4080.
Maskell, B. H., & Baggaley, B. L. (2006). Lean accounting: What’s it all about? Target, 22(1),
35–43.
Masocha, R. (2018). Does environmental sustainability impact innovation, ecological and social
measures of firm performance of SMEs? Evidence from South Africa. Sustainability, 10(11),
3855.
Mrugalska, B., & Wyrwicka, M. K. (2017). Towards lean production in industry 4.0. Procedia
Engineering, 182, 466–473.
Noto, G., & Cosenz, F. (2020). Introducing a strategic perspective in lean thinking applications
through system dynamics modelling: the dynamic value stream map. Business Process Man-
agement Journal.
Omogbai, O., & Salonitis, K. (2016). A lean assessment tool based on systems dynamics. Procedia
CIRP, 50, 106–111.
178 10 Beyond Lean

Sajan, M. P., Shalij, P. R., & Ramesh, A. (2017). Lean manufacturing practices in Indian man-
ufacturing SMEs and their effect on sustainability performance. Journal of Manufacturing
Technology Management.
Pigosso, D. C., Schmiegelow, A., & Andersen, M. M. (2018). Measuring the readiness of SMEs
for eco-innovation and industrial symbiosis: Development of a screening tool. Sustainability,
10(8), 2861.
Schmidt, F. C., Zanini, R. R., Korzenowski, A. L., Schmidt Junior, R., & Xavier do Nascimento,
K. B. (2018). Evaluation of sustainability practices in small and medium-manufacturing enter-
prises in Southern Brazil. Sustainability, 10(7), 2460.
Serrano, I., de Castro, R., & Goienetxea, A. (2009). Pacemaker, bottleneck and order decoupling
point in lean production systems. International Journal of Industrial Engineering: Theory,
Applications, and Practice, 16(4).
Siguenza-Guzman, L., Van den Abbeele, A., Vandewalle, J., Verhaaren, H., & Cattrysse, D. (2013).
Recent evolutions in costing systems: A literature review of time-driven activity-based costing.
Review of Business and Economic Literature, 58(1), 34–64.
Soltysova, Z., & Modrak, V. (2020). Challenges of the sharing economy for SMEs: A literature
review. Sustainability, 12(16), 6504.
Somapa, S., Cools, M., & Dullaert, W. (2011), The development of time driven activity based
costing models: A case study in a road transport and logistics company. In Current issues in
shipping, ports and logistics (pp. 431–445).
Leveraging Lean to Tackle
Uncertainty 11

11.1 Introduction

The pandemic of COVID-19 has heightened the importance of building resilient


organizations that can not only recover from the debilitating supply chain disrup-
tions quickly, but also leverage the crisis to their advantage.1 Even during the years
leading to the pandemic itself, there has been a steady rise in the disruptive events
globally.2 This predicament is expected to continue like this due to a myriad of
interlinked reasons such as environmental deterioration, networked economy, shift-
ing political forces, and shorter product life cycles, each of which is not only a
potential disruptive force, but also feeds others, potentially multiplying the overall
impact.
Some management thinkers believe that Lean deteriorates the organizations’
ability to respond to disruptions. This is only true, only if all that is done under
the umbrella of Lean is to drive out inefficiencies or redundancies (buffers such
as capacity, inventory, time) without due regard to underlying uncertainties. The
fact that the truly Lean firms such as Toyota continue to survive disruptions rather
graciously as against those firms which implemented Lean only as a mixed-bag of
tools or slogans, should tell us how misguided this judgement is. True Lean philos-
ophy is based on the principle that uncertainty/variability is the cause of buffers,
and the only way to truly (if at all) eliminate a buffer is by having a complete
knowledge (and therefore no uncertainty/variability). Since complete knowledge is
impossible, the next best thing is to create an organization that senses, processes,
and responds to change quickly—a learning organization. After all, in the first
place, Lean as a system was developed to deal with one of the worst disruptions

1 https://www.mckinsey.com/business-functions/risk-and-resilience/our-insights/the-resilience-

imperative-succeeding-in-uncertain-times#.
2 https://worlduncertaintyindex.com/.

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 179
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9_11
180 11 Leveraging Lean to Tackle Uncertainty

of past century, the ravaging effect of the Second World War on Japan especially
post the atomic bombs. And Lean or TPS proved its mettle as “THE” philos-
ophy to handle uncertainty during the next major global disruption—the 1973
oil crisis. There are always layers of assumptions/beliefs/past conditioning upon
which organizational buffers are based on. Whenever something happens to shake
these assumptions, Lean management takes notice of it, corrects its worldview, and
builds the necessary buffer. Therefore, Lean questions only just-in-case buffer and
not the redundancies deliberately designed into organizational value chains, as we
will see in some of the snippets in this chapter.
SMEs, most of which are at far end of supply chains, face the maximum brunt
of the uncertainties due to well-documented demand signal amplification effect
known as “The Bullwhip Effect” (Lee et al., 1997). SMEs therefore are a highly
vulnerable sector, who will be at a loss if their organizational design does not
help them tide over these uncertainties. Lean management principles aid building
flexibility into supply chain design and capabilities, a handy tool in times of unex-
pected events (Sheffi & Rice, 2005). Lean management helps SMEs to prepare for
shocks and adapt quickly to changing environment.
The chapter begins with an explanation of various terms related to uncertainty,
in order to bring clarity from among a bewildering array of jargon used in popu-
lar press and as well in our day-to-day conversations. We then look at examples
of some of the recent supply chain disruptions and their impacts. Finally, a few
examples of how organizations have effectively tackled uncertainty by leveraging
Lean are illustrated.

11.2 Definition of Terms

We classify various terms in vogue into two groups, viz. sources of disturbance and
the strategies adopted by firms. Disturbances could be on account of internal or
external situations presented to the organization. Strategies refer to the capabilities
the organizations develop in order to cope with the disturbances in a manner that
is favourable to the organization.

11.2.1 Sources of Disturbance

Various disturbances affecting organizations are presented in Fig. 11.1. The terms
are organized such that the degree of impact on the firm increases as one proceeds
outwards from the innermost circle.
We explain these terms from an operations perspective in the following
subsection.

11.2.1.1 Randomness
Randomness refers to the unpredictable variation in behaviour of every pro-
cess, equipment, or system. Truly random behaviour implies variation from
11.2 Definition of Terms 181

Fig. 11.1 Sources of


disturbance faced by firms
Disruption

Impact
Uncertainty

Variability

Randomness

mean/expected behaviour with some regularity, but without any discernible pat-
terns such as increasing/decreasing trends or seasonal/cyclical fluctuations. Ran-
domness is also called white noise, which is technically the residual component
in any data series, after all other components of variation are explained. With suf-
ficient past data, there are statistical techniques available to measure the extent of
randomness in a process and thereby recommend the achievable control limits.

11.2.1.2 Variability
The explainable component of variation in mean/expected behaviour over time
can be called as variability of the process. From this definition, one can expect
predictable patterns such as increasing/decreasing trends or seasonal/cyclical fluc-
tuations in the operations data such as demand, production quality, and raw
material quality. Presence of a pattern helps organizations perform root cause
analysis, so that it can either take corrective actions or be better prepared.

11.2.1.3 Uncertainty
Incomplete knowledge, understanding, and articulation (typically about future) are
referred to as uncertainty. While the term uncertainty implies both favourable and
unfavourable possibilities, we are concerned with unfavourable possibilities and
would like to cover from them. Uncertainty results from our inability to see future,
or understand/process the information objectively, or due to high complexity of the
system/environment.
Consequently, uncertainty can be reduced by reducing the lead time since keep-
ing the horizon short negates our inability to see the future to some extent. Lead
time is defined as the time required by the organization to get the product into the
182 11 Leveraging Lean to Tackle Uncertainty

hands of the customer from the time order is received in the make-to-order supply
chains, whereas it is defined as the number of days ahead of which different enti-
ties of the supply chain have to forecast in the make-to-stock supply chains. Lead
time can be reduced by improving the quality and timeliness of information collec-
tion, processing, and propagation within and through the departments and as well
to the entities outside the organization. Lastly, reduction in the complexity of sys-
tems and processes also significantly contributes to reduction of uncertainty. The
extent to which the lead time cannot be further reduced determines the present
irreducible uncertainty. However, with improvement in processes, understanding
of the environment, technology, and practices, organization can continue to keep
reducing their uncertainty.

11.2.1.4 Disruption
A completely unexpected and a rather sudden happening, such as a pandemic
breakout, Suez Canal blockade, and natural calamities or accidents, can be termed
as disruptions. There are always events happening across the world all the time,
which the erstwhile organizations were insulated from because the impact of such
events was localized and did not touch organizations’ network. However, with
increase in the interconnectedness of the world, either the supply, demand or both
ends of the organizations supply chains are increasingly getting impacted with such
happenings (Coe & Yeung, 2015). If the impact is on the supply end, it is called a
supply shock and a demand shock when the impact is on the demand end. A survey
of supply chain professionals conducted by World Economic Forum (WEF) found
that the top five sources of disruption are (1) reliance on oil, (2) non-availability
of shared data/information, (3) fragmentation along the value chain, (4) extensive
subcontracting, and (5) poor supplier visibility.3

11.2.1.5 Risk
Typically, individuals/organizations want to protect themselves against the
unfavourable events, and so efforts are made to quantify the expected impact of
these unfavourable events. Such a quantified measure is called risk. There are mea-
sures of risk pertinent to all the levels, viz. randomness, variability, uncertainty, and
disruptions. Popular risk measures are variance, standard deviation, value-at-risk,
etc.
A collection of practices that improve the preparedness of organizations to
meet uncertainty and disruptions is referred to as supply chain risk management.
Most common approach of risk management involves assessment of probability
of occurrence of an unfavourable event and its impact rating (on some nominal
scale such as 1 to 10, higher the rating, higher the impact). The product of these
two values is called as risk score, and organizations establish a threshold for such

3 https://www.weforum.org/agenda/2020/03/covid-19-coronavirus-lessons-past-supply-chain-dis
ruptions/.
11.2 Definition of Terms 183

risk scores. Any event with a risk score higher than the threshold is actively moni-
tored, and sensors /indicators of such events are put on a high alert mode. Alternate
plans are created that get triggered as soon as the symptoms are sensed. Organiza-
tions create a cadence to review and modify their risk assessment matrix including
addition/modification/deletion of events and adjustments to the probability/impact
values.

11.2.2 Strategies

Strategies that can be adopted by firms to tackle the source of disturbances are
presented in Fig. 11.2 in the order of additional investments by the firms. The
innermost circle presents the least investment, and the outermost, high investment
in building deliberate redundancies in capacity. It is to be noted that the sources of
disturbance (Fig. 11.1) and strategies (Fig. 11.2) do not necessarily correspond in a
one-to-one mapping, but some alignment exists. It is logical to improve capabilities
from the inner to outer circle, for higher-order investments stick well when lower-
order capabilities are firmly in place.
We explain these terms from an operation’s perspective in the following
subsection.

11.2.2.1 Reliable
A process is said to be reliable when it operates within the specification limits, i.e.
the capability of the process defined in terms of its upper and lower control limits
should be well within the customer specified upper and lower limits, respectively.

Fig. 11.2 Strategies to be


adopted by firms Resilient (High
investment)
Investment

Flexible / Agile

Responsive

Robust

Reliable
(Low
investment)
184 11 Leveraging Lean to Tackle Uncertainty

Various statistical quality control techniques including Six Sigma have been devel-
oped and implemented successfully to monitor process reliability. We encourage
the reader to refer any of the foundational statistics books on the subject matter
for additional details.

11.2.2.2 Robust
A process is said to be robust, when its outcome (signal) continues to be within
acceptable limits even when the environmental conditions (noise) change substan-
tially. Genichi Taguchi, a Japanese engineer, had developed statistical methods
which involve Design of Experiments (DoE), Taguchi Loss Function (TLF), etc.,
collectively called robust design methods. DoE helps the designer to develop an
optimal number of parameter settings, with which one can estimate the relation-
ship between inputs and outputs. TLF helps assign customer (dis)satisfaction value
to the extent of variation (deviation from expected value) in output. TLF value
rises slowly when the variation is small, but rapidly rises beyond a threshold. TLF
serves as a motivation to continuously endeavour to reduce the variation. Robust
engineering goes a step beyond reliability in the sense that it takes active control
of process design, so that the output stays within desirable levels.

11.2.2.3 Responsive
A firm is responsive, when it is able to meet the alternative delivery requests
from the standard delivery options offered by the company. The alternative deliv-
ery request can be in the form of product features, quantity, delivery lead time,
location, or lot size, etc. Ability to cater to such non-standard requests involves
deliberate cushion building in the organization’s supply chain such as signing up
with costlier suppliers, logistics service providers, keeping inventories at various
echelons, having quick access to spare capacity. Responsiveness is an effective
organizational strategy to meet uncertainty, if its products are innovative, enjoy
high margins and have short life cycles or fixed selling seasons (Fisher, 1997).

Zara: A Responsive Organization


Zara, a Spanish fashion retailer, is a case in point. It takes less than 6 weeks
for a new product from design stage to be positioned in its retail shelves.
Contrast this with the industry average of six months. Just like Toyota, every
process, infrastructure, and information technology investments, supplier and
distributor agreements are all tuned towards keeping this lead time as low
as possible, rendering Zara one of the most responsive supply chains. The
results are evident—in an industry where most of the players sell only 60%
of their production at full price, Zara is able to sell more than 85%, positively
impacting its profits.
11.3 Recent Supply Chain Disruptions 185

11.2.2.4 Flexible/Agile
Flexibility is a reactive capability, i.e. the organization is able to quickly change to
meet the changing environment, whereas agility refers to the organizations’ proac-
tive capability to capture the opportunity presented by the uncertainty faster than
its competitors. Organizations can build flexibility into their plans by regularly
undertaking a segmentation analysis of their products, component parts and creat-
ing an operations plan consisting of a mix of standard, optional, and rare products
right from design to delivery.4 The mixed model production capability recom-
mended in Lean management improves the organizational flexibility and agility
(Duggan, 2018; Takahashi et al., 2007). There are articles that place agility as a
strategic capability, under which flexibility is an operational capability (Abdelilah
et al., 2018; Lee, 2004). Flexibility and agility are structural capabilities, whereas
responsiveness refers to information and process capabilities.

11.2.2.5 Resilient
Despite of best of the systems and processes, disruptions are unavoidable. While
responsiveness, flexibility, and agility are all instruments to handle both uncertainty
and disruptions, resilience is an exclusive measure of the organizations’ ability to
quickly restore its normal functioning post-disruption (Ponomarov & Holcomb,
2009). Sheffi and Rice Jr (2005) popularized the notion of enterprise resilience
in their MIT Sloan Management Review paper. They recommend that the orga-
nizations should endeavour to achieve resilience by investing in flexibility rather
than in redundancies. Simchi-Levi and Simchi-Levi (2020) proposed supply chain
stress tests that evaluate the time to recover (TTR), time to survive (TTS), and
the performance impact (PI) as a measure of resilience for critical supplies. As
stated in the introduction section of this chapter, COVID-19 pandemic had sub-
stantially increased the necessity of organizational and supply chain resilience.
This resulted in the release of plethora of consulting and IT solution offerings
addressing resilience by providers. The solutions belong to two categories appli-
cable to both upstream and downstream of the enterprise, (1) usage of analytics
for better segmentation and risk classification of components and finished products
(Simchi-Levi et al., 2014) and (2) improving visibility through enhanced digital
connections.5

11.3 Recent Supply Chain Disruptions

Global uncertainty has been observed to increase steadily based on an index pub-
lished using text mining for word, “uncertainty”, and its equivalents on 143 country
reports published by Economic Intelligence Unit as seen in Fig. 11.3.

4 https://www.mckinsey.com/industries/advanced-electronics/our-insights/building-a-flexible-sup

ply-chain-in-low-volume-high-mix-industrials.
5 https://www.mckinsey.com/industries/advanced-electronics/our-insights/reimagining-industrial-

supply-chains.
186 11 Leveraging Lean to Tackle Uncertainty

60000

50000

40000

30000

20000

10000

0
1990q1
1990q4
1991q3
1992q2
1993q1
1993q4
1994q3
1995q2
1996q1
1996q4
1997q3
1998q2
1999q1
1999q4
2000q3
2001q2
2002q1
2002q4
2003q3
2004q2
2005q1
2005q4
2006q3
2007q2
2008q1
2008q4
2009q3
2010q2
2011q1
2011q4
2012q3
2013q2
2014q1
2014q4
2015q3
2016q2
2017q1
2017q4
2018q3
2019q2
2020q1
2020q4
Fig. 11.3 World uncertainty index (WUI) on the raise

While there are numerous disruptive events as seen in the above chart, to set
the context, we describe six examples of supply chain disruptions in the recent
past.6 These examples are representative of the lot in the sense they cover natural
disasters, accidents, a major sociopolitical disturbance, and a pandemic.

11.3.1 COVID-19

Since the first case of COVID-19 was identified in Wuhan, China, in December
2019, the pandemic took the world by storm. It broke out in waves in different
world regions, with every country experiencing two to three waves. Lack of an
effective vaccine forced countries to adopt social distancing, quarantining as the
only means to prevent the spread of the virus. This led to a series of lockdowns,
and countries sealed their borders avoiding goods and people movements. Even
within country, logistics was limited to essential goods during lockdown periods.
For globalized supply chains, this move was unexpected and devastating. Added to
the supply side challenges, due to extended work from home, the demand patterns
of number of product categories have changed substantially (e.g. SOHO equipment
demand shot up, and automobiles and component demand dropped). Companies,
small and big, stood up to the challenge, figured out alternate suppliers/materials,
changed products/production plans, implemented safe operating procedures, and
worked with their banks and supply chain partners to tide over the cash deficit
period.

6 https://www.aptean.com/en-EU/insights/blog/6-events-disrupted-maunfacturing-supply-chain.
11.3 Recent Supply Chain Disruptions 187

11.3.2 Semiconductor Manufacturing Factory Fire

On 19th March 2021 at 2:47 AM, a fire broke in the N3 Building (300 mm line),
Naka factory (Hitachinaka, Ibaraki Prefecture, Japan) of Renesas Semiconductor
Manufacturing Co., Ltd.7 An electric malfunction has resulted in a fire, which
was put off by 8:12 AM. The fire burnt down production equipment and damaged
the sensitive clean room. The company could resume its original production levels
only by 24 June 2021 resulting in production loss of approximately 100 days.8 The
company commands a market share of nearly 30% of worldwide semiconductor
sales to automobile industry. The loss of production of such a critical supplier was
a double blow to the automobile industry, which was already reeling with semicon-
ductor shortages induced by pandemic. A few companies (Nissan, Honda) had to
change their production plans completely to cope up with this supply disruption.9
The company exhibited an extraordinary alacrity in identifying alternate produc-
tion options, rapidly sourcing replacements to damaged equipment, and kept all
the stakeholders regularly updated about the status of recovery.10

11.3.3 Brexit

Britain’s exit from European Union (popularly called Brexit) began in 2016 and
culminated in the Trade and Cooperation Agreement (TCA) between both the
parties on 24 December 2020.11 This agreement finally brings to rest months of
uncertainty for supply chain partners on either side of the English Channel. The
agreement also brings clarity on how taxes will be applied on goods based on the
country of origin, requiring firms to operate transparent supply chains.12 One of
the most evident short-term impacts of Brexit has been the increased lead time and
transportation (includes clearance) costs between UK and EU, requiring firms to
overhaul their supply chain planning, execution, and monitoring systems.13 EU is
the largest supplier and customer base for UK enterprises. Considering the deep
interconnectedness of trade between UK and EU for several major industries,
McKinsey Consulting recommends that firms should (1) redefine their sourcing
strategy, (2) revise their supply chain footprint, (3) review inventory build-up

7 https://www.renesas.com/us/en/about/press-room/notice-regarding-semiconductor-manufactu

ring-factory-naka-factory-fire-summary-updates.
8 https://www.renesas.com/us/en/about/press-room/update-10-notice-regarding-semiconductor-

manufacturing-factory-naka-factory-fire-production-capacity.
9 https://www.reuters.com/business/autos-transportation/renesas-says-plans-restore-full-produc

tion-fire-damaged-chip-plant-by-end-may-2021-04-19/.
10 https://www.renesas.com/us/en/about/press-room/update-6-notice-regarding-semiconductor-

manufacturing-factory-naka-factory-fire.
11 https://www.chrobinson.com/blog/the-direct-impact-of-the-brexit-deal-on-supply-chains/.
12 https://www.pinsentmasons.com/out-law/analysis/brexit-deal-supply-chains.
13 https://www2.deloitte.com/nl/nl/pages/tax/solutions/the-impact-of-brexit-on-your-supply-

chain.html.
188 11 Leveraging Lean to Tackle Uncertainty

strategy, (4) prepare for changes in demand, (5) adjust product portfolio, and (6)
strengthen capabilities and talent.14 We find these recommendations are useful to
manage all supply chain disruptions.

11.3.4 Suez Canal Blockade

Containerization of cargo is perhaps the single most contributor to huge increase in


worldwide logistics. The growth in cross-border logistics led the shipping industry
to build bigger and bigger vessels. The biggest among the class of container ships
is called as Ultra Large Container Vessels (ULCV), which can carry more than
20,000 TEU containers.15 The ULCVs can cross Suez and Panama canals. Suez
Canal is the main artery of global logistics with nearly 50 ships crossing it in a day,
contributing to nearly 12% of global trade.16 Suez Canal is the shortest shipping
route between Europe and Asia. When the “Ever Given”, a 20,000 TEU ULCV,
ran aground due to strong headwinds on 23 March 2021, it effectively choked this
artery. A 24 × 7 recovery effort by a number of experts resulted in clearing of the
ship six days later, after a trade loss of nearly $9 billion every day.17 Shippers
used the alternative route around Africa via Cape of Good Hope with an increase
in transit time by 2 weeks.18

11.3.5 Drought in Taiwan

Semiconductor chips are ubiquitous. They are there practically in every consumer
durable product around us. With the increased drive towards smart “everything”,
the usage of semiconductors is going to keep increasing. Manufacturing of semi-
conductor chips involves serially binding many layers of silicon wafers one on
another to create Integrated Circuits (IC). Given the miniaturized manufactur-
ing with high precision, the raw materials, WIP and the finished chips need
to be thoroughly cleaned of any residuals. This cleaning is done by what is
called as Ultra-Pure Water (UPW). A single IC made on a 30 cm wafer may
require about 2200 gallons of UPW to get desired levels of cleanliness. Further,
deionizing approximately 1500 gallons of regular water produces 1000 gallons of

14 https://www.mckinsey.com/featured-insights/europe/brexit-the-bigger-picture-rethinking-sup
ply-chains-in-a-time-of-uncertainty.
15 https://en.wikipedia.org/wiki/Container_ship.
16 https://www.cnbc.com/2021/03/29/suez-canal-is-moving-but-the-supply-chain-impact-could-

last-months.html.
17 https://lloydslist.maritimeintelligence.informa.com/LL1136246/Suez-blockage-extends-as-sal

vors-fail-to-free-Ever-Given.
18 https://www.bbc.com/news/world-middle-east-56538653.
11.3 Recent Supply Chain Disruptions 189

UPW.19 Additionally, water is required for running the plant for purposes such
as AC cooling towers. All these make semiconductor industry one of the largest
industrial dependents of water. Normally, Taiwan is a country with abundant rain-
fall/typhoons. However, in 2020, it faced a worst drought in 56 years. Taiwan
Semiconductor Manufacturing Company (TSMC) is the world’s largest semicon-
ductor manufacturer, single largest contributor to Taiwan GDP, a user of about
150,000 tons of water per day, was diverted with much of the water available in
the country’s reservoirs, at the cost of other sectors, notably agriculture.20 Regard-
less, a worldwide semiconductor crisis ensued that may take a long time to recover
from, as setting up of alternate production facilities is at least a 3- to 4-years-long
capital-intensive project. Serious efforts are underway to increase the water recy-
cling ratio to reduce water consumption in semi-conductor manufacturing by many
researchers around the world.

11.3.6 The Texas Freeze

Texas suffered one of the worst winters in 2021, brought about by three successive
sever winter storms that swept the Americas during February 2021.21 So harsh
was the impact that the power grid failed, resulting in power outage lasting for
days in the state. Texas is home to a number of industries such as semiconductor
manufacturing, oil and gas, petrochemicals, food and agriculture, and aerospace
and defence, besides being a transportation hub.22 Samsung, Infineon, and Texas
Instruments are some of the large semiconductor manufactures in the state that
were affected by the outage, exacerbating the already delicate semiconductor sit-
uation.23 The industries using the by-products of petroleum refineries such as
plastics, adhesives, and chemicals were also significantly affected. Total economic
losses from this disaster were estimated to be $130 billion (Busby et al., 2021).
The catastrophe is yet another example of how one event in one place causes ripple
effects in different supply chains.

19 https://www.chinawaterrisk.org/resources/analysis-reviews/8-things-you-should-know-about-
water-and-semiconductors/.
20 https://www.forbes.com/sites/emanuelabarbiroglio/2021/05/31/no-water-no-microchips-what-

is-happening-in-taiwan/?sh=368aa62d22af.
21 https://en.wikipedia.org/wiki/2021_Texas_power_crisis.
22 https://www.ismworld.org/supply-management-news-and-reports/news-publications/inside-sup

ply-management-magazine/blog/2021/2021-03/the-texas-freeze-repercussions-and-risk-mitiga
tion/.
23 https://www.forbes.com/sites/willyshih/2021/02/19/severe-winter-weather-in-texas-will-imp

act-many-supply-chains-beyond-chips/?sh=1b10fbc2358a.
190 11 Leveraging Lean to Tackle Uncertainty

11.4 Lean: Counter to Uncertainty

Where Lean differs in philosophy vis-a-vis most management approaches is


to understand the sources of variability and initiate efforts to
reduce/eliminate/manage these variations. While this approach is more fundamen-
tal, active and involving in nature, passive approaches that take variability, lead
time as given are more common, perhaps due to the ease they promise in imple-
mentations. And yet executives are taken aback, when an unknown/unexpected
variation occurs, and their current set of tools is not equipped to handle the
situation. As noted by Spearman and Hopp (2020) in their case for unified
operations science, time as a buffer is interrelated with inventory and capacity
choices. Conventionally, organizations countered variation with high lead time
(time buffer), or inventories or capacity buffers. Ford and others managed by
cutting the demand side variation by choosing to serve a narrower and stable slice
of demand, and Sloan and other mass producers managed by creating a massive
capacity and inventory investments. These approaches are increasingly getting
untenable with the fleeting nature of today’s market demand for most products.
So, time is also a response measure that can define capturable demand.
In this context, we highlight a few examples how Toyota (Iyer et.al 2009) pre-
pare themselves for disruptions, and upon the occurrence of disruption, measures
they take to rebound. While most of these disruptions occur suddenly, yet there
are also real disruptions that are slow, such as environmental or technological.

11.4.1 Fire at Aisin Seiki Plant (1997)

When the lines producing p-valve, a machined product, at one of the supplier
plants, located at Aisin Seiki was gutted down on 1 February 1997, any other
organization would have taken weeks, if not months to put together a plan-B, but
not Toyota. Within hours, some 62 suppliers responded to calls by Toyota to pro-
duce one or more of the many types of p-valves that were needed by Toyota in the
months to come. P-valve being a standard part, Toyota maintained only 2–3 days
of production worth inventory in the system, implying that the entire production
not just that of Toyota, but also that of suppliers who depend on Toyota’s orders
were to come to a grinding halt, unless the situation is not addressed immediately.
Decisions were made quickly on which supplier will make which type of the valve,
and the corresponding manufacturing designs were faxed by next day. Within a day
or two, prototypes of the assigned valves started arriving at Toyota for approval,
which were mostly given the same day. Standardized processes as recommended
by Lean management are the most important reason as to how these companies
were able to design and produce completely new parts. By 8 February, volume
production of p-valves has commenced at about 50% of the responded suppliers.
Key takeaways from this incident for SMEs would be that (1) it helps if they are
part of an ecosystem, instead of being solo, (2) respond very quickly to any crisis
calls, and put together a response team, and (3) stay with the Lean management
process to steadily gain lost productivity.
11.5 Summary 191

11.4.2 Strike at the US West Coast Ports (2002)

Unlike the previous fire example, trouble was brewing for a while between the
US West Coast port union workers and the ports from early 2002, which finally
resulted in a lockdown of ports starting from 29 September 2002. Sensing this
challenge, Toyota took the following measures to keep its North American plants
up and running despite the lockdown: (1) built inventory for the imported parts
through the period leading up to lockdown, (2) book air cargo capacity ahead, and
(3) kept the option of re-routing ocean cargo alternate ports such as Mexico and
Canada. This example clearly illustrates the willingness to relax Lean norms such
as low inventory or frequent shipments in times of crises. Further, a hallmark of a
Lean-oriented firm is that it always has its ears to the ground and acts quickly.

11.4.3 Manufacturing Problems at Freescale (2005)

Freescale, a manufacturer of semiconductor chips, was having capacity problems


at its France factory during 2005. Semiconductor capacity enhancement takes years
to set up. Automotive boom was at its peak during 2005, implying that the chips
shortage could potentially dampen the overall sales. In addition to placing its own
team at the supplier’s premises to help them improve productivity and resort-
ing to air shipments to reduce pipeline inventory, Toyota this time realigned its
production mix such that the supply of chips to fast selling models was priori-
tized. This is an example of how a leading Lean organization shifts its managerial
focus depending on shifting bottlenecks and change the practices as the situation
demands.
Many other large firms such as Walmart, South West Airlines, and Seven-Eleven
have implemented some or all principles of Lean management successfully and
hence better prepared to meet uncertainty.24

11.5 Summary

In this chapter, we clarify various terms related to uncertainty and the strategies
adopted by firms to deal with the same. We also indicate a logical order for these
terms and also a recommended sequence of implementation of various strategies.
We looked at a few devastating supply chain disruptions of the past decade. Finally,
we saw how Toyota, the pioneer of Lean management tackled uncertainty, by
deliberately and knowingly relaxing its Lean norms. Such a relaxation holds good
for only as long as the danger of uncertainty is clear and present. Once the event
passes, the buffers will be gradually removed.

24 https://iveybusinessjournal.com/publication/going-lean-as-a-solution-for-navigating-uncert
ainty-and-a-crisis/.
192 11 Leveraging Lean to Tackle Uncertainty

With all positive public press, it might be tempting for the reader to imagine
that Toyota’s Lean journey was a bed of roses. On the contrary, the company
met with obstacles/challenges at every turn of this arduous but very rewarding
journey. And some very serious wake-up calls too, notable among these are the
Japan Earthquake (2010) and the accelerator pedal-related recall in North Americas
(2010), both resulted in this great company to pause and reflect. Toyota, in fact,
has a name for this phase of Lean implementation (hansei: pause and reflect). The
learnings from hansei from every Kaizen event make the organization more aware
of its own capabilities, and the environment it operates in. The essence of removal
of wastes by a Lean organization improves its visibility by that much more than its
non-Lean peers. With this improved visibility, a Lean firm can assess the impact
of disruption, look at alternate options, and trigger them as appropriate, better than
its counterparts.

References

Abdelilah, B., El Korchi, A., & Balambo, M. A. (2018). Flexibility and agility: evolution and
relationship. Journal of Manufacturing Technology Management.
Busby, J. W., Baker, K., Bazilian, M. D., Gilbert, A. Q., Grubert, E., Rai, V., Rhodes, J. D., Shidore,
S., Smith, C. A., & Webber, M. E. (2021). Cascading risks: Understanding the 2021 winter
blackout in Texas. Energy Research & Social Science, 77, 102106.
Coe, N. M., & Yeung, H. W. C. (2015). Global production networks: Theorizing economic devel-
opment in an interconnected world. Oxford University Press.
Duggan, K. J. (2018). Creating mixed model value streams: Practical lean techniques for building
to demand. CRC Press.
Fisher, M. L. (1997). What is the right supply chain for your product? Harvard Business Review,
75, 105–117.
Iyer et. al, A. V. (2009). Toyota: Supply chain management. MGH.
Lee, H. L. (2004). The triple—A supply chain. Harvard Business Review, 82(10), 102–113.
Lee, H. L., Padmanabhan, V., & Whang, S. (1997). The bullwhip effect in supply chains. Sloan
Management Review, 38, 93–102.
Ponomarov, S. Y., Holcomb, M. C. (2009). Understanding the concept of supply chain resilience.
The International Journal of Logistics Management.
Sheffi, Y., & Rice, J. B., Jr. (2005). A supply chain view of the resilient enterprise. MIT Sloan
Management Review, 47(1), 41.
Simchi-Levi, D., Schmidt, W., & Wei, Y. (2014). From superstorms to factory fires: Managing
unpredictable supply chain disruptions. Harvard Business Review, 92(1–2), 96–101.
Simchi-Levi, D., & Simchi-Levi, E. (2020). We need a stress test for critical supply chains. Har-
vard Business Review, 28.
Spearman, M. L., & Hopp, W. J. (2021). The case for a unified science of operations. Production
and Operations Management, 30(3), 802–814.
Takahashi, K., Yokoyama, K., & Morikawa, K. (2007). Integrating lean and agile strategies into
the production control system for mixed-model production lines. In: Advances in Production
Management Systems (pp. 405–412). Springer, Boston, MA.
Case Study 1: Seeds of Growth

Background

PAN Seeds (PAN) has established a strong presence in the rice dominated eastern
belt of India. The main seed processing plants are located in West Bengal, the
largest rice-producing state in India, with an estimated production of 15 million
tonnes per year. The company commenced operations in a small way in a few
decades back and subsequently expanded both its field production and seed pro-
cessing to multiple plant locations. PAN sells both basic and premium varieties
of paddy (rice) seeds catering to farmers in West Bengal and its neighbouring
rice-producing states.
Mr. Alok Marodia, the founder, has handed over the day-to-day sales and oper-
ations functions to his two sons and spends his time on Research and Development
and other organization building activities. Anshuman (the elder son) takes care of
operations and is always on the lookout for ways and means to improve opera-
tional performance. With their progressive mindset, PAN has already implemented
and is regularly using an ERP for all its functional transactions. In addition,
the management has set up a system of measuring, tracking and analysing per-
formance on select key performance indicators (KPIs) on Google spreadsheets.
Weekly review cadence with functional heads aid performance monitoring and
timely decision-making.
An invite to a one-day seminar on Lean management sparked Anshuman’s
curiosity. Though he could not attend, he received the presentation material from
the seminar and went through it. Impressed at the operational improvements
reported at various case study companies, he called for a meeting with the Lean
Sensei (Ganesh, one of the co-authors) from the seminar.
The two met up in person in February 2020, in which Anshuman introduced
himself and PAN and asked Ganesh “What can you do for my company”? Ganesh
laid out a general approach to implementing Lean management in a seed process-
ing plant. However, he said that a more specific approach could be presented only
after a diagnostic visit and an interaction with the key people at the plant. The

© The Editor(s) (if applicable) and The Author(s), under exclusive license 193
to Springer Nature Singapore Pte Ltd. 2023
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9
194 Case Study 1: Seeds of Growth

first week of April’20 was decided as ideal for this visit, as plant operations for
Kharif season would be in full swing (See Fig C1.1). In most parts of India, the
monsoon rains arrive in June or July and farmers sow their major crop during this
period, called Kharif . Post harvesting, a second smaller crop is sown in the Rabi
season that occurs in the winter months. April would therefore be the best time
for observation and assessment of current operations. And then COVID struck the
world, resulting in massive lockdown and travel bans. By the time the restrictions
were removed, Kharif season was over. Persistent that Anshuman is, he ensured
that the initiative was back on track by July’20.

The Seed Industry

Lean is a universal philosophy but for successful application of its concepts one
needs to understand the features of the industry. Seeds are an agricultural com-
modity and as such of a highly seasonal nature. The seed industry is quite unique
in many respects as against regular manufacturing in the sense that different pro-
duction and distribution processes must happen at specific times of the year as
shown in Fig. C1.1.
It therefore becomes critical that all the resources are aligned such that the
narrow window of opportunity is not lost. The key aspects to be considered when
designing Lean improvement initiatives in this industry were:

• High peak load coupled with a short window of opportunity of planting season
implies entire downstream plant output needs to be made available during two
to three months of Kharif season and one to two months of Rabi season.
• There are no permanent workers. Seasonal workforce come to work in the plant
only in the operating months while rest of the year they work elsewhere either
on the fields or doing other jobs.
• The plant executives and supervisory staff are permanent employees but without
active engagement for several months of the year. This fixed cost is unavoidable,
and to keep it to a minimum, employees are hired from nearby villages and
towns, are relatively less qualified, and with almost no exposure to industrial
set-ups.
• Machines run for two months, twice a year with a four months’ gap between
the seasons. The long idling of plant machinery with sudden spurts of usage is
again a challenge from the maintenance of equipment performance.

Every single characteristics of the industry seem to go against every known notion
of where would Lean fit. It is with this background that Ganesh had to view Anshu-
man’s question “What can you do for my company” which in his mind translated
to “What should be our approach to implement Lean in this industry”.
Case Study 1: Seeds of Growth 195

Nov-Dec
Seed harvesting, Reducing
moisture in seed through
natural drying (Sunning)
and storage for Kharif
requirement

Oct-Nov
Seed grading, Jan-Feb
processing, packing
Storage of seed
& dispatch for short
Rabi season

Jul-Sep
Seed sowing in fields, Unsold Mar-Apr
seed return to plant - Seed grading, processing
repacking & storage for use in and packing
Rabi Season

Apr-Jun
Dispatch to market, farmers
plant as monsoon sets in. Seed
harvesting, Sunning & storage
for Rabi season

Fig. C1.1 Annual cycle of seed process

Approach to Lean Initiative

Given the unique characteristics of this industry, there were understandably


doubts on the applicability and usefulness of Lean management. Specifically, the
questions that needed to be answered while designing the approach were:

• How can Lean thinking be applied in an industry where there is no production


for eight months and peak production during the remaining four months?
• Standards are a key foundation for improvement—where there is no fixed work-
force, how will we get them to standardize? How will they take ownership of
the process?
• Inventory is the biggest Muda as per Lean, but here storing of seed is the norm?
• Even if we do manage to design the initiative and implement Lean concepts,
will there be any significant tangible business benefit?
196 Case Study 1: Seeds of Growth

• Bring in the mindset of Lean thinking


• Learn by doing
Awareness

• Make equipment "Breakdown free" to maximize output during the season by initiating
Equipment Autonomous Maintenance (AM) during off season
Reliability

• Rapid improvement of grading and packing line within one week of start of season
• Observe, improve and set standards for all the other processes as and when they take place during
Process the year
Improvement

• Put in place spefic metrics for monitoring operational performance


• Create sufficient data for analysis and provide inputs for further improvement during the next
season
Validation • Fortnightly review by Anshuman

Fig. C1.2 Broad roadmap for Lean

Lean involves looking at processes through the philosophy of “Doing more with
less”. Tools and techniques can always be adapted. Keeping in mind the above
factors, it was decided to dive straight into the implementation rather than fol-
low the conventional diagnosis, roadmap, and implementation approach. With the
understanding of the nature of operations, a four-stage roadmap was formulated at
the outset to implement Lean at PAN Seeds as shown in Fig. C1.2.
The largest plant of PAN Seeds was located at Bardhaman (West Bengal) and
contributed about 75% of the seed output. Two more plants located in other
districts of West Bengal contribute the remainder. In order to disseminate Lean
thinking across all the plant locations, it was decided to involve key people from
the other locations in focused improvement workshops to be conducted at Bard-
haman. The plant managers of the other two plants would then implement the
learnings in their respective plants. Anshuman would review the progress in all
the plants during his fortnightly visits. The organization structure for the plants is
depicted in Fig. C1.3.

Stage 1: Initiating Lean Thinking

With the four-stage approach decided, the plant teams gathered for a three-day
quick improvements workshop led by Ganesh. The goal of this initial workshop
was to bring in Lean thinking through working on a few quick improvement
projects. Anshuman decided to participate himself as a team member to not only
understand the Lean philosophy but also to highlight the importance being placed
Terminology Description

Plant Manager
Godown – a shed that includes plant machinery and storage facilities.
Bardhaman has 14 godowns of varying production capacities (Tons
per hour) with different levels of automation, and storage (Tons)
Case Study 1: Seeds of Growth

Senior Supervisor
Quality Control Logistics Administration
(3)
Senior Supervisor – manages operations of multiple godowns. There
are three supervisors.

Godown
Supervisor (1 per Godown Supervisor – in-charge of the operations of one godown
Godown)

Godown Boy – runs the grading and processing plant (a.k.a operator)
Contract Labour Godown Boy (1
Gangs per Godown)
Labour Gangs– contracted seasonal workers who load and unload
trucks, dump the seed in the grading plant and run the packing lines.
They also manage the Sunning.

Fig. C1.3 Organization structure—Bardhaman plant


197
198 Case Study 1: Seeds of Growth

by PAN’s management on this initiative. The participants included plant in-charges


and senior supervisors from all three plants.
The Kharif season having concluded in May, only a few manual packing activ-
ities were in progress at the time of the workshop. In one Godown, mustard seeds,
a minor product, were being packed and dispatched. In another of the Godowns,
unsold market returns were being repacked into gunny bags for storage. Vegetable
seed packing was in progress in another plant location at Fatehpur. The rest of the
plant locations were in complete shutdown. (Refer Fig. C1.1).

Day 1—Awareness
Initial orientation was provided to the team on core Lean concepts such as
customer focus, internal supplier–customer relationships, and how to observe a
process through the eyes of the customer. Value addition and the wastes known as
Muda, Muri, and Mura (3 M) were discussed. Three cross-functional and cross-
plant teams were then formed from among the participants. Each team then went
to observe one of the three packing processes under the Lean paradigms. Teams
shared their observations in the end of day session.

Day 2—Understanding
The teams were encouraged to think about “why” the observed wastes happen.
A second round of detailed observations was done to understand the Whys. In
addition, concepts of cycle time measurement and productivity calculations were
introduced and applied to the running processes. Ganesh then challenged each team
to increase productivity by 50% in these simple manual packing processes.

Day 3—Excitement
Teams took up the challenge and applied the concept of waste elimination and
immediate countermeasures to improve flow. Study and fine-tuning of line imbal-
ances after establishing basic flow involved adjusting worker positions and number
of workstations. By evening—a huge result! All three teams reported in excess of
50% increase in productivity and hourly line output after making improvements.
Mr. Alok Marodia, the MD, had been invited for the closing session and presen-
tation by the teams and was pleased with the enthusiasm of team members and
excited by the results shown.
This three-day action-oriented learning workshop served to motivate the key
people, enabling them to gain confidence through their actions and brought in
the faith that Lean works. More importantly, it created the enthusiasm among the
participants to pick up the later stages of Lean implementation with jest.

Stage 2: Making Equipment Reliable

Preserving the germination properties of seed is critical to productivity of the farm-


ers (customer). Hence, the grading, processing, and packing is done just when seed
is needed in the market. This means a narrow 45–60-day window for completing
Case Study 1: Seeds of Growth 199

Table C1.1 OEE computation


Regular OEE computation formula Rough OEE Computation Formulae used at
PAN
OEE is a measure of the percentage of total Daily plant stop—start time = Available Hours
available time that a machine spends in (AH)
actually adding value to the customer. It is Grading (supplier rated) capacity = G
computed by removing the six major losses Theoretical Output in Available Hours (TO) =
from the total available time, categorized as G x AH
below: Raw seed dumped (charged) = RM
Availability: total time minus breakdowns and Grading Loss (undersize, husk, foreign body)
changeovers =L
Performance: available time minus short stop Actual Plant Output (AO) = RM–L
and speed losses Overall Equipment Effectiveness (OEE) =
Quality: produced quantity less rejection and AO/TO
yield losses
OEE: availability % × performance % ×
quality %

the plant activities in the main season. Any time lost due to plant breakdowns
implies poor product availability in some markets. In Lean parlance, this translates
to a focus on plant Overall Equipment Effectiveness (OEE). However, as is often
the case in most SMEs, the data on downtime, short stops, plant performance, and
changeovers was not available. The daily data captured in a large register did at
least record plant start and stop time and the plant output for each day.
The bias of Lean as we saw earlier is towards action. While accurate OEE data
on availability, performance, and quality definitely helps in focusing efforts, the
team did not want this to be a roadblock in the pursuit of improvement. Hence,
the OEE was computed using a rough formula as against the regular computational
method as shown in Table C1.1.
This came to approximately 70% for a few high capacity Godowns with semi-
automated packing lines, providing a lot of scope to reach the benchmark level
of 85% and above. Most Godowns are dedicated to one or two seed varieties
and therefore experience very few changeovers, if any. Hence, the team’s focus
was on avoiding breakdowns and minimizing short stops. As the plant was under
shutdown, it was a good time to kick off autonomous maintenance (AM) activities.
Initiated in August’20, AM was developed, implemented, and practised through till
the end of the next Kharif season in May’21.
The primary purpose of autonomous maintenance is to have the operators
develop ownership for their own machines. This is achieved by training them to
identify machine abnormalities, rectify (or get rectified) these abnormalities so as
to minimize equipment related problems. As we saw, there are no full-time per-
manent operators in PAN. But it was still essential to have some form of AM in
place. The first phase involved training the supervisors and Godown boys. In the
Rabi season they personally took ownership for the daily CLRI activities, filling
up the abnormality registers and corrective actions.
200 Case Study 1: Seeds of Growth

• AM Step One initiated


August 20 • Observation, tagging and rectification of abnormalities
• Unfulfilled basic condition and hard to access were also focused upon

• Second round of abnormality identification- practice makes perfect


• Development of draft CLRI standards
Sep '20 • Dry run practice of CLRI on shut machines

• Orientation of seasonal workers to develop sense of ownership of machines


• Observe, identify sources of contamination (seed dust, spillages) and take immediate countermeasures
within week of plant restart
Oct-Nov'20 • CLRI followed every evening on shutting down the plant for the day
Rabi season • Senior supervisor, godown supervisor and godown boy doing CLRI

• Why why analysis of breakdowns recorded in rabi season


• Implement permanent solutions for dust an spillages
• CLRI visual identification completed on machines
Dec-Feb '21 • Revise draft CLRI standards

• CLRI training on the job to labour gang


• Daily CLRI on all machines involving all workers
Mar-May'21
• Recording of breakdowns and abnormalities in register for further analysis
Main season

Fig. C1.4 AM implementation schedule

In the next phase, the labour gangs who helped in the grading operations and
ran packing lines were trained on the job on CLRI. In the Kharif season, daily
cleaning and following CLRI was done by entire workforce, and this also enabled
slashing the time needed for this activity.
The steps and the corresponding timelines involved in AM implementation are
shown in Fig. C1.4.
By end of the May season, the plant teams had embraced CLRI whole heartedly.
Every evening, the team members religiously clean and inspect their respective
machines. In fact, when Anshuman asked the team at the end of a year of Lean
implementation on which aspect they learned from the most, the unanimous answer
was “CLRI”.
In June, data was compiled to understand the impact of CLRI on plant per-
formance. OEE of the main Godowns had risen to 85% as a result of significant
reduction in breakdowns and short stops. The grading plan was completed ahead
of schedule even though the output was higher than in the previous year.

Stages 3 and 4 Process Improvement and Validation

The entire cycle of operations in the seed processing plant repeats twice a year,
once each for the Kharif and Rabi season, respectively. As discussed earlier, the
Case Study 1: Seeds of Growth 201

1.Raw Seed 2.Natural 4. Loading of


3. Grading &
unloading drying packed seed
packing
(Sunning) for dispatch

Fig. C1.5 Plant process sequence

operations are sequential such that different processes are performed at different
points in time. Hence, we are compelled to take up each process for improvement,
as and when it is running and complete the improvement and standardization cycle
simultaneously given the short window. Hence, Stages 3 and 4 of the roadmap were
happening together throughout the year. The overall sequence of plant processes
is depicted in Fig. C1.5; each box was taken up as a separate improvement project
within the Lean roadmap.
Given that the Lean initiative commenced in July, the seeds for Rabi season
were already in storage after drying. Grading (3) is done through a mostly auto-
mated set of equipment whose improvement was taken up through the equipment
reliability project. We therefore commenced our process improvements with the
packing line (3) in October, followed by unloading of seed (1) and sunning (2) in
November month. Peak dispatches happen in summer months, and we focused on
loading and dispatch (4) in April’ 21.

Packing Line Improvements


One of the major constraints for a seasonal industry such as PAN Seeds is the
inconsistent availability of workforce. The plant operates with contracted labour
gangs, each gang being allotted one or more Godowns. The members are involved
in all the activities of the Godown right from unloading incoming seed, sunning
related activities, operating packing line, and loading packed product. In the peak
season months of November and April–May, gangs were often found shifting from
one activity to another leading to disruptions in the packing line output. Data of
the past season highlighted multiple line stoppages as the entire gang went to help
out in sunning or dispatch loading.
The team was set with a goal of establishing packing line operations free of
Muda and Muri, such that the line could meet the target output with the minimum
“fixed” workforce. Then these workers would continue to run the line while the
other gang members would work on the other activities in parallel. The initial
workshop had already exposed the plant team to the concept of 3 M and waste
elimination. As Rabi season plant operations commenced in early October, the
team was dedicated to improving packing line within the first week of operation.
The typical seed packing process flow is depicted in Fig. C1.6.
There are three types of packing lines across the PAN Seeds plants.

1. Manual: All the activities are done by hand.


2. Semi-automatic: Mechanized conveyor line but a man has to support by holding
the bag in place for filling, sealing, stitching, etc.
202 Case Study 1: Seeds of Growth

Dumping Coating
of raw Fine Secondary (if
Seed Cleaner Grader needed)

Grading

Pre-Cleaner Primary Destoner Transfer to


Grader Silo

Graded Sealing
seed from /stitching Weighing Transfer
silo of bag the bori to FG area

Packing

Seed Putting Stitching


filling bags into the bori
into bag bori
(outer
bag)

Fig. C1.6 Seed grading and packing process flow

3. Fully automated FFS line: Operator involvement is minimized to machine


setting and outer bag (bori) operations.

However, irrespective of the type of packing line, the Lean concept to be applied
was one and the same—single-piece flow. Hence, three teams were formed, one
for each type of packing line and the teams executed the improvements in parallel
during the first week of the Rabi season. The approach taken to implement and
refine flow is detailed below.

Step1: Calculate Takt time

The packing line has the target of packing whatever has been graded. Hence,
grading capacity determines the takt time for the packing line. The team for each
Godown first computed the respective takt times.

Example:
Godown 11 Takt time
Grading capacity = 4.5 tonnes/hour
Pack size = 6 kg bags
Target = 45000/6 = 750 bags per hour
Actual working time per hour @ 85% effectiveness = 50 mins
Takt time = 50*60/750 = 4 seconds per bag
Case Study 1: Seeds of Growth 203

4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0

Fig. C1.7 Cycle time study of semi-automatic line

Step 2: Observe current operations

Team then observed the current line operations—and measured the cycle times
of each of the operations. The cycle times are depicted pictorially in Fig. C1.7.
The simplest way to identify bottlenecks in the flow is to observe where material
is piling up. In this case, the team observed that several bags were waiting in line
for operation 2: bag stitching. Every now and then, operators had to stop the filling
machine and remove the extra bags from the conveyor, and later during breaks,
they would put them back at the stitching station and stitch.

Step 3: Countermeasures

Ideas were discussed on improvements that could be done to reduce the waste in
the bottleneck operations using the Eliminate, Combine, Simplify, and Rearrange
(ECSR) principles. Solutions implemented include.

• Reduce incoming variability: In this case, it was observed that four filling heads
were in use with cycle times varying from 9 to 12 s. The filling head settings
were adjusted to ensure uniform cycle time of 9 to 10 s. Since the takt time
if 4 s, one filling head was put into standby mode. With three filling heads,
the cycle time was 3.3 s which ensure smooth flow. The smooth flow of bags
resulted in cycle time of the stitching operation also reducing to 3.5 s as operator
was undisturbed by bags piling up.
• Short cycle operations of weight checking, writing, and bori stitching were
combined such that a single operator could do the same. Each bori has 36 kgs,
i.e. 6 bags of 6 kgs each. Hence, the takt time for this is 4 s per bag × 6 bags =
24 s. With workstation arrangement and reducing distance between the stages,
one operator was able to complete this set of activities within the takt time with
minimum strain.
• In the manual line, a bori fixture to hold the bulk bori into which individual
packets are filled—a single operator is now able to stitch the bag and drop into
the bori held by the fixture. See Fig. C1.8 for the improvement done in bori
packing.
204 Case Study 1: Seeds of Growth

Earlier process Improved process

Fixture for
Bori’s – manually Packed Seed Bags holding Bori
held for packing

Fig. C1.8 Improvement in Bori filling

Step 4: Validate results

The packing lines thus improved in the first week of the Rabi season were
operated with the new methods for the rest of the season. The smooth flow with
the minimum required workforce was thus confirmed, and the outcomes mea-
sured showed a clear improvement in terms of both productivity and reduction
in wastage. As the short season concluded towards the end of November, the plant
team was confident of sustaining the new process in the next season.

Step 5: Making standards

Based on the data collected during Rabi season, process standards were drawn
up for the Kharif season. The standards were made product SKU wise for each
Godown packing line and incorporated the target output rate, resources (man and
machine) to be allocated, preparation activities to be done before start, and the
quality checks to be carried out. Following the standards over the Kharif season
of 2021 helped realize productivity gains across all the lines as shown in the Table
C1.2.

Table C1.2 Comparison of packing line performance


Parameter Manual line Semi-automatic FFS
Number of people per line Before lean 11 11 6
After lean 7 9 4
Seed wastage (average) Before lean 60 kg per day 100 kg per day 200 kg per day
After lean <10 kg per day 18 kg per day 60 kg per day
Case Study 1: Seeds of Growth 205

Sunning Improvements
West Bengal is a warm and humid place. Most of the seeds are harvested post-
monsoon and arrive at the plant in November and December. These seeds are
processed and packed only in April for farmers to sow by June. Seed has to ger-
minate only when planted; hence, this property has to be preserved for over half a
year. Moisture content (MC) of the seed is the single most important parameter to
control germination and needs to be below 12%. Incoming seeds may have up to
18% MC, and a natural drying in the sun is used to reduce this to acceptable level.
In past seasons, the team recorded that over 15% of incoming lots had to undergo
at least one “resunning” as the required MC level could not be reached in one day
of sunning. Sunning process depicted in Fig. C1.9 involves a lot of manual labour
and therefore additional cost if it has to be redone.
As per Lean paradigm, the only value-adding activity here is the process of
spreading exposing the seed to the sun for drying. Observing the process, the team
noted that spreading was generally completed by 10.30 am, and the seed is again
collected in heaps by 2.30 pm and repacked by 3.30 pm. This was due to the
winter months having early sunset; by 4.30 pm, it is dusk, and dew formation is
a threat to the moisture content of exposed seeds. Early morning fog also has the
same impact, and seeds cannot be spread before 8 am. For effective sunning, the
two main targets for process improvement were therefore to ensure that:

• Seed should spend maximum time in the sun.


• Maximum surface area of all seeds should be exposed to the sun.

The team implemented two major changes based on core Lean concepts.

Step 1: Seamless flow in the cycle of activities from bringing the stored seed
to spreading thereby reducing the “throughput” time for spreading. Table C1.3
compares the earlier and improved processes and their outcomes.

Bring 50 kg bags from Arranging bags in rows in the Open each bag and dumping
storage to the yard allotted sector of the yard the seed on the ground

Moisture > 12%

Spread the seed using rakes


Take bags to dumping area Check Moisture of seed
for grading when planned
Moisture < 12%
Allow seed to dry in the sun,

Take bags back to storage Shovel seed back into Raking the seed into
area the bag and restitch it heaps.

Fig. C1.9 Sunning process


206 Case Study 1: Seeds of Growth

Table C1.3 Throughput time of sunning process


Earlier process Improved process
Schedule Activity Schedule Activity
8–9.30 am Labour gang lays out bags of 7.30 am First lot of bags brought out
seed in the yard by labour gang
9.30–10.30 am Women workers cut open the 8.00 am Women cut open first lot,
bags; gang will start dumping second lot brought out by gang
the seed in parallel. Sample in parallel. Moisture check
collected and moisture done as soon as first row of
checked after complete cutting bags of a lot are cut open.
of bags for the lot
10.30–11 am Women rake the seeds of all 8.15–8.30 am First lot of seeds dumped in
the lots one after another tandem with cutting open the
bag. Raking done immediately
for each lot before moving to
next lot
Lot sunning begins from 10.30 to 11 am Lot sunning begins from 8.30 to 9.30 am
Average hours in the sun = 4 Average hours in the sun = 6

Step 2: The second target was to expose maximum area of seeds to the sun.
However, the yard area was limited and needed to be optimized. For two weeks,
data was captured on the depth of the spread seed layer and the incoming and
reduced moisture content. Through analysing this data, a correlation was estab-
lished—higher input moisture content seeds need more exposure to sun which
can happen through wider spreading to lower the depth of seed layer. Through
further trials, the team could establish a standard table which gave the depth
(inches) to be maintained for each range of incoming moisture.

But each lot has different quantities. So how do we know to what extent the seed
should be spread to maintain the recommended depth? For this the team divided
each yard into rectangular sections of standard 12 feet width, visually marking off
the length at 5 feet intervals as shown in Fig. C1.10.
For each depth, the square feet per kg of seed was computed based on practical
measurement of the spread seeds and standards fixed. The standards ranged from
2 to 5 inches layer depth, with increments of half an inch. For example, seed
with incoming 14–15% MC should be spread with a layer thickness of 3 inches.
The formula developed is based on kgs of sheet per square foot for a particular
layer thickness. Using this calculator made available on each Godown supervisor’s
mobile, he easily computes the surface area of spreading needed for the specific
lot quantity. Now since the width of each section is fixed at 12 feet, area/width
gives the length of the yard section to be occupied by the lot. Based on this he
instructs his gangs to first place a line of bags at the end to mark the border for
the lot. The gang then evenly spaces out the rest of the bags in the lot within the
boundary and once spread they cross-check and confirm the layer depth.
Case Study 1: Seeds of Growth 207

60 1D 1C 1B 1A
55
50
45
40
35
30
25
20
15
10
5

12 FT

Fig. C1.10 Yard marking

A small change in the procedure being followed for moisture checking had
to be done to enable this new method to be followed. Earlier, samples would
be collected after all the bags were opened before seed is dumped out and the
moisture checked later. Now, as soon as the first row of bags is placed, they are
immediately opened, moisture checked, and the area calculation done immediately
so as to give the labour gang clear instruction.
The outcome was validated within the months of November and December. The
resunning quantity dropped from 9% in the previous year (2019) to less than 5% in
the current year (2020). Only abnormally high moisture content batches needed a
second round of sunning.
The team has standardized this improvement to the extent that they revalidated
the entire standards for the Sunning which happens in May. The hot and humid
conditions of May are different from the cold and drier December. Tweaking of the
standards followed and now there are two sets of standards, one for each season.

Loading and Unloading Improvements


May is a hectic month for the plant. Most of the activities are going on simultane-
ously and are at peak load; unloading of raw seed and Sunning (for Rabi), grading,
packing, and dispatches for Kharif . Since the same labour gangs are more or less
involved in all the activities, there is a lot of firefighting in allotting and monitoring
work. One major concern was truck movement and handling. On any given day,
more than 50 trucks are found in the plant premises either for unloading raw seed
or for dispatch of finished product.
A cross-functional team of senior supervisor, plant manager, and logistics exec-
utive was formed to observe, analyse, and improve truck related operations. While
in Lean language, the entire operations are Muda as they involve only handling
and transport of material within the factory, it is a necessary part of the supply
chain. The main purpose of the activity is therefore deemed as value adding. In
208 Case Study 1: Seeds of Growth

Table C1.4 Comparison of


Unloading truck TAT (minutes per MT) Before 24
truck TAT
After 19
Loading truck TAT (minutes per MT) Before 12
After 7

this case, it is either “unloading” or “loading” which occurs when a bag of seed is
either removed or placed in the truck. So only that part of the time when people
are doing this is considered value adding.
The team observed the entire truck turnaround process (TAT) for both receipts
and dispatches and tabulated these in the form of operations analysis tables.
The observations showed that the actual unloading time was about 4 min per
tonne which was comparable to the benchmarks for such manual operations. How-
ever, 40% of the total time spent by a truck in the plant premises was for other
activities including security check, waiting at various places such as the weigh
bridge (before and after) and for documentation at logistics department.
The team brainstormed and came up with several actions that were imple-
mented within a couple of weeks. Major improvements in work flow and processes
are seen in eliminating duplication of documentation and streamlining the weigh
bridge operations.
Physical observations also highlighted a few bottlenecks in the infrastructure.
In a couple of places, trucks were unable to pass when there is already a truck
being unloaded or loaded. The roads were modified with small changes while a
second weigh bridge was installed considering the increased volume of operations
spanning 14 Godowns. The docking point in some of the Godowns was modified
with extension platforms such that two trucks could be placed and handled at the
same time as compared to only one being handled so far. All this enabled reduction
in the waiting time of trucks thereby improving the TAT as shown in Table C1.4.

Some Outcomes of One year of Lean

Some tangible gains


With the completion of Kharif dispatches, PAN had been through one entire cycle
of implementing Lean concepts in various seed processes occurring at different
points of time. The plant team tabulated the impact of this intervention on key
metrics that are being regularly tracked by the management. A comparison with
the previous year which was already considered the benchmark year at PAN is
shown in Table C1.5.

Empowering people
Given the industry sector, it is no surprise that the plants are located in remote
areas with no connection to any industrial zone. The extremely seasonal nature
of work means that fixed costs need to be kept to a minimum. Consequently, the
team running plant operations are from nearby villages and towns and none of
Case Study 1: Seeds of Growth 209

Table C1.5 Plant


KPI 2019–20 2020–21
performance comparison
(paddy seed—Bardhaman Total packed output (MT) 11,411 14,177
plant) Godown OEE 70% 82%
Productivity—labour cost per MT Rs.967 Rs.880
% rework in sunning 8.91% 4.76%

them are engineers or management graduates. They have never been exposed to
other industries or the work practices elsewhere and are cocooned in their own
world of seeds. Most of them have risen up through the ranks from being Godown
boys to senior supervisory positions.
But what has stood out right from the outset of this Lean journey is their hunger
and willingness to learn and try out new things. Having grasped the paradigm
of Lean thinking, they soon understood and applied the concepts be it “waste
elimination” or “flow” or “autonomous maintenance”. Realizing the benefits of
application first hand only served to motivate them further along this journey.

Sustaining and Moving Ahead

Seeing their involvement and interest, PAN rolled out a key result area (KRA)
linked incentive model for the first time in the company’s history. The team is
rewarded if they sustain the improvements made through the season and achieve
these KRAs. And they did it!. PAN’s management has seen the uptick in plant
performance and the growth of their people. At the plant, the labour gangs are
earning more by expending less effort. So, all the stakeholders see a good reason
to keep improving through Lean.
In the 2021–22 year, the plant team have already commenced the next round
of Lean activities. The Kharif data on plant performance including OEE, break-
downs, and short stops was analysed and specific problem areas identified. Teams
have been formed to work on these. For example, the short stops in the FFS pack-
ing lines in the plant’s Godown 14 were an area of concern. The team underwent
a problem-solving training workshop in August’21 and completed a root cause
analysis of the top two reasons for short stops. Actions have been taken and the
results are being validated during the November ’21 Rabi season run.
Pre-Rabi season, the team was engaged in strengthening the AM activity by
working on the repeated abnormalities, sources of seed spillage, and dust observed
in Kharif season. The CLRI sheet was fine-tuned and visual controls on the grad-
ing and packing machines are enabling the CLRI time to be reduced. The target
for next season is to do it in less than ten minutes every day.
And most recently in December ’21, the plant has formally launched 5S
activities starting with the stores. The packing materials, machinery spares and
consumables and the chemical stores have all been taken up for improvement.
210 Case Study 1: Seeds of Growth

Anshuman is now in the process of extending Lean to other aspects of the


business starting with the corporate functions of HR, finance, and logistics.
Case Study 2: Business Transformation
Through Lean

Background

Linkwell Telesystems has been operating in the electronics industry for over three
decades and has managed several product lifecycles since its inception. The com-
pany’s strategy is to develop potential high volume products typically ordered
through government tenders and supply these in bulk quantities. As a product
nears the end of its lifecycle, Linkwell’s strong R&D is ready to roll out and scale
the next product. Through this timely new product development (NPD) and launch
strategy, Linkwell was able to enjoy the fruits of telecom boom of the 1990’s and
early 2000s. Linkwell pioneered the point of sale (POS) devices in early 2000s
and is currently riding the energy meter wave owing to the rapid and sustained
growth of household and commercial establishments in the country.
While energy meters have been in the market since the past decade, there has
been a significant spurt in demand from 2017. That year, Linkwell had to ramp
up production fivefold from about 100,000 units to 500,000 units per month to
meet the requirements of the numerous contracts that it won in rapid succession.
This rapid growth was managed mainly by lateral expansion of production lines,
addition of facilities and developing multiple subcontractors. However, this being
a competitive and low margin business, the company’s bottom line was under pres-
sure in spite of the large growth in top line. Secondly, the product was expected
to soon reach the end of its life cycle as the next-generation smart meters were
expected to enter the market, rendering some of the investments in capacity expan-
sion useless. In early 2019, Linkwell’s top management started to feel the need to
improve profitability by reducing unwanted expenses while at the same time create
an operating model flexible enough to match the fluctuating demand conditions.
Way back in 2008, Linkwell had undertaken a pilot Lean intervention over a
six-month period under our guidance, in its coin operated phone assembly lines.
While the intervention gave a significant boost to the throughput, the Lean journey
was not continued then, due to various internal considerations of the plant man-
agement. But this glimpse of the power of Lean remained ingrained in the mind of
Ms. Radha, the Executive Director. And so it happened that her team once again

© The Editor(s) (if applicable) and The Author(s), under exclusive license 211
to Springer Nature Singapore Pte Ltd. 2023
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9
212 Case Study 2: Business Transformation Through Lean

reached out to us by the end of 2018. A preliminary visit to understand the cur-
rent state of operations in April 2019 was quickly followed up with an agreement
to embark on a year-long Lean program to streamline the entire manufacturing
operations for energy meters.

Diagnostic

We began with the current state assessment of the existing operations across
Linkwell’s four plants all located within a five-kilometre radius.

• Units 1 and 2 are in the main factory complex, covering meter calibration, final
assembly to packing, an SMT(Surface mounted technology for Printed Circuit
Board <PCB> assembly) line and the main stores.
• Units 5 and 6 have additional lines for product calibration, final assembly, and
packing lines.
• Unit 3 makes sheet metal parts while the adjacent Unit 4 produces the injection
moulded components.

Linkwell was working with over 40 subcontractors including six for SMT, ten
for leaded process (manual component assembly onto PCBs) and more than 20
mechanical assembly units, most of them located within the same 5 km radius
of the main units. Subcontractor management was handled in a decentralized man-
ner by respective final assembly units (1,2, 5, and 6). Each subcontractor is linked
to specific units. The production schedulers in these units issue schedules, coor-
dinate with the stores for material issues and follow-up (with subcontractors) for
the finished items. Oracle ERP’s material procurement, inventory, and accounting
modules are used while production planning was done on spreadsheets.
The diagnostic exercise brought together functional heads and unit in-charges
to observe their respective processes and compare the current state of operations
vis-a-vis organization goals and process targets. The cross-functional teams formed
comprised of the respective unit or line in-charges, the warehouse in-charges, qual-
ity team leaders, and members from support functions including HR, logistics, and
stores. These teams dedicated three full days under our guidance to complete the
current state assessment and frame an improvement plan for their respective areas.
The existing overall material and information flow diagram is shown in
Fig. C2.1. As it an be seen, the flow is quite complex, especially because a number
of intermediate processes take place at subcontractors. The complete set of com-
ponents is received at the main stores. The core of the product is the PCB; this
is initially processed at the SMT line at Unit 1 or at external SMT subcontractor
units. The PCB is then sent to the stores from where it is issued to various leaded
(PCB) subcontractors who complete the PCB assembly, test it for functionality,
and then send them to the mechanical assembly subcontractor units allocated by
the respective Linkwell unit scheduler. Each final assembly unit (1, 2, 5, and 6)
has their own set of leaded and mechanical assembly subcontractors with whom
they coordinate. Once the subcontractor has completed the meter assembly, the
Case Study 2: Business Transformation Through Lean 213

assembled units are dispatched to the respective Linkwell units for calibration,
final assembly and packing. The packed meters are stored in the finished goods
areas of the respective units awaiting customer inspection and dispatch clearance.
Unit 3 fabricates sheet metal parts that are mainly used by the mechanical assem-
bly units, while Unit 4 provides injection moulded components used both at the
subcontractor and final assembly units. All parts made by units 3 and 4 are routed
through a separate store located in the same premises. Throughout this case study,
we show Linkwell’s own units in light grey colour and subcontractors in medium
grey colour, in figures and various other exhibits.
The product flow involving interim processes at different subcontractor loca-
tions had resulted in non-value-adding material transport, handling, and documen-
tation. In addition, plan follow-ups and coordination required dedicated expeditors
who were firefighting all the time. The fivefold increase in capacity within a year
had been made possible by adding additional units 5 and 6 and rapidly devel-
oping multiple small subcontractors to support them. This sudden scaling up of
operations left no time for the management to design and implement ideal pro-
cess flow. The team concluded their diagnostic exercise with a listing of the major
issues identified and framed a Lean roadmap to tackle these issues. With the man-
agement’s focus on reducinfg expenditure having been clearly communicated, the
teams defined the process-level goals as detailed in Table C2.1.

Roadmap

With the above goals in mind, the roadmap was broadly designed to address the
overall value stream including the subcontracted processes.

Improve Flow

1. Focus Area#1 (Downstream processes): Connecting operations in flow, line


balancing, workstation improvements to enhance productivity.
2. Focus Area #2 (Subcontractor operations): Improve flow, workstations to
enhance output per subcontractor. With the above in place, reduce number of
subcontractors to simplify overall operations.
3. Focus Area#3 (Upstream processes, sheet metal, injection moulding and
SMT): Reduce changeover times through SMED and thereby WIP inventories.
Do problem solving and Kaizens to resolve specific issues.

Implement Pull

4. Focus Area #4 (Link up the entire value stream): Lean scheduling based on
simple pull systems to link these units to their customer lines be it subcontractor
or the final assembly lines.
214

Monthly Procurement
Suppliers through ERP & manual Customer
Monthly
Plan
Storage of components
at Unit 1 Main Stores Daily
Schedule
SMT –
PCB
Soldering Dispatch against
Customer schedule

SMT – PCB Leaded Unit


Unit
Soldering PCB 1,2,5,6
Assembly
Unit 1 process Calibration
Final Product
FG Storage &
Assembly &
Unit 1,2,5,6 TPI
Sheet Metal Packing Unit
Welding Unit 1,2,5,6
Fabrication 1,2,5,6
Unit 3

Injection Unit
Moulding 3/4
Monthly Unit 4 Stores
Schedule

Fig. C2.1 The material and information flow of current process. Notes Units 3 and 4 make their own schedules based on the monthly plan requirement for
assembly units
Case Study 2: Business Transformation Through Lean
Case Study 2: Business Transformation Through Lean 215

Table C2.1 Process goals for lean intervention


Section Parameter UOM Current state Target
Materials Space utilization Cubic feet (CFT) To be computed >90%
management (CFT)
Customer service Kit issue against <40% >90%
level JO
Stores Visual No Search free
management
Unit -3 (Sheet metal Material transport Distance >50 m <20 m
operations) moved (metres)
Quality First time right 98% >99.5%
Unt -4 (Injection OEE Ratio 70–80% >80%
moulding)
SMT OEE Ratio 60–70% >80%
PCB handling No. of touches 4 1
Lines (1,2,5, and 6) People Value adding % 50–60% >80%
productivity
Output per Meters per day 1600–2000 3000–3500
sealing line
Space utilization VA % <25% >50%
Facility utilization Distance 250–450 ft <100 ft
travelled
Rejection % rework 2% <1%
Finished goods Space utilization % Utilized To be computed >90%
operations Inventory Visual No Search free
management

Stabilize Processes

1. Focus Area #5—Standardize the improvements done through 5S.


2. Focus Area#6 (Equipment reliability) Autonomous and planned maintenance
to minimize short stops and breakdowns.

Implementation

The first three months were spent in connecting the process islands through a judi-
cious combination of flow and pull. In accordance with Lean philosophy, Linkwell
started the process improvement/waste elimination from the customer end with ini-
tial focus on the final processes comprising of calibration, assembly and packing.
After packing, the product is stored in the Finished Goods warehouse waiting for
dispatch clearance by the customer.
216 Case Study 2: Business Transformation Through Lean

Focus Area#1—Establishing Flow at Final Processes

All the finishing operations of the product are done in-house in manufacturing
units 1, 2, 5, and 6. A cross-functional team was formed at each unit to work
on improving the existing lines. After a couple of days of observation, the teams
made note of the following:

• Most assembly and packing operations are short cycle with less than 10 s cycle
time.
• As calibration operates in three shifts, inventory is automatically built up after
it is completed—this material searching, handling and transport.
• The meters for calibration come in from the mechanical subcontractors in plas-
tic crates with 72 meters per crate while post calibration 100 meters are placed
in a trolley for assembly in order to maintain the meter serial number. Each
meter has a unique number that is tracked all the way to the end customer point
and it is mandatory to pack the finished meters in the same order.
• The layout is such that there is significant transport of plastic components to
the line and assembled meters to the final packing area.
• Any short supply of the meters coming from mechanical subcontractor meant
that the calibration and assembly lines remain idle. This was found to occur
frequently and was one of the reasons why calibration was operating in the
night shift as well, despite having adequate capacity.

Based on the observations, three improvement projects were identified for


implementation.

(1) Improving assembly-to-packing flow for better productivity,


(2) Reduce calibration cycle time and thereby reduce the number of shifts,
(3) Link input materials and all processes through visual management and pull
systems.

Project 1—Assembly Line to Cell Conversion


Figure C2.2 depicts the basic value stream map (VSM) of the existing process.
An assembly line operated by 11 people with a maximum operation cycle time
of 12 s was delivering around 2100 meters per day. Process observations showed
significant imbalance between operations with inventory piled up at some stations
and other stations being idle. Each meter also had to be picked up multiple times
and pushed from station to station. This excessive handling not only was a wasteful
movement for the operator but also a cause for damaging the product. The team
worked over two months in a systematic manner to improve the process layout.

Step 1—Study existing line set-up and identify Muda, and Muri.
Step 2—Line modified from existing straightline table set-up to a U-shaped table
set-up.
Disconnected
High WIP ~ Disconnected processes
3 shifts processes

Assembly & Carton Sealing


Calibration Welding Testing packing
2 Machines
10 Benches 3 Machines 3 Jigs 3 lines 2 per
2 per 1 per 1 per 11 per
C/t : 240 sec.
C/t : 40 min C/t : 10 sec C/t : 30 sec C/t : 12 sec max. Batch Size: 20 units
Batch Size: 96 Batch Size: 1 Batch Size: 3 Batch Size: flow per carton
Available: 20 hrs Available: 7hrs Available: 7hrs Available: 7hrs Available: 7hrs
Case Study 2: Business Transformation Through Lean

40% NVA time Short Stops due


for loading and to input short
unloading supply from SC Imbalance –cycle
times varying from
4 to 12 sec

Fig. C2.2 VSM of existing process


217
218 Case Study 2: Business Transformation Through Lean

Packing boxes
Input trolley

Carton box stand boxes


Plastics

Fig. C2.3 Two-person cell workstation

Step 3—Line balanced and new workstations were set-up to minimize Muri such
as, picking up of large plastic components from bin kept on the floor, reaching
out for folding boxes for packing, etc.
Step 4—Trial done with two single person workstations on opposite sides of
an assembly table with one person assembling and second person packing. The
existing operations were redistributed between them to balance the cycle times.
Figure C2.3 shows the layout of such a cell.
Step 5—Both the improved line and the two-person cell were run for one month
along with the existing lines. The team observed work practices, monitored
daily production and quality, and concluded that the cell was most suitable and
effective method. Table C2.2 compares the performance of existing line with
the cell.
Step 6—Existing workers were retrained for multiskilling so that any of them
can assemble and pack the whole product.
Step 7—Two-person cell workstation standardized and replicated to form ten
such cells by dismantling the long tables used in the existing lines. This group
of ten cells was able to deliver the same output as the three lines in operation
before Lean.

The conversion of three long assembly lines into ten small cell workstations
necessitated a relook at the overall layout of the assembly floor. The earlier layout
depicted in C2.4 had certain other constraints such as zig-zag material movement
and excessive material handling.
Only the material planned for immediate dispatch can be stored in the Fin-
ished Goods (FG) area shown above. Bulk of the material had to be sent out in
forklifts to an adjoining FG warehouse. The plastic product covers are bulky and
are transported from Unit 4 in large specially made plastic lined boxes. These
boxes are transported by hand trolleys across the factory floor.

Actions Taken
Layout was modified to reduce zig-zag material movements. The major changes
here were,
Case Study 2: Business Transformation Through Lean 219

Table C2.2 Comparison of performance of assembly line and two-person cell


Process parameters Modified line Cell
Number of operations Assembly—7 Assembly—6
Packing—3 Packing—4
Cycle times Min.—4 s Assembly—36 s
Max.—12 s Packing—30 s, plus loading
carton box on conveyor every 20
pieces
Pick up and place time Pick up and place—3 s per Pick up and place—3 s per
operation operation
Total throughput time 120 s 66 s
Pick and place time 30 s = 25% NVA 6 s = 9% NVA
Outcomes
Number of people 11 2
Output per shift 2100 meters 700 meters
Productivity 190 meters/person 350 meters/person
WIP At any time at least 100 meters on Max. of 5 meters per cell
line due to imbalance
Rejection 1–2% <0.5%

Calibration

W W W
Input
WIP Post- Testing
Testing Testing
Calibration
LINE 1
LINE 3

LINE 2
LINE 4

Plastics Plastics
Laser M/c
SC Testing &

Box
Clearance
Packing Boxes

Staff Tables

Sealing
Finished Goods

Fig. C2.4 Earlier layout of the overall production unit


220 Case Study 2: Business Transformation Through Lean

• Welding and testing synchronized—the cycle times were more or less balanced
and hence it was easy to link these two processes without any further process
improvements. Tested meters are placed in movable trolleys in sequence and
parked in standard WIP area—marked for 2 h stock (16 trolleys) Further, areas
were designated for input, plastic, and packing materials .
• A readily available belt conveyor, taping, and strapping machines which were
not in use were renovated and put in place to complete the set-up. Online sealing
and strapping meant the carton boxes could be directly moved to FG store
thereby avoiding an extra handling.
• With the space freed up by the new layout, almost 75% of the FG pro-
duced could be accommodated in the same premises. Only items with
delayed or unclear dispatch dates are shifted outside.

The simplified layout and the improved work processes are shown in Fig. C2.5.

Calibration

Weld + test Weld + test W Input Input

C1 C6 WIP for
Packing Boxes

cells
C2 C7 Empty
SC Testing

Staff Tables
Space for
C3 C8 Second
C4 C9 Line Set up
C5 C10
Laser M/c
Test M/c

Box Output

Plastics

Finished Goods Finished Goods

Fig. C2.5 Simplified layout and improved work processes


Case Study 2: Business Transformation Through Lean 221

Project 2—Calibration Cycle Time Reduction


Calibration involves loading of software into the meter and validating. It is a criti-
cal process since it determines the proper functioning of the meter post-installation.
Each calibration bench, monitored by one operator, has two sides with 48 slots
each thereby processing 96 meters per cycle. The team wanted to reduce the
cycle time in order to complete more cycles per shift and reduce the number
of shifts operated. After observing five running cycles across various benches,
the team made an operations analysis table and identified value-added (VA) and
non-value-added (NVA) activities as shown in Table C2.3.
As can be seen from the above table, Muda activities occupied more than 40%
of the 45 min cycle. The team analysed the reasons for this and came up with the
following solutions to reduce this Muda.

1. Loading the next batch immediately after the activity 4 (instead of activity 6)
to start the software loading process. Then complete the sign off of previous
batch.
2. All the benches do not get loaded at same time and the operator of the adjacent
bench is now roped in to unload and load one side of the bench to which (s)he
has direct access.
3. Modifications done to software loading and verification reduced the time from
24 to 20 min.

Outcome of implementing and standardizing the above actions is shown in


Table C2.4.
Each bench can now calibrate 15 batches x 96 meters per batch = 1440
meters in a shift. With the available 7 benches, by ensuring 100% input material

Table C2.3 Operations analysis of calibration process


S.No. Activity Time taken (min) VA/NVA (Muda)
1 Loading the units from the trolley onto the bench 6 NVA
2 Software loading 20 VA
3 Verification 4 VA
4 Unloading the units onto bench table one by one 6 NVA
5 QC check sign off on each unit 7 NVA
6 Placing the units in sequence back on the trolley 2 NVA

Table C2.4 Comparison of


Parameter Before After improvement
calibration process before
and after improvement NVA time 21 min 7 min
VA time 24 min 21 min
Total cycle time 45 min 28 min
No. of batches per shift 10 15
222 Case Study 2: Business Transformation Through Lean

availability, the calibration output could match the day’s target of 10,000 meters.
Calibration operated in three shifts earlier. With this reduction of cycle time, it is
now able meet the upstream demand within general shift operation alone, enabling
it to go online with the assembly processes. Further, the room has five air condi-
tioners (1.5 T each) which are needed to keep the temperature below 24 °C as per
process requirement. Reducing to a single shift operation resulted in significant
power cost savings as well as the air conditioners are switched off after the day’s
work.

Project 3—Visual Management and Defining Rules to Link Processes


In line with the new layout and single shift operation, new material handling
norms were defined. All input materials are stored subcontractor-wise. Each SC
was allotted plastic bins of a different colour for visual identification. Movements
to and from calibration area were done only through numbered trolleys containing
100 meters each. Trolley parking locations were marked with space provided for
the defined WIP only.

Focus Area#2—Enhancing Subcontractor Capacity and Linking


with the Final Processes

In addition to higher cycle time, another reason for calibration operating all three
shifts was variability in lead time in receiving the assembled meters from all of
the upstream suppliers (both own units and subcontractors). So logically, the next
set of projects taken up by Linkwell were to ensure 100% on time in full (OTIF)
delivery by the upstream suppliers to the calibration process (in units 1, 2, 5, and
6). The overall material flow among the own production units and subcontractors
leading to the final process is shown in Fig. C2.6.
All the upstream processes in the figure above were taken up for improvement
over a three-month time frame. The SMT is a specialized process and as such
Linkwell had to depend on large outstation SMT units to obtain its net requirement
beyond what was made in house.

3.SMT – PCB 2.Leaded PCB 1.Mechancial SC Unit 12,5 and


Soldering Unit 1 SC process Unit Assembly 6 Calibration

SMT – SC 5.Sheet Metal 4.Injection


process Fabrications Unit 3 Moulding Unit 4

Fig. C2.6 High-level material flow between own production units and subcontractors
Case Study 2: Business Transformation Through Lean 223

Project 1—Capacity Enhancement of Mechanical Assembly


and Leaded PCB Subcontractors
The team proceeded to work with the immediate supplier, the mechanical subcon-
tractors, so that they could match the daily demand rate in a synchronous manner.
The aggregate requirement was catered to by a cluster of 20+ mechanical subcon-
tractors who were in turn supplied by a group of eight leaded PCB subcontractors.
Managing the quality, logistics, planning, and delivery for these many units was
challenging and involved a dedicated team of supervisors mainly for coordina-
tion and follow-ups. As a pilot, two subcontractors each from mechanical and
leaded PCB were selected and cross-functional teams were formed to observe and
improve their work practices.

i. Mechanical Unit Assembly.

Typical high-level process VSM of mechanical units is shown in Fig. C2.7,


The gemba observations showed testing, last operation, to be the bottleneck
with large number of PCBs piled up before it. Secondly, the assembly and sol-
dering operations were set up to operate in flow but the line was not balanced
resulting in small piles of inventory in between some of the operations. Soldering
was divided into four workstations causing multiple hand-offs of the PCB, one of
reasons for product failures detected during testing. The teams worked on each of
these areas as described below.

i (a). Testing: Complete cycle observed and operations analysis done to identify
value-added (VA) and non-value-added activities (NVA) as shown in Table
C2.5.

It is seen that the actual testing or value-adding time is only 30% of the total cycle
time. Further, PCB was handled multiple times during the test. The team came out
with a modified test jig with only three slots and reworked the test software to work
with this jig. This jig was now attached to the end of the assembly line thereby
integrating testing into the single-piece flow line. An added benefit was immediate
feedback on the type of failure to the respective workstation enabling a reduction
in the percentage of failures. With the new jig, the loading and unloading time has

Component Component Component


Unit Assembly Testing
Preparation Assembly Soldering

4 operators, 1 6 operators, 1 4 operators 1 6 operators, 1 2 operators,


day Batch piece flow piece flow piece flow Batch of 48

Fig. C2.7 High-level VSM of mechanical units


224 Case Study 2: Business Transformation Through Lean

Table C2.5 Operations analysis of testing process


Sl.No. Activity Time taken Remarks (VA/NVA)
1 Loading the meters from the 3 min NVA
crate one by one and locking
each meter in place
2a Testing 3 meters at a 12 min (1 min per set of VA time = 6 min
time—moving the probes, 3 meters) NVA time = 6 min
connecting and recording
parameters, comparing with
specs and pass/fail status
automatically generated
2b Visual inspection of VA
display—in parallel
3 Unloading the passed 3 min NVA
meters from the testing
machine and placing them into
plastic crate
4 Moving output to designated 2 min NVA
area and fetching next input
crate

reduced further and 3 meters are tested every 36 s. So, the improved process now
operates at an effective cycle time of 12 s per meter.

i (b). Unit Assembly: For the unit assembly line to be able to operate at 12 s takt
(time for the new testing process), the following changes were made:
• Workstation improvements to enable easy picking up of components,
• Line balancing using Eliminate, Combine, Simplify, and Rearrange
(ECSR) principles to ensure flow close to ideal single-piece flow.
i (c). Soldering

Earlier, soldering was split up among four operators as shown in Fig. C2.8.
This resulted in each PCB being handled four times which was detrimental
to both the quality, with increased chances for component being detached, as
well as to the output with an increased cycle time (due to non-value-added move-
ment of material between operators). The work flow and layout were changed to
accomodate single person workstations to enable parallel processing as shown in
Fig. C2.9:

Operator 1 Operator 2 Operator 3 Operator 4


PCB breaking Earth wires Lead wires breaking Connections

Fig. C2.8 Earlier soldering process


Case Study 2: Business Transformation Through Lean 225

Operator 2 Full Soldering


Operator 1 PCB Cycle time < 36
breaking. Splitting Operator 3 Full Soldering seconds per
set of 3 PCBs operator
Operator 4 Full Soldering

Fig. C2.9 Parallel processing in soldering

i (d). Component Preparation

The workstations were converted from line form to single person workstation
tables and shifted to a separate area. Each workstation was given the daily target
as per 12 s takt time. Standard inventory of half day was defined for prepared
components. While this is being consumed on the line, the requirement for the
second half of the day would be prepared and placed in the designated zone.

i (e). Standardization and Outcomes

Implementing Lean lead to a significant increase in the output of the subcon-


tractor units. Having perfected the flow in the two pilot subcontractor facilities,
the same was replicated across other selected subcontractors. Linkwell then fil-
tered out the poor performers and reduced the number of subcontractors thereby
making it easier to manage the supply chain. Table C2.6 summarizes the results.

ii. Leaded PCB

The through-hole process for soldering components onto PCBs used in the
energy meter was being carried out at eight different subcontractor units. The
tested and passed PCBs are placed in electrostatic discharge safe (ESD) bins
and sent to the main stores at Unit 1. At Unit 1, the PCBs are kitted with
other components needed for the meter, and are distributed to the Mechanical
Assembly subcontractors. Two PCB subcontractors were selected for piloting Lean

Table C2.6 Summary of


Before After
improvements in mechanical
unit assembly subcontractors Output per day 900–1000 meters 1500 meters
(potential of 2000)
Productivity 40 70 meters/person/day
meters/person/day
No. of 20 + 12
subcontractors
226 Case Study 2: Business Transformation Through Lean

Wave soldering machine


idling frequently

Component Component Manual


Wave Soldering Testing
Preparation Stuffing Soldering

4 operators 6 operators 2 operators 4 lines (11 4 set ups


1 day Batch 1 piece flow 1 piece flow operators each) 3 operators
1 piece flow 1 piece flow

PCBs piled
up, waiting

Fig. C2.10 Typical VSM of leaded PCB subcontractor

implementation. The typical process flow and main observations at the largest
subcontractor, Lalitha, are shown in Fig. C2.10.
The four manual soldering lines (and attached testing set-ups) were supplied
by a single wave soldering machine, which was producing around 4500 PCBs
per day. However, the wave soldering machine had an installed daily capacity of
12,000 PCBs. Testing was identified as the bottleneck operation and the team
began removing wastes from testing. After resolving this bottleneck, the team
worked on improving the utilization of the wave soldering machine.

ii (a). Testing

Testing of PCBs was done on a specially developed jig connected to a PC in


which the testing software was loaded. The operator fits the PCB in the jig and
gives the test command. All the parameters are checked through the software and
green lights show up on the monitor as each parameter is cleared. Once the test
is complete, the system alerts the operator whether the PCB has passed/failed and
the jig lifts up (error proofed) automatically. The tester signs on each passed PCB
and places it in the ESD bin, while failed PCBs are kept separately for analysis
and rework. Following were the team’s key observations:

• Testing set-up is physically located apart and operates on a standalone basis


necessitating supply of PCBs in bins from the manual soldering line to this
location.
• The testing cycle time is 120 s for a set of 3 PCBs. This translates to around
750 PCBs per day per jig. Two jigs were allocated to each line but the actual
output was around 1200 PCBs per line or about 4500 PCBs in total (from four
soldering lines).
• Out of the 120 s cycle time, operator’s work was only about 30 s which includes
loading, unloading, and signing on each PCB.
• Almost 10% of the PCBs are retested due to failure—actual failures were about
1% while the rest were due to connectivity issues in the jigs. This repeat testing
meant a wasted effort.
Case Study 2: Business Transformation Through Lean 227

Table C2.7 Operations analysis of PCB testing process


S. No. Activity Time taken (s) Remarks (VA/NVA)
1 Loading the 3 PCB board on to the jig 10 NVA
2 Testing preparation—erasing memory 30 VA
3 Loading the required protocols 20 VA
4 Testing the parameters 50 VA
5 Unloading the board from the jig 10 NVA

These observations led to immediate Kaizens that could significantly improve


the testing process.

• Jigs modified slightly to solve connectivity issues and avoid false alarms.
• Testing jigs attached to end of the manual soldering line so that the PCBs are
tested and packed inline and in flow mode. To match the line target output, two
test jigs were positioned such that a single operator can handle both test jigs
thereby increasing productivity and reducing the handling related Muda.

With the above changes, the testing output went up to 1500 PCBs per day per
line.
The line target was then increased to 2000 PCBs per day. Line balancing
through ECSR-based improvements was carried out and the soldering output
started meeting this target. This meant that the testing once again became a bottle-
neck. The team now conducted an operational analysis of the actual testing activity
as shown in Table C2.7.
Interestingly, the proportion of Muda was very less in this operation and there
was not much scope to reduce the loading and unloading time. The team therefore
started probing the value-adding activities and identified two improvement ideas:

(1) Erasing memory did not actually need the test jig. There was a possibil-
ity of doing it with a much simpler set-up, which was then designed and
implemented.
(2) Some of the testing protocols could be speeded up through tweaks in the
software.

As a result, the total cycle time on the jig was finally reduced to 72 s, resulting
in an output of about 1050 PCBs per day per jig, i.e. 2100 PCBs per day per line.

ii (b). Wave Soldering

To cater to four manual soldering lines each now operating at 2000 PCBs per
day, the output of wave soldering needed to be in excess of 8000 Nos per day.
Observations showed that the wave soldering was frequently idle as the rate of
output from the upstream process, component stuffing was lower. This manual
228 Case Study 2: Business Transformation Through Lean

Table C2.8 Improvements in output of the PCB subcontractors after Lean implementation
Output per day Before (PCBs per day) After (PCBs per day) Increase in output
SC Unit 1 4500 8000 75%
SC Unit 2 2000 6000 (extra line) 200%
SC Unit 3 1000 2500 150%
SC Unit 4 1800 2500 40%
Number of subcontractors 8 4 Reduced by half

stuffing was done in a typical moving conveyor with six workstations with each
operator stuffing (fix) a set of components as the PCB came along the conveyor.
Further observations of this process presented the following reasons for the slower
pace of component stuffing:

• Frequent movement of the first operator to fetch PCB boards placed a little
away from the line,
• Imbalance of the work content of the six operators in the line,
• At stations with a higher work content, the operators were frequently pulling
back the boards moving on the conveyor to complete their work.
• Several PCBs had to be reworked after visual inspection due to components
being missed out.

To address the issues observed, a single person workstation was created for
stuffing where one operator stuffs all the components as per PCB design. The
components were placed in front in small bins in order of the stuffing sequence
to avoid any errors of missing or wrong stuffing. After completing the operation,
the tray was placed onto the conveyor and moved to the visual inspection station.
An output of 1600 PCBS per day was achieved at the trial station, and five more
such workstations were set up with the sixth operator being assigned the role of
a material feeder to ensure each station is supplied with all the components and
bare PCB boards.

ii (c). Standardization and Outcomes

To match with the line demand, the output of wave soldering was standard-
ized at around 8000 Nos per day. The improvements done at pilot subcontractor
were replicated at other selected leaded PCB subcontractors, and the number of
subcontractors was reduced to four. Table C2.8 shows the enhanced capacity.

Focus Area#3—Improving Capability of Upstream Processes

With the final assembly and its immediate supplier processes reorganized along
the Lean paradigm with clear improvements, it was time to turn attention on the
Case Study 2: Business Transformation Through Lean 229

further upstream processes. As mentioned in the introduction, Linkwell has three


major upstream units—the SMT line (surface mounted technology) for production
of PCBs, sheet metal fabrication for making child parts such as the meter name
plate, fasteners, etc., and the injection moulding unit that made the body parts such
as meter casing, outer box, and other plastic moulded parts.
While capacity of these processes appeared to be sufficient to meet the targeted
output, their ability to actually deliver the components exactly when needed by
downstream processes was in doubt. Examples of these shortfalls are delays in
the name plates supplies by Unit 3 or in the injection moulded casings from Unit
4. Further, the SMT line appeared to be overloaded leading to Linkwell planners
having to outsource PCBs from external sources located in a different city. The
cost of outsourcing (subcontracting charges plus logistics) was higher than the
in-house production cost on the SMT line.
As before, cross-functional teams involving the concerned unit managers,
customers (assembly line leaders), quality team members and planning cell
member were formed, who observed the processes, and executed the following
improvement projects.

Project 1—OEE Enhancement of SMT Line Through SMED


The SMT team was already tracking parameters such as down time and quality, and
also, the SMT line performance data was also available from the machine software.
The Lean team’s initial computed Overall Equipment Effectiveness (OEE) was
only 60% which meant there was a significant scope for improvement. From OEE
analysis, changeover time emerged as the biggest loss, as there was at least one
changeover per day taking anywhere between 90 to 240 min depending on the
complexity.
The team participated in a three-day Single Minute Exchange of Dies (SMED)
workshop under our guidance. As part of this workshop, the team first observed
the existing changeover procedure, analysed the Muda, Muri, and Mura, imple-
mented solutions to reduce these wastes and thereby reduce the changeover time.
The current layout and SMT line process flow is shown in Fig. C2.11 with the
following four main processes:

1. Paste application—done through the printer using a stencil specific to each


PCB,
2. Component placement and assembly—two high speed pick and place machines,
two semi-automatic machines and one tray feeder,
3. Reflow soldering—done through an oven,
4. AOI visual inspection with predefined reference templates loaded.

The following are the team’s observations of changeover procedure.

1. Changing stencil and paste on printer—Movement from stencil preparation to


the printer location,
230 Case Study 2: Business Transformation Through Lean

2. Component loading on feeder reels and verification—Each product has anywhere


from 30 to 50 components most of them unique which means loading of 30 to
50 feeder reels per changeover. The component reels are brought from the stores
located behind the SMT.
3. Multiple crisscross movements between the jig, rack, and pick and place
machines were repeated for each component loading. Each component reel goes
to the jig for fixing in the feeder and these feeders loaded into the slots provided
in the auto feeder machines
4. After component reels are loaded in the feeder, each component is verified by a
quality assurance team member to ensure there is no error during loading. For
this, each reel has to be partially removed, the component number checked and
noted against the feeder location. These are then verified against the program
in the feeder display.
5. Program has to be called at each pick and place station while the PCB template
has to be called in the automated optical inspection (AOI) station
6. The first piece inspection procedurecommenced with visual inspection of first
board before reflow, then AOI check and clearance followed by passing another
four boards through the AOI. After this, clearance for production run was
given and this entire process took a good 20 minutes.

The team brainstormed to come up with a series of improvements in terms of


layout, method changes, and deploying additional available resources to speed up
the changeover process. These improvements included the following:

• Layout of feeder jigs, racks, stencils, and other supporting activities changed
to eliminate unwanted motion and strain. Stencil inspection workstation made
with provision for paste preparation, and the regularly used stencils were shifted
next to it.
• Modification of racks into slim trolleys capable of holding multiple component
reels with visual identification. The trolley is filled up with component reels
received from the main stores, reels needed for the next changeover are placed
in the same sequence as required in external mode, i.e. done while the SMT is
running, the trolleys are then moved to ear marked location next to the respec-
tive feeders. This eliminated the need to have a component store within the
SMT area.
• Verification by line operator eliminated, and it was found adequate to have
only one verification process by QA person after loading feeders. In a few
months’ time, the software provision for barcode scanning of component reel
and matching it with feeder location was activated to eliminate manual QA
verification.
• Deploy available resources—In addition to line operator, AOI operator was also
involved in changeover. The component preparation was completed in general
shift by supervisor and kept ready for changeovers.
• First piece inspection—This was replaced by passing the first five boards all the
way to AOI and clearing in one go.
Case Study 2: Business Transformation Through Lean 231

Stencils
Component Stores

TF
U
L N
O L
A Paste HSF HSF AF1 AF2 Reflow Soldering AOI O
D Printer 1 2 A
E D
R E
R
Monitor

Paste 3 level Component Reels HS AF 3 level Component Reels


Prep Rack Jig Jig Rack

b TF
U
L N
O L
A Paste HSF HSF AF1 AF2 O
Printer 1 2 Reflow Soldering AOI
D A
E D
R E
R
Monitor

Stencil + HS HS Pre-placed AF AF Pre-placed component


Paste Prep Jig component trolley Jig trolley

Fig. C2.11 a Earlier (above) and b modified (below) layout of the SMT line

Figure C2.11a, b depict the layout of the SMT line before and after the SMED
workshop. The dotted lines indicate operator movements for doing the changeover
activities
Once the benefits of SMED were validated, a changeover SOP and checklist
was prepared. Changeover times were displayed on a white board next to the
SMT line monitor to ensure visibility. Further additional feeders were procured to
facilitate external preparation of reel components.
Outcome

• Average reduction in changeover time by 40–50%.


• This resulted in increase in OEE from 60 to 70%. Further increase in OEE was
later taken up by an improvement project for reducing short stops.

Project 2—Component Flow in Sheet Metal Fabrication Unit


The most critical component made in the sheet metal fabrication unit is the
name plate (top/front cover) of the meter. The name plate carries all the prod-
uct details and a barcode for reading the meter. The process flow for name plate
manufacturing is shown in Fig. C2.12.
232 Case Study 2: Business Transformation Through Lean

Punching Bending Pad Printing Barcode printing

c/t < 5 sec c/t <10 sec c/t < 2 sec


c/t 10-15 sec
2 machines 3 machines 3 machines
4 machines
1 shift 2 shifts 1 shift
3 shifts
Batch production Batch Kit requirement
Batch production
production

Fig. C2.12 VSM of name plate preparation process

A monthly requirement plan is given by the assembly units (1, 2, 5, and 6). The
delivery is coordinated with the mechanical sub-assembly subcontractors who fit
the name plates to the meters. There are two name plate models in terms of design
and size. At the pad printing stage, customer-specific information is printed and the
name plate becomes specific to the specific customer order. The barcode printing
machines are located at the final assembly units. After pad printing, the Unit 3
hands over the name plates in plastic crates to their Stores. From stores these
are sent in small vehicles to the respective assembly units for barcode printing.
The unit coordinators then arrange to send these name plates to the mechanical
assembly subcontractors.
The complexity of the current process had resulted in the following major
problems:

• Name plates not available as per requirement—Unit 3 was manufacturing and


printing the name plates as per monthly plan. However, changes in customer
delivery schedule would lead to shuffling of the monthly schedule at the assem-
bly lines. But name plates would not be available requiring several follow-ups
and expediting with Unit 3 to urgently make the required name plates.
• High WIP of customer-specific printed name plates—At times, orders are held
up due to financial or other customer related issues and the name plates would
lie idle. Reworking of such printed name plates, by erasing the print using a
cleaning agent, and reprinting with other customer orders was also a regular
practice.
• Multiple handling and transport between Unit 3 to stores then to assembly units
for barcoding and onwards to subcontractor for mechanical assembly resulted
in complications in terms of material reconciliation, handling damages and
constant need to track the items, increasing the logistics cost.

In addition to the above observations, the team also captured the quantitative
data of the process in the following Table C2.9.
Based on the above observations and data, the team identified the following
areas to focus upon.

• Reducing number of touches, handlings, and transporting of nameplates,


• Minimize generation of defectives.
Case Study 2: Business Transformation Through Lean 233

Table C2.9 Initial performance of name plate preparation process


Parameter Nos Observations
Value adding operations 4 Cycle times are very less; bending and
pad printing variations in cycle time
Number of direct touches 8 Additional touches for cleaning of
printing defects, rearranging at pad
printing machine
Handling of crates containing name plates 12 At each stage, output is put into plastic
crates of standard size and moved to
intermediate storage location. Later, they
are taken to next operation when needed.
Additional handling at for loading and
unloading in vehicle
Distance moved 5 km Moves to units 1,2 and 5,6 for bar coding
Inventory points 5 After each stage
Defectives 2.5% Mainly printing defects, pieces are
cleaned and reprinted. A separate team of
four people were engaged in rework
activity

Project 1: Linking Sheet Metal Punching, Bending, Printing,


and Barcoding
This future state map of the name plate printing operations was visualized by the
team as shown in Fig. C2.13.
The target was to punch, bend, and keep a standard WIP before pad printing
since up to this stage, there are only two variants of name plates. Then based
on the weekly schedule, name plates would be printed and barcoded inline and
shifted to stores. Any changes in priority required one day’s advance notice. With
this redesigned VSM, material handling and transport would be minimized as the
barcoding is also to be done online. Several Kaizens were undertaken, as described
below, to help implement the target state VSM.

Kaizen 1: Bending process cycle time reduction


Observation of the bending process showed the actual value-adding time is only
about 2–3 s, while there was a lot of variation in unloading time. Figure C2.14

Punching Bending Pad Printing Barcode printing


c/t < 5 sec c/t 8 sec c/t 4 sec c/t 3 sec
2 machines 4 machines 3 machines 3 machines
1 shift 2 shifts 1 shift 1 shift

Fig. C2.13 Future state VSM of name plate preparation process


234 Case Study 2: Business Transformation Through Lean

shows the operations analysis with value-added (VA) and non-value-added (NVA)
processes identified separately.
The team performed a “why-why” analysis as shown in Table C2.10, to under-
stand the root cause of the variation in time for component removal from the
machine.
Following this Kaizen, the overall bending cycle time was reduced from the
earlier 10–15 s to a consistent 8 s.

Picking up sheet Press button ith Removing


Die punches and Placing in the
and loading the both hands component from
bend s component output bin
machine (safety) the machine
2-3 seconds 2 seconds
3-4 seconds 2 seconds 2-6 seconds

Fig. C2.14 Operations analysis of bending process

Table C2.10 Why-why analysis of process time variation (component removal)


Problem Unloading time varies from 2 to 6 s per piece
Why Difficulty in removing piece after punching
Why Piece getting stuck in locator pins at the bottom face
of the die
Why is it getting stuck? The bent part has to be lifted up and above the
locating pins on the bottom die, and this is a little
difficult to do due to less clearance
Why are pins needed? Pins help locate the part for punching
Why have they been designed like that? No specific reason. So, the height can be reduced if
needed
Temporary countermeasure Finally, it was seen that there is adequate reference
provided for centering by the pins at the back of the
die and the front pins are not even required. So, the
front pins were removed
Result The part could be removed manually and
consistently in 2 s
Permanent solution A Kaizen done by fixing a flexible air pipe to the
machine and connecting it to the available pneumatic
line. This was then linked to the upward stroke of
the punch. As the die goes back up after completing
the punch, the air nozzles blow a short 1 s burst of
air that blows out the bent parts directly into the bin
now located behind the machine
Case Study 2: Business Transformation Through Lean 235

Picking up 15-20 parts


Operate machine - pad
from crate on floor Place one piece on the Removie printed part
Clean surface by hand punches and prints on
and keep next to machine fromthe machine
2 seconds the part
loading point 2 seconds 2 seconds
2-3 seconds
1-2 seconds

Fig. C2.15 Operations analysis of pad printing process

Kaizen 2: Pad Printing cycle time reduction


As per the proposed process, a supermarket of bent name plate parts would be
available for pad printing. However, pad printing would have to work in continuous
flow with the barcoding machine which operates through a belt conveyor. This
would reduce the multiple handling of printed parts thereby reducing defects as
well. In order for this idea to work, it was necessary to reduce the pad printing
cycle time so that it can feed the faster barcoding machine synchronously. The
team’s operation analysis of the pad printing process with value-added (VA) and
non-value-added (NVA) processes identified separately is shown in Fig. C2.15.
The above observation led the team to undertake the following two major
process improvements:

1. Reconfiguring the workstation to avoid the first activity–the crates were placed
on a stand such that operator could directly pick up and load the part without
strain.
2. The cleaning of the surface manually with a cloth was consuming time and
not addressing the issue wholly. Smudged printing leading to reworks was still
happening. A foam roller was developed and attached to the machine itself
enabling it to move back and forth linearly as part of machine cycle. Once the
part is placed, the roller cleans the part, and as it goes back, the pad comes
down and prints on the part.

With these two changes, the cycle time of pad printing came down to below 8 s. On
the two pad printing machines, this meant that two name plates are printed every
8 s. In addition, defective generation dropped to <0.5% as the cleaning became
effective. The rework team was no longer needed as the few defective pieces could
be easily cleaned and reprinted at the end of the day by the regular production
operators.

Kaizen 3: Linking pad printing and barcoding


Now that the cycle times had become manageable, the assembly units were con-
vinced to move barcoding machines with operator to the sheet metal Unit 3. The
236 Case Study 2: Business Transformation Through Lean

layout inside the air-cooled printing cabin was changed to bring these two pro-
cesses in line thereby eliminating several handlings and touches. The earlier and
modified layouts are shown in Figure C2.16a and b.
Following two major benefits resulted from making these improvements:

• Avoid 5 km transportation for barcoding (see Fig. C2.12)—also related 4 extra


handlings of bins avoided, i.e. loading and unloading of transport vehicle,
shifting to location for barcoding,
• Online inspection gives immediate feedback to pad printing operator—elimi-
nated additional inspection in second shift and related handling.

a
Inspect (2nd
Move to Stores shift output)

1. Zig Zag material flow


2. Multiple handling
P1 P2
3. 2nd shift inspected next day
Inspect

Inspect
Input

Input
Output

Output

Inspect

Output
P3

Input

b
Move to Stores

P1
Output

Barcoding conveyor
1. Online barcoding
Input

2. Online inspection
3. Uni-directional flow
4. One touch handling
P2
Output

Barcoding conveyor
Input

P3
Output

Barcoding conveyor
Input

Fig. C2.16 a Earlier printing room layout. b Modified printing layout


Case Study 2: Business Transformation Through Lean 237

Kaizen 4: Standard WIP and Pull-Based scheduling


With the completion of physical linking of the processes and ensuring line balance,
it was time to link up all the sheet metal processes as per the envisaged future
state (see Fig. C2.13). Towards this goal, two supermarkets were created with
their specific rules as follows.

1. Post bending—Three days’ worth of production was to be maintained as WIP


for the two variants in proportion to their demand. Bin quantities were fixed
(500 Nos per bin) and floor marking for storage was done accordingly. As and
when the bins are pulled by the pad printing process and one day slot becomes
empty, punching and bending processes operate to make and fill up the emptied
slot as per the next requirement in the plan.
2. Printed name plates stock for next three days assembly schedule is maintained
in the Unit 3 stores. Here, a separate area was designated and floor marking
done for keeping the name plates customer-wise.

With the flow and supermarket pull systems in place, the shortages of name plates
were minimized and assembly units were achieving close to 100% of the plan.

Focus Area#4—Linking Up the Entire Value Stream

By now, the physical material flow had been streamlined and the various produc-
tion stages nearly balanced in terms of daily capacities. To ensure that the flow
sustains on a day-to-day basis, it was time to synchronize the supporting activities
such as materials management, production planning and quality systems. The next
three months were spent on synchronizing all support functions and aligning them
with the core process flow. The information flow was accordingly reconfigured.

Project 1: Materials Management—Ensuring on Time in Full (OTIF)


Supply of Kits
Linkwell was using a mix of Oracle ERP and excel worksheets for planning pro-
duction and material procurement. The overall flow of material planning is shown
in Fig. C2.17.
The obstacle to flow now was material shortages. The team analysed the short-
ages and grouped them under procurement and internal stores categories. Further,
they brainstormed, identified, and implemented solutions to address the shortages
as described below.

Procurement related
Monthly production plan is manually done in excel sheets separately by the assem-
bly units. This was then consolidated in another sheet by the supply chain team.
The plan is fed into the ERP and the computer program needs to run in the night
to generate shortage report, plan and other outputs. To overcome this delay, the
shortage report is often manually prepared by importing the stores stock into a
238 Case Study 2: Business Transformation Through Lean

Consolidated ERP run to update


Monthly Plan by
product wise material
Assembly Unit In-charges
month plan requirement

Prepare Shortage
Take Approval report
indents

Place Purchase Stores Receives


Order Materials

Quality Keep aside


check and return

Receive into
stores stock

Unit Not in stock Try to adjust from


indents as stock with other
per plan Subcontractors
Issue available
materials

Fig. C2.17 MRP run and execution

spreadsheet and matching against the plan. This method was cumbersome and
also error prone and leading to either excess stocks or stockout situations. For
example, when the ERP run was made for the July, using the manually generated
shortage reports, over 30 items were found to be short. At the same time, several
items were found to be in excess of three months inventory.
The systems engineer joined the core Lean team for this project and tweaked
the monthly run process, so that the shortage report from the ERP, instead of the
manual report, would be the basis for any procurement. For the common items,
which are used in most of the product variants, minimum and maximum stock
levels were set up in the ERP and triggers were set to alert the supply chain team,
whenever stock-on-hand either falls below or exceeds these thresholds respectively.
Only select items would be subject to manual intervention and indent approvals
post-discussion with the operations head. Even for these, a pop up was provided in
the ERP suggesting substitutes. If an item was in shortage, the supply chain team
would first look for substitute availability before planning a purchase.

Material Storage and Issue Related


The stores itself had a lot of materials lying without specified location, stocks of
material were also outside stores premises. The mechanical items were especially
Case Study 2: Business Transformation Through Lean 239

hard to find as materials were stored in a mixed-up condition. As we saw earlier,


there are two stores; the main store where all bought out items are received and
stored and a sheet metal and plastic components stores for the output of Units 3
and 4. Material was issued in kits to the subcontractors. A kit consists of all the
components needed to make the product and kit quantities are generated by the
ERP using Bill of Materials (BOM) based on the planned finished product quantity
to be entered by the planner. Kit issues to mechanical subcontractors was a com-
plex process—they would collect some items from the main stores, some from the
sheet metal unit stores and the PCBs that come in from the leaded subcontractors
issued by the final assembly units. This meant multiple handling, transport and
coordination among the two stores and the assembly units. Subcontractors were
also found making multiple trips using small vehicles resulting in further delay
and confusion at the stores. This was mainly because the kit would not be ready
and partial quantities had to be picked up for which using a larger vehicle meant
higher cost.
To begin with, a physical reorganization of the stores was carried out using
5S principles. The team decided that all mechanical sub assembly items would
henceforth be kept in the Unit 3 stores to have a one stop shop for kit collection
by these subcontractors.

Step 1(1S): Removed all non-moving, damaged, and unwanted items and
disposed them off as per management direction and cleared space in both stores.
Step 2 (2S): Shifted extra racks from main stores to sheet metal and plastics
stores and reorganized all the materials on 2 S principles with designated areas
for name plates, plastics, bought out items and other items.
Step 3: The kitting area was marked in both stores with provision of up to 3 kits
at any time. Kits kept ready in the morning and subcontractors given pick up
time slots in the afternoon, thereby minimizing the time spent by each vehicle
in the store’s premises. Once the full lot of kits was in place, subcontractors
were asked to bring larger vehicles to collect entire kit in one shot. The follow-
ing outcomes as shown in Table C2.11 were realized on completing the stores
reorganization.

Table C2.11 Improvement in kit collection time by SCs


Parameter Earlier After reorganization
Number of trips 1 trip to Main stores, 2 trips to Single trip to unit 3 stores with a
Unit 3 stores in small vehicle larger vehicle
Distance travelled 40 to 60 km 15 km
Turnaround time at stores Anywhere from 1 to 3 h 30–45 min
240 Case Study 2: Business Transformation Through Lean

Table C2.12 Why-why analysis for poor OTIF of kits


Why Why Why Why Solution
Full kit not Specific item Delay in Wrong data Manual excel ERP changes
available shortage getting stock working
SC item SC capacity Improve SC
delay process
Issued to other Issued Defined in Change SOP
SC quantity as current SOP
per month
plan

Kitting Practice Improvement


As noted previously, main reason for the delays and fire-fighting to achieve targets
was the non-availability of full kit of materials required for mechanical assem-
bly. During the initial diagnostic, it was found that the On Time In Full (OTIF)
against these indents was almost zero. Every indent had some or the other mate-
rial shortages. In many cases, materials were issued in part based on availability,
and to somehow ensure production targets are achieved, the unit coordinators were
diverting common components from one subcontractor to the other. There was no
trace of such material movements in the ERP. A why-why analysis was done to
identify root causes and develop actions to mitigate them as shown in Table C2.12.
The changes in ERP have been explained in the procurement-related issues
section while Focus Area 2 detailed out the improvements done at various subcon-
tractor processes. We now look at the planning SOP and how the team redesigned
the kit planning, issue, and reconciliation process to address this issue.

Existing Process for Kit Planning


Based on monthly plan, unit production planner would split the quantities to each
subcontractor and instruct stores to issue them kits in accordance to the same. The
stores would issue kits based on the plan and quantities would vary from 10,000
to 25,000 kits—anywhere from one week to entire month production. Since each
subcontractor would collect the kit on different days, a subcontractor collecting
material at a later date would find a shortage in an item due to less stock in
the stores. However, this item may have already been issued in full to another
subcontractor. The unit planner would then intervene and transfer part quantities
to keep production going at both the subcontractors. This sort of adjustments and
firefighting over a period of time had resulted in complete mismatches of stock of
various components at different subcontractor locations.

Revised SOP for Kit Planning


To ensure even distribution of kits and better controls, a new SOP was defined as
follows, using the “production levelling” principle of Lean manufacturing:
Case Study 2: Business Transformation Through Lean 241

1. Kits equivalent of three-day production capacity of the subcontracting unit


would be issued. For mechanical kits this would be of 5000 meters and for
leaded kits this would be either 10,000 or 20,000 PCBs. This is strictly driven
by the weekly schedule given by PPC every Thursday for the next week.
2. Kits would be issued twice a week and when the subcontractor still has a day’s
safety stock remaining. Different subcontractors were scheduled on different
days to level the kit preparation workload at the stores. For example, Unit 3
store has to issue kits to 10 subcontractors which means 20 kit issues per week.
The schedule is made for 3 kits per day.
3. The output of the subcontractor units is also routed through stores system. To
collect a fresh kit for 5000 meters, the subcontractor has to have delivered to
the stores 5000 assembled meters from the previous kit.
4. As schedule and quantity is fixed, kits for the day are prepared in the morning
and placed in designated kitting areas. SCs are given time slots post lunch to
pick up their kits. So, there is no waiting for them and they are able to carry
all the items in a single trip.

Project 2: Lean Production Scheduling


One of the biggest Muda seen in the organization was the large number of
follow-ups taking place across the operations—emails, calls, reviews, fire-fighting
meeting, personal visits to SC premises and interactions with stores. The root
cause for this was identified to be the current production planning, scheduling,
and monitoring system whose key issues are listed below:

• Upstream units make their own daily schedules keeping in mind equipment
changeovers, operator availability, raw material status, etc. They only ensure
that they operate within the monthly requirement plan which is also made on
a zero base each month. The plan does not factor in the stock of the already
produced components lying in the Unit 3 stores.
• As we have seen, the stores also operate independently issuing kits as per
monthly plan provided. They do not give feedback to upstream units on stock
levels or shortages.
• The assembly unit managers monitoring the final product actuals against the
plan are often clueless as to where things have deviated in the whole chain.
Sudden changes are made in assembly schedules due to material shortages or
customer priority changes and these are not in sync with upstream plants output.
Hence, further delays occur in dispatch schedules.

To summarize, all the stakeholders in the value stream be it supply chain, stores,
upstream units or final assembly units, were operating in silos once the monthly
requirement plan was given by the assembly unit managers. The disruptions,
changes, and constraints faced while executing the plan were communicated either
through emails or in weekly review meetings chaired by the Operations Head.
242 Case Study 2: Business Transformation Through Lean

Actions Implemented
A two-member production planning and control (PPC) cell was formed consisting
of one member each from the stores and production functions. This cell was ini-
tially entrusted with supply side planning for a month. After setting this in order,
the responsibilities were extended to cover the complete production scheduling,
monitoring and coordinating across all the stakeholders to ensure smooth flow
of material and information and ensure achievement of 100% against plan. Lean
scheduling practices introduced to facilitate their work are:

• Monthly requirement plan is translated into a weekly final assembly schedule,


• This weekly schedule is discussed in a meeting with all concerned unit man-
agers and functional heads, fine-tuned, and released every Thursday evening,
• Stores issues kits to subcontractor unit’s according to this schedule; each kit
equal to three days output,
• Unit 3 prints name plates from its standard WIP according to this schedule.
Recall that three-day standard WIP is already fixed with name plates,
• Unit 4 produces plastic components based on the fixed minimum stock levels
(MSL, Kanban trigger). As the components are issued and stock level reaches
MSL, a trigger goes to the Unit 4 production in-charge and who then plans to
refill the stock with a defined production run for the component,

Outcomes of Improvement in Planning and Scheduling

• OTIF for kit issue increased from <10% to >70% within six months.
• Assembly units achieved more than 95% against plan for three consecutive
months.
• With central PPC control, the firefighting, follow-up and related Muda have
been minimized and eliminated the need for multiple coordinators at each unit.

Standardize and Sustain

The last stage of the Lean transformation journey at Linkwell entailed setting up
systems and standards that would help the organize consolidate the gains obtained
through Lean and ensure sustenance of the Lean process in the long run. An oper-
ations excellence team was set up to internalize the Lean approach and be able to
take it forward once the consultant withdrew. One of the assembly line managers,
Vishnu, had already been given the role of Lean champion following the initial
improvement and consolidation of the assembly lines through cell concept. Now,
two more employees were added to this team (see Fig. C2.18 for the organization
chart) to handle the 5S and maintenance aspects that are part of the stabilization
phase.
The next level operational excellence concepts implemented over the last few
months were:
Case Study 2: Business Transformation Through Lean 243

Fig. C2.18 Lean


organization chart Operations Head

Unit & Functional


Heads

AM & PM
Lean Champion 5S Champion Coordinator

(1) Autonomous maintenance across all manufacturing units to reduce break-


downs, quality issues and short stoppages, and
(2) 5S assessment through a comprehensive checklist covering all aspects of Lean
implemented till date to enable sustenance and further improvement.

Autonomous Maintenance
With the flow established, quality systems set and pull-based scheduling in place,
the core operations were now able to deliver more output at a faster rate while
maintaining a minimum standard WIP across the value stream. Changeover times
were also reduced to support small lot sizes as per customer demand. But all this
could easily be disrupted by either a machine breakdown or variations in machine
performance. Hence, it was imperative to maintain the equipment in a reliable
condition and aim for zero disruptions. Linkwell implemented the first four steps
of Jishu Hozen (autonomous maintenance) over a three-month time frame.

Step 1: Initial cleaning and identification of abnormalities were carried out


across all the key equipment in all the four units and the utilities supporting
them. A total of 940 abnormalities were noted and more than 800 were closed
within a month’s time.
Step 2: Generation of countermeasures. Some of the abnormalities such as dust
in electrical panels or bent connector pins in the calibration equipment were
found to repeat across the plants. Why-why analysis was done to identify and
arrest the sources of contamination/root causes of the problems.
Step 3: Preparation of CLRI standards.
A cross-functional team consisting of maintenance and engineering staff,
unit production and line supervisors brainstormed to prepare cleaning, lubri-
cation, retightening, and inspection (CLRI) standards for their respective
equipment. These were then vetted by actual practice on the machines.
244 Case Study 2: Business Transformation Through Lean

Step 4: General inspection.

The CLRI standards were pasted at each machine centre in local language and
routines established.

• Operators were trained to follow the standards.


• Every morning before start of work, 10 min was allotted for CLRI.
• Abnormality registers were introduced at each section for operators to note
down any abnormalities identified.
• The Lean team audits the CLRI adherence physically once a month.
• To ensure that CLRI standards and machine maintenance activities are regu-
larly inspected, they have been incorporated into the monthly 5S assessment
checklist.

5S for Sustenance
Linkwell had already been implementing 5S over the past couple of years and had
a core team led by a 5S champion. This team was mainly doing gemba audits
for housekeeping, unwanted materials and equipment, abnormalities and capturing
these with photos. These were then discussed with concerned section heads who
would then make an action plan to improve upon the observations. To make 5S,
a vehicle for sustenance of the Lean initiative, a new assessment checklist was
created and implemented. The entire factory area was divided into zones, each
zone with two 5S champions or zone leaders and 5S activities under the Lean
initiative were implemented in the following four stages:

Stage 1—Self-Assessment.
Following a training on the checklist and its criteria, each zone leader
assessed his or her own zone along with a 5S core team member; they reported
the score, observations, action points for improving their score.
Stage 2—Core Team Assessment.
Two core teams were formed, one for the production and another for the
support function zones. After a fortnight, these core teams assessed all their
respective area zones and reported their own scores and the progress made by
the zone teams with respect to the action points.
Stage 3—Consultant Assessment.
A month later, we conducted a formal assessment along with the core
team and reported the first official scores for all the zones. The best per-
forming zone in the manufacturing units and best support function were each
awarded the rolling 5S trophy by the executive director in a formal get together.
Zone Leaders were recognized individually for their efforts. Each zone made
their respective action plans to improve upon their performance by the next
assessment cycle.
Case Study 2: Business Transformation Through Lean 245

Table C2.13 Consolidated benefits of Lean implementation


Process Measure Before After
Finishing lines Productivity (per person per 180 350
day)
Leaded subcontractor Output (Nos/day) 1500 per line 2000 per line
No. of subcontractors 7 4
Mechanical assembly Output (Nos/day) 1000 1500
subcontractor No. of subcontractors 20 12
SMT line OEE 58% 70%
Quality Rejection % (Top 8 defects) 1.23% 0.55%
Overall WIP inventory (Value Rs. 520 180
Million)
Space utilization 13,000 SFT vacated
Overall cost savings INR million per month 1.60
Labour 0.75
Power 0.08
Subcontractor 0.75
Logistics 0.02

Stage 4—Sustenance.
A roadmap was framed wherein a formal assessment followed by recognition
of the best 5 S zones would be done on a quarterly basis. The checklist would
help each zone frame a continuous improvement plan till the zone scores and
sustains a minimum of 90% for two successive assessment cycles. We anticipate
this to happen in about two years after which the checklist could be upgraded
and used for further improvement activities.

Conclusion

This relatively short Lean journey of nine months commenced with current state
assessment and roadmap framing, went through the improvement phase and cul-
minated in the sustenance phase marked by the 5S assessment. The team is
now empowered to continue the journey of continual improvement under Lean
paradigms. The team consolidated the benefits of this initial round of Lean
implementation over the year as shown in Table C2.13.
The monthly reducing inventory trend (in Rs. million) is shown in Fig. C2.19.

Post Script—The pandemic Era


Within months of the Lean journey detailed above, the entire world was hit with the
impact of the Corona virus. Being firmly on the Lean path gave a tremendous boost
to Linkwell in battling with the business and operational complications brought
about by the pandemic. While our engagement had already ended with the handing
over of the initiative to the Lean team, they rose to the occasion to take up the
246 Case Study 2: Business Transformation Through Lean

1600

1400

1200

1000

800 Total Inventory


WIP
600

400

200

0
May-19 Jun-19 Jul-19 Aug-19 Sep-19 Oct-19 Nov-19

Fig. C2.19 Declining inventory month-on-month

challenges posed by the pandemic situation and worked on further improving and
standardizing different aspects of the process.
For instance, they extended the visual management concepts to their file man-
agement systems in all the departments such as HR, procurement, administration,
etc. that improved the ease of retrieval, reduced the space usage, and reduced filing
errors. A number of low-tech, low-cost, but effective poka-yoke-based procedures
have been put in place to combat COVID, such as hands-free door opening/closing
of all office/wash rooms, visual marks for social distancing, etc. This indicates how
deep the Lean-based problem-solving methodology has been ingrained into the
thinking process of every employee of the organization. In fact, the daily routines
of every department have Lean practices instilled within them, thereby avoiding
any notion that Lean is something different from the daily responsibilities of the
individuals. Finally, the organization created a performance measurement system
that cascades from the top to lower most levels, in which Lean goals are explicitly
indicated to every level. In effect, the company has made Lean as a way-of-life in
every aspect of its daily operations, problem-solving, learning, decision-making,
and managing in internal and external processes.
Case Study 3: The Lean
Restaurant—Serving Customers
Effectively

Background

Liberty exclusive is a leading hospitality service provider running restaurants (A’la


Liberty) and a banquet facility at prime locations of Hyderabad, India. Liberty is
well known for high quality vegetarian catering services for a wide range of func-
tions and events. Over the last decade, the restaurant has been adjudged the Best
Vegetarian Restaurant in Hyderabad multiple times by the prestigious Times Food
Guide awards. Recently, Liberty has opened its first cloud kitchen for takeaways
and is planning to expand this as a franchise operation across multiple locations.
Mr. Vishal runs this family promoted business, handles the overall operations
and personally manages the restaurant at Banjara Hills. His mother manages the
catering services and Rishabh, Vishal’s younger brother, handles the day-to-day
operations of the restaurant located at the upmarket Tech Park area. Liberty has a
loyal and growing customer base and a strong brand image and Vishal saw a huge
potential for growing the business and extending the brand to cater to the booming
online food delivery business of service providers like Swiggy, Zomato, etc. He
also felt the need to improve bottom line as expenses had grown at a higher rate
than sales in the restaurants.
Vishal then roped in Prasad, his ex-colleague from Taj Hotels to handle oper-
ations and drive improvements so that he could focus on growing the business.
To realize the potential, Vishal was looking to strengthen the core operations and
delivery processes and establish a standard operating model that can be scaled
easily. During a conversation with one of Liberty’s regular customers, Vishal was
referred to us. After a preliminary meeting, Vishal engaged us to help Liberty
improve and standardize its existing operational processes using the concepts of
Lean and Kaizen. A six-month time frame was fixed for the first Lean intervention
with a focus on the restaurant business segment.

© The Editor(s) (if applicable) and The Author(s), under exclusive license 247
to Springer Nature Singapore Pte Ltd. 2023
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9
248 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Management Goals

Following a gemba walk through of the restaurant facility, we sat down with Vishal
and defined the following goals to better the restaurant operations:

1. Increase resource productivity of People (revenue per employee), Material


(Zero wastage), and Space (revenue per square foot),
2. To reduce employee stress and strain while handling the day-to-day opera-
tions—avoid “firefighting”,
3. Improve customer experience by reducing their waiting time from entry to
exit, and ensuring consistent delivery of specified and expected quality in food,
facilities, and service,
4. Establish standards for the core processes and a system for monitoring adher-
ence. The standard practices should help business scale up smoothly and
quickly.

The Approach

Unlike manufacturing, restaurant operations are highly variable, with peaks and
troughs. Liberty operated lunch and dinner buffets all seven days of the week while
also offering guests choice of a’la carte menus. Past data showed that over 80% of
the guests prefer the buffet. There was also a marked difference in terms of vol-
umes through the week. Being a family-oriented restaurant, weekends contributed
to bulk of the revenues. Additionally, volumes spiked during festival holidays, with
guests having to wait for a while to be seated.
A core team comprising of Vishal, Rishabh, Prasad, the customer care manager,
floor manager and the head chefs was formed to work with us on the project.
The Lean intervention began with a current state assessment during which the
entire restaurant operations were viewed through the Lean paradigm of customer
perspective. In hospitality parlance, the customer is known as guest and he or she
pays for the food and service. The unique feature of the restaurant business is that
the customer also participates in the process, specifically in the case of a buffet
meal. Hence, the initial observation focused on guest parties such as couples, small
groups (4–8 people) and large gatherings (>10 people). The entire value stream
was observed from entry of the guest party at the reception desk located at the
entrance till the guest party leaves the restaurant after paying the bill, and the
table is cleaned and set up for the next guest party.
From this initial observation, the entire set of operations could be sorted into
three distinct flows.

• Guest flow,
• Food flow,
• Service staff flow.
Case Study 3: The Lean Restaurant—Serving Customers Effectively 249

The overall process flow of restaurant operations integrating these three flows is
shown in following Fig. C3.1.

Obstacles to Flow

After observing each of the three flows, the team was able to arrive at the main
impediments to smooth flow and frame a roadmap of improvement projects to
address these constraints.

Flow of Guest

Under Lean paradigms, the guest should flow smoothly and seamlessly through
the service environment without having to wait anywhere. The following obstacles
were identified where guest flow was disrupted and the guest was inconvenienced:
Waiting for table due to.

• Reservation mix ups,


• High table set-up time.

Waiting at the table.

• Waiting for staff to take order, serve water,


• Waiting for soup/starters to be served.

Issues observed at buffet counter.

• Difficulties in picking up items,


• Waiting for crockery (katoris*, soup bowls, dessert plates) and cutlery,
• Waiting for food items which are stocked out,
• Early arrivals wait as some items are not yet ready at buffet.

*A katori is a small bowl used for items like lentils, curd, etc.

Flow of Service

Majority of the delays for the guests were due to non-availability of service staff
near the tables. The team observed that the staff were busy either getting starters
from the buffet area or crockery/cutlery from the washing areas. Starters are gener-
ally prepared on order and served hot; they include a mix of Continental, Chinese,
and Indian items. All crockery and cutlery used by the guests are transferred to
the dish washing area periodically, where they are rinsed, washed and dried before
being returned to the buffet counter or guest tables. It was felt that addressing the
250

Materials Ingredients
Management issue

Food Preparation

Food Pans
Table service items refilling
– soup, starters

Guest Entry Guest


Guest Guest Billing &
at Welcome Seated&
eating Buffet eating Guest Exit
Desk Ordered

Table Service of
Reservations Clearance Bill
setup water Side
station Preparation

Crockery, cutlery Washing &


replenishment Drying

Fig. C3.1 Guest, food, and service staff flows in restaurant operations
Case Study 3: The Lean Restaurant—Serving Customers Effectively
Case Study 3: The Lean Restaurant—Serving Customers Effectively 251

delay in starter preparation and speeding up washing process would free the ser-
vice staff to attend the needs of the guests in a timely manner and ensure better
guest service levels.

Flow of Food

Cooking of food as per recipe is the core value-adding process of a restaurant. The
team observed the activities that make up this process and identified the following
impediments to flow.

• Delays due to non-availability and late issue of ingredients from the stores,
• Time lost in searching for items in the kitchen,
• Constraints in starters preparation process in the live kitchen.

Implementation

The obstacles identified were analysed further to arrive at the root causes and a
four-month roadmap was formulated prioritizing the focus areas for improvement.
Lean focuses attention on the customer experience, and hence, the first priority was
to improve the service levels through ensuring increased staff availability and atten-
tion to the guests. How Liberty achieved these major operational improvements are
discussed in the following sub-sections.

Increasing Service Staff Availability at the Guest Tables

A simple why-why analysis was helpful in identifying the reasons behind guests
waiting for service. Table C3.1 describes the analysis.
Based on the above analysis, the following process improvements were identified;

1. Streamline and ensure flow of the cutlery and crockery from point of use to the
cleaning and back—one touch concept,
2. Early warning of stockouts of the cutlery and crockery at the buffet counter and
refilling through pull-based replenishment using tray Kanban,
3. Redefine roles of service staff—separated focus on guests, buffet and washing
areas.

We present the process of analysis, deriving the required changes to be made


to achieve the improvements, implementation, and the resultant outcomes in the
following sub-section.
252 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Table C3.1 Why-why analysis


Issue Why Why Why Why Why Why Solutions
Guest Lunch (for Chef cooks food As per existing Revised staff
waiting for staff) not ready after preparing buffet practice lunch
Gone to eat lunch
steward/wa before 1 pm rice preparation
iter to order time to 12 noon
Staff not available Gone to the soup/ Responsible for Roles as defined by Redefined roles
near the table area starter pick up serving these restaurant of service staff
counter items to the management (see below)
guest
Gone to the Delay in getting Delay in drying of Multiple handling Batch system One touch
washing area for clean cutlery/ washed cutlery and strain water for washing & handling and
fetching cutlery/ soup bowls between washing, drying; NVAs flow in washing
crockery drying and taking the in method and drying
items out(?) followed
Guest Waiting for During rush time, Pull based
waiting to crockery and crockery not replenishment
pick up cutlery available at the using trays
items at buffet counter
buffet Item stockout in pan Delay in refilling No mechanism No specific Person Redefined roles
food pans to check stock assigned of service staff
level (see below)

One Touch Handling Method and Flow in Washing and Drying


The current flow of crockery and cutlery is depicted in Fig. C3.2. The process
operates in a batch mode with no fixed schedule or quantity. Each section works
to complete whatever is available with them at the time, when it is free. Hence,
multiple stacks of crockery are observed in between the processes.
Preliminary observation showed stacks of dirty crockery waiting to be washed
and this made it clear that washing was the bottleneck and was not able to keep up
with the demand, especially during peak hours. An in-depth observation revealed
that the existing layout of the washing area was leading to multiple touches (han-
dling) and crisscross movement of staff. Please see Fig. C3.3. Dotted lines and
dark grey boxes show glasses flow, and the solid lines and light grey boxes show
the other crockery and cutlery flow.
The glasses were washed in a separate line as they needed to be handled in
a delicate manner. These were mainly used for juices and water, and there were
no delays observed in availability. However, both the ceramicware (crockery) and
stainless-steel cutlery were being handled in the main washing line. While only
four operations—clean, rinse, wash, and dry are done, it was seen that each item
was touched about 10 times. The washed crockery is stacked on a shelf in the
order they are washed. This meant a mix of bowls, katoris and plates which have

Periodically Take back


Use Clear from take and Wash -
table, place to point of
crockery/ dump in continuous Segregate Dry
on side use when
cutlery washing flow
station asked
area

Fig. C3.2 Process flow of current operations


Case Study 3: The Lean Restaurant—Serving Customers Effectively 253

Entry / Exit
Clean & Rinse

Wash
Dirty Stack
Load tray

Wash
Dry

Glasses
Stack

Dirty
Segregate Dry & Stack
Glasses

Fig. C3.3 Existing operations flow and layout

to be again segregated when they are unloaded into a basket. This basket is placed
on the floor. A staff member, sitting on a low stool, picks up one piece at a time,
dries and places it in a stack on another table. This stack is then placed in a tray
by one of the restaurant staff and carried to the buffet counter where once again it
is removed from the tray and stacked.
Using Lean flow principles, the team changed the layout to improve flow and
drastically reduce handling, movement and employee strain. In the new layout, the
table for placing dirty crockery was shifted adjacent to the rinsing sink thereby
avoiding the need to lift and move crockery. An additional table was located next
to the stacking location. The washing staff were now asked to place the items
in separate baskets or drying stands as they are removed from the washing pro-
cess. Another staff member, stands at the drying table, and continuously dries a
set of items from one basket placing each dried item directly on trays meant for
transporting the item back into the restaurant. The filled tray is then placed at the
buffet counter and an empty tray brought back to the drying stand for the next set
of washed items. Now all operations are carried out at waist height thereby avoid-
ing the need for staff to bend down or sit in uncomfortable postures. Modified
operations flow and layout is shown in Fig. C3.4.

Pull-Based Replenishment Through Trays


Crockery relevant to buffet such as soup bowls, katoris, dessert plates, and juice
glasses are kept in the shelves of the buffet counter. In line with industry norms,
Liberty had crockery of about 1.5 X (times) the restaurant capacity, in circulation.
At peak times (say 1.30–2 pm on a weekend lunch hour), there was a lot of
firefighting to restock the washed crockery resulting in guests facing a delay at the
buffet counter. The restocking additionally inconvenienced the guest as the service
254 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Entry / Exit
Dirty Rinse

Wash
Online drying Segregated
and load on trays stacks

Dry and load on


tray

Wash
Lunch table for
staff

Dirty Glasses

Fig. C3.4 Modified operations flow and layout

staff was found to be blocking the guest movement to pick food; he would take
about five minutes to stack soup bowls or katoris one by one at the earmarked
location. The team worked out a solution to reduce this time taken to replenish
crockery at the buffet counter thereby minimizing the inconvenience to guests.
Table C3.2 summarizes the improvement done.
The stock position of the crockery item and call for replenishment is based on
visual observation by the floor supervisor at the buffet. As a tray gets empty, he
signals the waiter to get a filled tray from the drying station. An empty tray left
at the drying station is a signal for the back-end staff to dry and place 24 fresh

Table C3.2 Existing and improved cutlery replenishment processes in the buffet area
Existing process New process
1. Random number of Bowls brought in a 1. Fixed number of 24 bowls bought in a holding
tray from the washing area tray
2. Waiter crouches down at the counter and 2. The waiter bends down momentarily, removes
places each bowl from the tray onto the empty tray, and replaces with full tray
shelf 3. Takes empty tray and returns it to the drying
3. On completion waiter takes empty tray workstation for replenishment
and leaves it on a side station#
Maximum number of bowls stacked: 64 (2 Total number of bowls stacked: 96 (4 trays × 24
rows × 16 × 2 high) height restricted due to bowls each, 2 trays high)
direct stacking

Cycle time per tray: 5 min avg Cycle time per tray: 30 s
#Each service section has a side station (cabinet with table top) used by the staff for supporting all
their work
Case Study 3: The Lean Restaurant—Serving Customers Effectively 255

bowls in the tray. A similar system was fixed for katoris and plates each with its
own defined quantity.

Service Staff Roles Realignment


The why-why analysis has shown that current attempts by the service staff to
somehow fulfil their duties is also a reason for the delay in serving guests. We
stationed ourselves at the door leading from the restaurant to the back area (kitchen
and washing) for about ten minutes during peak hour and observed that the door
was opened more than a hundred times in just these 10 min. Almost every member
of the staff—steward, waiter, chef, and even the restaurant manager walked in and
out multiple times during this period. The fact that the service staff including the
section steward and floor manager keep disappearing into the kitchen area meant
that they were often not found by the guests when needed. Not only was this
leading to poor service to the guests but also affecting the kitchen hygiene as
multiple people kept trampling in and out. The staff were also under a lot of strain
due to this constant movement from their service area.
The restaurant layout below gives us a picture of the extent of motion and
non-value-added activities taking place. The movement of one set of staff serving
Section B is shown—the other section staff also have a similar set of movements
for service, food pick up, and clearance. Existing layout and staff movement is
shown in Fig. C3.5.
To reduce the movement and increase the service staff presence in the dining
section for the guests, the roles of some of the staff were redefined. One senior

Dining Section - A Live Kitchen


1

Main Kitchen
Buffet Counter

Dining Section - B
Washing

Dining Section -
Dining Section - C D

Fig. C3.5 Existing layout and staff movement


256 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Buffet Food
1. Replenishment of food in buffet pans, and
2. Starters delivery to stations

Kitchen Coordinator
1. Visual check on stock of food, crockery and Clearance
cutlery at buffet, 1. Clear used items from station to washing, and
2. Starter orders to kitchen, and 2. Refilling and supplying water to side stations.
3. Assist guests at buffet

Crockery / Cutlery
1. Replenishment of crockery at buffet counter, and
2. Replenish of cutlery to service area stations

Fig. C3.6 Redefined responsibilities of coordination team

captain and two stewards were pulled out from their respective service sections
and were formed into a separate team. The role of this captain or kitchen coordi-
nator was to synchronize the kitchen, washing area and front end (dining sections)
seamlessly. The roles were redefined such that only three people were allowed to
pass through the door into the kitchen area thereby freeing up the section service
staff to be in the vicinity of their guests. One person each was allocated to clear-
ance from side station, for replenishment of food and replenishment of crockery
and cutlery, respectively. The new team structure is shown in Fig. C3.6.
Following this realignment of staff roles, the movement of the section service
staff was hugely minimized as none of them were expected to enter the kitchen
area. All their movements were up to the nearest side station only. Figure C3.7
shows this improved flow with movements of section service staff and the newly
created coordination team.

Reducing Guest Wait Time for Table

On an average, the restaurant can service two parties of guests per table. The
lunch/dinner time is for about 3 h, and data shows that one party of guests occupy
a table for about 75–90 min. The guests who have made reservations for the latter
time and the walk-in guests are often found to be waiting at the reception area
for up to 30 min. Sometimes, to placate the irritated guests, the hostess would
hurriedly have them seated at the table while the table setting process was still
incomplete.
Part of this delay was attributed to time taken to clear and set up the table after
the previous party has finished their meal. Persistent observation revealed that the
Case Study 3: The Lean Restaurant—Serving Customers Effectively 257

Dining Area - A Buffet Kitchen

Back-end
1

Kitchen
Buffet Counter

Dining Area - B

Washing
Dining Area - D

Dining Area - C

Fig. C3.7 Post-improvement staff movement

speed of the table setting activity was affected by two things and the team worked
on improving both these factors that were causing a delay in table setting.

Unavailability of All Required Items in the Section Side Station


Each service section has a side station (cabinet with table top) used by the staff for
supporting all their services. The Lean core team went through these side stations
and observed the following:

• Variation in type of item and quantities from one side station to the other,
• Items found in mixed-up condition,
• Unwanted/unused items occupying the drawers,
• Muda of motion due to non-availability of certain items. For example, hand
wash bowls are available but the steward has to go to the kitchen area for hot
water.

The team used 5S to improve and standardize the side stations:

1S: The side stations were completely emptied and unused and unwanted items
were immediately discarded.
258 Case Study 3: The Lean Restaurant—Serving Customers Effectively

2S: The service team brainstormed and decided on the list of items that are
mandatory for each station. It was decided to stock the following items and the
corresponding quantities:
• Crockery and cutlery: 1.5 times the number of guest covers#
• Table cloths, napkins and other items: 2 times the number of covers
• Housekeeping items—cleaning sponge, dry cloth, spray, etc. one set per
station
#A cover is one seat on a table.
This was to ensure zero delay in resetting up of the table after one party of guests
leave.

• The side station has three levels—table top, drawer, and cupboard below. A
place for each type of item was fixed based on the frequency of use, size and
weight of the item and space available for fitting in the required quantity. All
items were placed as per labelled location only.
• A master list of items to be found in the side station with required quantities
was made and stuck to the inside of the cupboard door.
• Hot water kettle procured for each side station—this can be used both for hand
wash bowl as well as for guests who ask for warm drinking water.

3S: Daily morning practice of cleaning the side station and checking the
contents instituted.

Cleaning and Table Setting Procedure


The Single Minute Exchange of Die (SMED) technique was adopted for improving
table setting process. Firstly, a video of the entire process of cleaning and setting
up a typical 6 cover (seater) table was shot. The whole process was then observed
in detail with the concerned staff and analysed for Muda, Muri, and Mura. Some
key observations made by the team members are:

• Variation in practice from person to person,


• Variation in sequence of activities from one set-up to the next,
• Cleaning done by housekeeping staff and table set-up by steward. There was
waiting time for in between both activities as concerned people were doing
other jobs.
• Shortage of some items—stewards moving to other stations and sometimes to
the kitchen area to get the items (glasses/plates/cutlery),
• Time taken to refold each napkin—it is placed in the side station as received
from laundry but while placing on table the orientation has to be changed.

The team then zeroed on the following improvements:


Case Study 3: The Lean Restaurant—Serving Customers Effectively 259

• Side station reorganization (as detailed in previous section),


• Defining best and simplest way of setting table and having the service staff
practice the same,
• Napkins pre-folded by laundry service provider as per Liberty specification.

Outcome
Earlier table set-up time recorded was about 7–8 min for a six-seater table; after
the improvement it was reduced to 3–4 min and clear communication protocol
extended between the steward and the front desk to ensure that guests are sent into
the restaurant for seating only after the table set-up is completed.

Speed and Efficiency of Food Preparation

While Liberty had won multiple awards for food quality and taste, customer feed-
back highlighted issues with respect to delay in service of food. Having worked on
and enhanced the service staff attentiveness to the guest, the core team next turned
its attention to this issue, and was able to observe and identify the few specific
causes that were contributing to this delay. Kaizen projects were drawn up and
implemented to address these issues.

Delay in Service of Common Starter Items


There are some starters which are on the menu every day without exception. Pizza
is one such item with only the vegetable topping changing from day to day. Obser-
vations made during the buffet peak time showed delays in supplying the pizza
slices to the pick-up counters. Stewards and captains would often be seen waiting
for the pizza, and customers repeatedly asking them on why they have not yet
been served. The Lean concepts of VSM and material flow were used to better
understand the issues related to the pizza preparation and delivery process.

Process Analysis, improvement, and implementation

Analysis
At the peak hour about 60 people would be having starters at the same time.
Past data analysed showed an average consumption of 1.5 slices per person which
translated to peak requirement of 90 slices or 15 pizzas. 1.30 ~2.00 pm is peak
time. Delivering 15 pizzas in this half an hour implies a takt time of 2 min per
pizza. The value stream map is shown in Fig. C3.8. Each oven can accommodate
2 pizzas side by side and can therefore bake 1 pizza every 90 s which is less than
takt time of 120 s.
The above data and observations show that there is sufficient capacity to deliver
the peak requirement. However, the actual output was constrained by the following:
260 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Crisscross
movements of staff
Many bases between prep, oven
lying and pick up counter
around

Oven: Place on
Cut Oven: Remove
Prepare Apply Final pickup
Veggies Pre-bake and slice
base toppings bake for counter
and grate for 3 min 2~3 min hot stand
cheese

Trim
edges if
Variation in burnt
baking time Rework of 1
– operated in 3 pizzas
manually

Fig. C3.8 Value stream of pizza making

• Multiple movements of the chef and assistant to prepare and deliver a pizza—
also this movement was clashing with the chefs preparing fried starters and
tandoor snacks.
• Mismatch of base preparation and oven baking results in inventories of prepared
bases or stockouts. Inventories cause loss of freshness and stockouts cause idle
time on the oven.
• Variation in baking time due to exigency—at times 3 pizzas are stuffed in the
oven resulting in low quality (underbake). Other times, there is only one pizza
or temperature is turned up to speed up the baking leading to burnt edges. This
in turn resulted in additional rework activity of trimming the edges causing
further delays.

Improvements

(1) Flow—The starter area layout was modified to reduce movements of chefs
and their assistants and streamline the flow of starter preparation and delivery
to pick up point from where the service staff would collect their starter orders.
In the existing layout there was a crisscross movement of the assistant chef
for baking the pizza, again taking it across the room for slicing and moving
the sliced pizza to the pick-up counter. Similar movements for preparing tan-
door (a mini earthen oven) snacks as well as cold cut vegetable preparation
can be seen in Fig. C3.9a (Earlier layout diagram). Since the pick-up point
for starters was located right in the middle of the buffet, the guests picking
up their food were forced to go around the service staff waiting at the pick-up
Case Study 3: The Lean Restaurant—Serving Customers Effectively 261

a b
Cut Sink wash Veg rack

Cut
Sink wash

Prep

Pizza Prep
Pre p

Table
Tandoor + roti Cold

Tandoor
Pre p

Cold
Pizza Prep Table
Roti

Oven

Oven &
Cutting
Pizza 3
Cutting 1
1 2 3

Pick Up
Pick Up Buffet (Salad, Food Pans)
Buffet (Salad, Food Pans) Point
Point

Buffet (Main Course)


Buffet (Main Course)

Congestion
Guest Flow

1 – Hot Tawa (Uttapam)


2 – Fryers for snacks
3 – Pasta Live station

-------- Steps for pizza preparation


-------- Steps for Tandoor preparation

Fig. C3.9 a Earlier (Left) and b improved (Right) layout for live kitchen

counter. The congestion here was also inconveniencing guests picking up food
items kept close to the pick-up point.

With the layout modification, smooth flow of work for all the starters made
in the live kitchen was established while reducing strain on the chefs and their
assistants. The tandoor snacks and rotis were also now delivered in tandem with
customer requirement by adding one more tandoor. As the diagram shows, the
new layout has also helped ease the congestion at the buffet counter and reduced
inconvenience to the guests who were picking up their food. This set-up of smooth
work flow for pizza enabled the implementation of the concept of customer pull.

(2) Pull—The oven is divided into two plates—one is used for pre baking and
the other for finishing. Each plate can accommodate 2 pizzas side by side
and gives a consistent quality. Hence, a simple FIFO lane-based pull system
was put in place with a fixed quantity of two pizzas. This ensured that no
unnecessary WIP of pizza bases builds up and no pizzas are baked in advance
and become stale. This visual PULL mechanism is shown in Fig. C3.10. At
any time, two sliced pizzas are on the hot plate at the pick-up point. Two pizzas
are baking in the oven, if the hot plate is not empty, these pizzas remain in
the oven in warm mode. As the hot plate becomes empty, the assistant chef
removes two pizzas from the oven and transfers the two from pre-bake to bake
while putting in two new prepared pizzas into pre-bake oven plate. He slices
the removed pizza, places it on the hot plate, and then turns his attention to
preparing two pizza bases to replace the ones shifted into pre-bake oven.
262 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Oven 1 Oven 2 Slice and


Ingredients place

Grated cheese
FIFO
Pizza Base 7 5 3 3 1
Pizza Prep
Cut Veggies
8 6 4 4 2

Fig. C3.10 FIFO-based pull of pizzas

Through this combination of layout change and establishing a visual FIFO lane,
the team ensured that hot pizza slices are available at the pick-up point at all time.

100% Item Availability at Buffet at Restaurant Opening Time


Guests coming in just after the restaurant has opened were seen to wait as some
of the food items were not yet ready in the food pans at the buffet. While the
guests were not happy, their table occupancy time was going up reducing the time
available for serving the next party of guests. The team first prepared the process
flow as in Fig. C3.11 for the kitchen activities and then observed each activity
to identify the constraints and reasons for delay in food preparation. The main
factors contributing to this delay were receipt of indented material from the stores,
or incomplete indents by the chefs. This led to last minute rush to stores to get
the missed-out ingredients resulting in delays in pre-cooking activities. The team
came out with solutions to address each of these issues and implemented them to
good effect, as detailed in the following sections.

Delays in Kit receipt from stores


This was identified as the root cause for delays in food preparation. The Liberty
kitchen is structured on food type basis—separate chef teams handle Continental,
Chinese, Indian, tandoor, sweets, and deserts. Each kitchen chef places an indent
for all the items needed for the next day’s cooking as per the agreed menu. The
items are picked off the racks and placed in a basket by the store’s assistant; each
basket is a kit for the respective kitchen. The assistant chef arrives at the stores by
9 am to collect the kit meant for that day’s cooking. The team identified several
gaps in the material procurement, storage, and issue processes and acted on each of
them to improve the on time in full kit availability. These are listed in Tables C3.3
and C3.4:
The kit issues system is set such that each kitchen’s kit items are placed in
a basket and the baskets placed on a rack at the stores entrance. At 9 am, the
assistants pick up their baskets, check the items against their indent and move on
Case Study 3: The Lean Restaurant—Serving Customers Effectively 263

Chef Prepares indent for


next day menu

Items checked
against indent by
Stores Officer

Not available Available


Stores PI raised and Kit prepared by stores
PO prepared assistant in the night

Items got Assistant chef collects kit


from market from Stores between 9-10 am
Urgent

Emergency purchase
from nearby shops in Assistant chef collects
the morning utensils from washing area

Assistant chefs start– cutting,


pre-cooking etc from 10 am
onwards

Chefs Collect Semi


finished items from
freezers

Main chefs start the food


preparation 10.30 am
onwards

Fig. C3.11 Food preparation process flow

Table C3.3 Stock availability-related issues


Observations Actions taken
• Many items stock not available in the stores, • Introduced reorder levels and defined the
and this is not monitored in advance same in system with 10-day time based
• Stock data not updated in the system—in replenishment
some cases stock is available and its location • All the items in the stores master list were
known only to the stores assistant sorted kitchen-wise with procurement lead
• After issue, stock is updated on paper and time, typical daily consumption, number of
this data is latter re-entered into the days of stock to be maintained
software. There were several instances when • Daily online updation every evening
this update is missed out
264 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Table C3.4 Storage and issues of Kits


Observations Actions taken
• Items found mixed up in racks and • The entire stores was first reorganized to
cupboards and several items are found on make it search free, count free and strain
the floor as racks were overfull free while enabling FIFO as the items are all
• The procedure involves weighing of major of perishable nature
ingredients before issue against the indent • Out of 12 racks, three racks removed and
given. In practice, the staff found it replaced with pallets for storing bulk items
strenuous to bring heavy flour bags from its like rice, sugar, and flour
location to the weighing scale, lift them on • Change in layout of weighing scale to
and take them off for issue facilitate weighment and sealing
• For several items, the indented quantity has • Pre-preparation of bulk pack items to smaller
to be made up from bulk quantity pack 1 kg/500 gm packets during the non-busy
received from the supplier. This involved afternoon hours using bag sealing machine
additional handling of bulk bags to and from • Change in storage method—instead of item
weighing scale, transferring part quantities type wise the racks were reorganized
to the smaller bags and tying up these bags kitchen-wise (see Figs. 12 and 13 for new
to avoid spillage store layout below)
• Kit preparation takes up a lot of time and
effort—the assistant has to move around the
entire store 6–7 times to fetch the items for
one kit

Empty Desks with cupboards shelves on top (Files) Junk


(Vaastu) Scale Items
Office
Entry Material
Cupboardd IN /Out
Freezer 1
Cooler

Bulk Items
(mixed up)
Seal Freezer 2

Fig. C3.12 Stores layout and an example kit preparation for one Indian kitchen
Case Study 3: The Lean Restaurant—Serving Customers Effectively 265

Empty Desks with shelves on Spices


(Vaastu) top (Files) Packing
Kits Material
IN /Out
Continental

Scale

Indian Seal

Common Items
Chinese Deserts

Pallet 1
Tandoor

- flour

Pallet 2
- rice
Freezer 1
Freezer

Cooler
Pallet 3
- others

Fig. C3.13 Revised layout (blue dotted line indicates repacking into small packs which is done
during non-peak hours)

to their kitchen to start food preparation. The reorganization of the stores layout
helped reduce the movement of the stores assistant for getting the kits ready and
simplified the process thereby enabling him to have all the kitchen kits ready
well in time. The layout change was done keeping in mind Vaastu considera-
tions; Vaastu is an ancient Indian convention of planning spaces within building
structures. Figures C3.12 below shows the earlier lay-out and material movement
path and C3.13 shows the revised lay-out.
The following results were seen after implementing the above improvements.

• It took just about 15 min per kit and motion reduced to less than half.
• Item stock-outs and emergency purchases reduced by more than 50%.

Incomplete Indents
Incomplete indents are second major source of delay at the time of food prepara-
tion. It was observed that the chef would miss out on a particular ingredient in the
indent and realize that it is not available with him only at the time of food prepa-
ration. An assistant chef would be sent rushing to the stores to get the item issued;
in case of stockout, an emergency purchase had to be initiated further delaying the
cooking process.
Most recipes have been developed in house by the chefs and a lot of information
is in their minds. Especially for menu items that appear irregularly, it is quite
possible to miss out on the ingredients while preparing the indent. Vishal would
266 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Table C3.5 BoM for example recipes


S. No. Item No. of covers 10 Basket
Kitchen UOM Std Req
1 Herb guava Juice ML 500 Guava juice, guava crush,
chilli flakes, tabasco
sauce, oregano
2 Manchow soup Chinese ML 1000 Ginger, garlic, green
chilli, beans, cabbage,
carrot, coriander, salt,
pepper, sugar, soya
sauce, corn flour
3 Corn samosa Tandoor PCS 35 American corn,
capsicum, coriander,
green chilli, jeera, red
chilli, coriander powder,
paneer, maida, and oil
4 Italian mushroom salad Continental GM 500 Mushrooom, bel pepper,
balsamic vinegar, salt,
pepper, crushed pepper,
salad oil, chilli flakes,
oregano, and basil
5 Aloo Chana Chaat Indian GM 250 Potato, chana, tomato,
onion, chat masala, chilli
powder, cumin powder,
lemon, coriander, salt

work closely with the head chef in the menu setting and development of new
recipes and he realized the importance of having the recipes on paper.
The team spent several days in developing a Bill of Materials (BOM) master
using Microsoft Excel. As each day’s menu was decided, the team sat down to
list down the ingredients in details. The chef would do a dummy run through of
the preparation and each ingredient was placed on the table in front of him in
sequence to create the recipe. These details were inputted into the master sheet
and a drop-down menu was created to link the master sheet with the menu sheet.
Now, as the day’s menu is made, the ingredients needed for each item auto-
matically appear next to the item. The sheet has a provision to add the estimated
number of covers (guests) and calculates the food quantities to be made. Using
this ready reckoner, the chef only needs to verify/input the quantity needed and
prepare the indent. Table C3.5 shows a few sample entries from one of the menu
sheets.

Time Wasted in Searching for Semi-Finished Items Kept in Freezers


A standard practice of the food industry is to prepare in bulk, certain semi-finished
items and stores them in freezers. These items are taken out as per required
quantities, thawed and used in the final preparation. This reduces the daily work
and speeds up cooking without compromising on quality. Some examples include
Case Study 3: The Lean Restaurant—Serving Customers Effectively 267

Indian gravy, chutneys, pre-cut veggies, etc. Every morning, once the chefs begin
their work, they send their assistants to get the required items from the common
freezers located at a corner of the kitchen.
The core team observed this set of kitchen activities and listed the Muda below:

• Assistants have to walk some distance through the kitchen to reach the freezers.
• They search for the item and are able to locate it after having to remove various
other items in the compartments.
• Several times the chefs had to come over as the assistants could not find the
required item.

A meeting was held with the head chefs, and on their advice, a common training
session was held for all the kitchen staff on 5S and visual management. After this,
the core team in conjunction with kitchen staff implemented the following actions:

• Re-layout of freezers to reduce movement distance


• Within a freezer, a set of compartments was allocated to each of the kitchen
sections, for e.g. Indian, Chinese, Continental, etc.
• Each section cleared out the unwanted/non-moving items from the compart-
ments
• Each kitchen then segregated the required items within the allotted compartment
shelves in the following categories:

a. Unprocessed items such as cheese, vegetables, milk packets,


b. Semi-finished (cooked) items such as gravy, chutneys, etc., and
c. Finished (excess) items to be reheated and consumed on the same day.

To prevent accumulation of unwanted items again in the freezers, the food


management system was defined clearly with the two SOPs explained below.
Replenishment of buffet to avoid food wastage—Time and quantity-based replen-
ishment was defined. At 2 pm during lunch and 9.30 pm during dinner, the
restaurant manager and head chef check the food available in the food pans. Based
on the reservation position and this food stock, a decision is taken on which items
should be replenished and in what quantity. This is then communicated to the
kitchen and only those items are cooked again while the remaining semi-finished
items are put away into the freezers. Special care is taken for items that cannot
be stored at all such as, juices and certain cooked items that cannot be reheated.
A leftover management was created (sample in Table C3.6) to update the stock of
finished items.
Daily work checklist made and put up in respective section—This is especially
critical for the Deserts section as some of the recipes involve overnight setting.
The checklist sheet has the plan which is filled up by the head chef and actual
completed work is updated by the section chef. A sample is shown in Table C3.7.
268 Case Study 3: The Lean Restaurant—Serving Customers Effectively

Table C3.6 Leftover management board


Date: 30-Oct
Prep date Item Quantity Exp date Action plan Status
29-Oct Pineapple pastry 1/2 kg 2-Nov Consume in buffet on Pending
31-Oct

Table C3.7 Work checklist for deserts


Date 29-Oct
Activity No Item Shift Plan Actual Status
Preparation 1 Baked rasgulla 1 120 Nos
2
Finishing 1 Chocochip tart 1 100 Nos
2 Chocochip tart 2 80 Nos
3
Advance prep 1 Pineapple pastry 2 2 kg 1.5 kg Material shortage
2 Pineapple pastry 2 2 kg
3
Dispatches 1 Chocochip tart 1 150 Nos 120 Nos Shortage of base
2 Orange cheesecake 3 kg 3.2 kg OK
3

Food Preparation Time


Once all the items have been brought from the stores and required utensils are in
place, the assistant chefs start the preparation which includes washing and cutting
of vegetables, boiling/cooking of rice/noodles/pasta, and preparing spice mixes.
Among these, the maximum time and effort is spent in vegetable cutting. A number
of non-value-added activities (muda and muri) were observed in this process.

• Moving to and from the sink to wash and clean vegetables,


• Scraping, throwing away the outer skin and other waste etc., and
• Variation in cutting speed from person to person.

To reduce the time spent by chefs on this activity and free them up for the more
value-adding cooking work, a preparation section was formed to cut vegetables,
with two assistants drawn out from the existing kitchens. The layout was modified
to make a cellular arrangement as shown in Fig. C3.14.
Each kitchen notes their requirement of cut vegetables in a register maintained
in this preparation cell. Based on priorities given by head chef, the two-member
team wash and cut the vegetables and place them in the cut vegetables rack from
where the respective kitchen assistant chefs collect them.
Case Study 3: The Lean Restaurant—Serving Customers Effectively 269

Fig. C3.14 Preparation


W Cutting table
cell—vegetables cutting

Cut
Veg.
rack

Input
Sink - wash
Veg Rack

By combining the Muda done by each section into one preparatory section and
reducing this Muda further through cellular layout about half an hour time was
freed for each kitchen chef.

Standardization of Operations

Once the improvements were completed, new standards which would help in sus-
tenance of improved processes were created. Following standards were developed
and implemented:

• Restaurant audit checklist—A weekly checklist using 5S as the base, this was
used to check the level of cleanliness, organization, and service level of each
section. The highest scoring section was recognized.
• Roles and responsibilities were clearly defined for the restaurant manager, host-
ess, kitchen coordinator and head chef so that they can function in tandem
without duplicating or interfering in each other’s functions. In line with Lean
philosophy, their primary function is to support the value adders such as, chefs
and service staff.
• Restaurant closing SOP: The process of closing at night is one of the most
important factors that affect the setting up of the restaurant the next morning.
The restaurant closing process was observed in detail on a peak weekend night
and several improvement actions identified. To standardize the whole process,
a restaurant closing checklist was also prepared and implemented. Within days
the closing time had reduced from 90 min to less than 60 min with complete
cleaning and quality checks.

Overall Outcomes

The success of any process improvement is determined by practical measurement


of defined metrics which in turn would result in the improvement in the stated
business goals mentioned earlier. While we discussed process level outcomes in
270 Case Study 3: The Lean Restaurant—Serving Customers Effectively

each section, here we mention two key overall business-related gains seen in this
project:

1. Record increase of turnover per day by 40% was achieved without firefighting
and not adding a single resource,
2. Improved guest service which is reflected in 20% rise in customer satisfaction
levels measured through feedback forms and online reviews.

Conclusion

Lean and the standardized processes that emerged out of the intervention have
given Liberty the confidence to expand their core kitchen business. In the pan-
demic year, Liberty has already opened their first cloud (takeaway) kitchen to
great success; the kitchen SOPs are based on the same principles of speed of ser-
vice (order to delivery time) while maintaining the famed Liberty taste and quality
of food. Within months, Vishal has started looking at franchise owned kitchens to
expand the Liberty brand.
Case Study 4: Lean Design for New
Product Manufacturing

Background

While FS (original name disguised) is a well-known toy brand in India, the domes-
tic segment is actually a small part of its operations. The major portion of its
operations is devoted to third-party manufacturing for some of the world’s leading
toy brands.
The two manufacturing units located in the western and southern parts of penin-
sular India have been in operation for more than 25 years. Both the plants were
operating pretty much at full capacities especially during the peak season. In 2017,
a US based global toy retailer approached FS to manufacture and supply a new
range of plastic moulded toys designed based on a popular cartoon show. This
range of toys were expected to sell in large volumes in the global market and was
currently catered to by suppliers in Vietnam and China. The global toy major was
now looking to develop an alternate source to counter the effects of the escalating
trade conflicts between USA and China. After several months of discussion and
negotiations, FS was finally contracted to supply the range of toys to be made in
a new factory, to be built to as per customer standards.
In a recent plant visit to the existing factories, the chairman of FS did not mince
his words on the current state of operations there. He was upset about the excessive
material transport, piles of inventories, haphazard movements, and chaotic layout
of the existing factory. His communication to the top management was clear—
unless the new facility was designed to work efficiently and look world class, he
would not give permission for the new project. The CEO was negotiating at the
time to finalize a new site which had a newly constructed but unused factory shed.
He had to now live up to the Chairman’s expectations in the design and execution
of the new factory. A chance meeting with a former employee led the CEO to be
referred to us.
The brief to us was clear—study the existing plant to understand the plastic
moulded toy operations and then work with the new product development team

© The Editor(s) (if applicable) and The Author(s), under exclusive license 271
to Springer Nature Singapore Pte Ltd. 2023
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9
272 Case Study 4: Lean Design for New Product Manufacturing

in designing and executing the layout, process flow and systems at the new fac-
tory. By the time we were engaged, the team had already completed a lot of the
preparation,

• The product development team had already started its work. They had, enabled
by the customer, visited an existing supplier site in Vietnam and understood the
manufacturing process requirements of the product range. The product range
comprised a set of 6 different toys of a single product family; each toy had its
own variations in terms of colour, features, and functionalities.
• The site with pre-existing building was finalized and all factory compliances
taken—it was now available but had to be made ready for operation since it
had been unused for a few years.
• Some of the key machinery had already been ordered based on approximate
capacity calculations done with inputs from the customer.

Now, the challenge was to get the plant operational within four months for the
pilot production run. Once the customer approved this, the bulk production would
have to start. FS management could allot only limited funds for the site work as the
company was having working capital constraints. Hence, the goal was to design
and set up the operation within the pre-existing shed with minimum infrastructural
enhancements.

The Challenge

The brief was to design a layout and implement process flows for a new product
which was being manufactured for the first time. Hence, there was no past data or
experience to help the team working on this project. What it had were a few sam-
ples of the six toy variants, some videos, and notes of the manufacturing activities
from the team’s Vietnam visit.
The team commenced the design with the available information. Firstly, the
sample toys were disassembled and put back to understand the components and
the assembly steps involved. Each of the six toy variants had two main parts—one
part was a vehicle such as a car, truck, helicopter, etc. and the second part the
rider of the vehicle. Each toy variant was unique in terms of both vehicle and rider
with different colours, designs, and functionalities. With this understanding, they
proceeded to the initial stage of process flow design.

Step I—Process Flow Design

Define Product/Service Groups Through Product—Quantity (P–Q)


Analysis Based on Projected Business
P–Q analysis helps identify runners, repeaters, and strangers which have a major
implication on process and layout design. In this case, the six toys of the
product family were analysed in terms of customer demand projections. The
peak quantities during season (Table C4.1) were considered in design capacity
computations.
Case Study 4: Lean Design for New Product Manufacturing 273

Table C4.1 P–Q analysis


S.No. Toy variant Demand Takt time (Sec)
with Takt time (for moulding)
1 Blue 4000 15.8
2 Red 4000 15.8
3 Grey 2000 31.7
4 Yellow 2000 31.7
5 Green 1500 42.2
6 Orange 1500 42.2
Total 15,000

Calculate Takt Time and Check Feasibility of Dedicated Flow Lines


Table C4.1 shows the demand data and the corresponding takt time. Moulding
was the preliminary and time-taking process, and hence was planned as a three-
shift operation. 22 planned working hours and OEE target of 80% were assumed
for the moulding. For the rest of operations (manual intensive) single shift was
planned which meant that the takt time was one-third of the moulding takt
time. The monthly demand was used to a compute the daily moulding production
requirement as shown in Table C4.1.
For example, the blue toy had a daily requirement of 4000 units.
Available time per day = 22 h @ 80% OEE = 22 × 60 × 60 × 80% = 63,360 s.
Takt time = Available time/Demand = 63,360/4000 = 15.8 s.
From the table, it was evident that there was no one runner (volume
leader) product, and in fact monthly customer demand projections showed that
the proportions of the six toys varied from month to month. Therefore, the option
of creating dedicated flow lines (for runners) was not feasible. So, the team
decided to dig deeper and identify process/resource commonalities by making use
of product-process (P-P) matrix for the entire family of toys.

Prepare Product-Process (P-P) Matrix and Group Resources


for Products with Similar Flows
The P-P matrix helps group the set of machines/workstations/processes that are
needed to make a group of products and helps identify the operating cells that can
be formed to process them. The key process flows for the components are shown
in Fig. C4.1.
The exercise now involved detailing the product-process linkage for each com-
ponent. This being a new product for FS, there was no historical process data
available. Hence, each sample toy was carefully dismantled into its constituent
parts and for each part the possible process sequence was brainstormed and iden-
tified through team discussion. The inputs of the project manager and design
member based on their visit to the Vietnam plant helped in clarifying doubts about
the process. Estimated cycle times based on either observation in Vietnam or based
on similar operations being done for other toys were then filled into an easy to
274 Case Study 4: Lean Design for New Product Manufacturing

Fig. C4.1 Typical process Molding Molding


flow chart for toy

Vehicle process
manufacturing

Rider process
Painting Painting /Stickering

Tampo Printing Sub Assemblies

Sonic Sealing Vehicle Assembly

UnitToy Blister Packing

Carton Box Packing

understand P-P matrix in visual format. The matrix for one variant (Blue Toy) is
shown in Table C4.2.
The numbers in the cells represent these estimated cycle times, and the colour
of the cell is the actual colour of the paint or design print at the respective process.
While the base colour of the toy is blue, both the vehicle and rider have designs of
different colours that are either painted using spray painting in booths or printed
using Tampo (pad) printing machines. For example, a rider component such as
head, may have brown face and black hair done through painting while the finer
work of black eye balls is easier done through printing. A paint booth can be
used for any operation or colour. Paint and mask (used to mask the not-to-be-
painted parts of the component) have to be changed and spray gun cleaned for
every component change-over. Similarly, pad printing machines can also be used
for any component by just changing the pad and colours. These cycle times helped
in the computation of number of work stations required for manual activities such
as painting, assembly, and packing and machine requirement for machine centric
processes such as moulding and Tampo printing.
Let us consider an example. As we saw in Table C4.1, the blue toy has a takt
time of 15.8 s for a three-shift operation which translates to a takt time of 5 s for
the single shift painting operation. This toy has multiple components and we take
as an example the equipment planning for the rider components named Body RHS
and Body LHS. For each of these components, the estimated cycle time for each
of the four painting operations is about 10 s which means these four operations
can be planned in a single-piece flow line comprising of four paint booths located
adjacent to each other. One line would deliver one painted component every 10 s.
Since the takt time is 5 s, we need to plan for two such lines or eight paint booths
for Body LHS and eight paint booths for Body LHS.
Case Study 4: Lean Design for New Product Manufacturing 275

In this manner, the P-P matrix gave the team complete clarity on the line bal-
ance requirements for flow and enabled the team to move on to the next step of
calculating machinery requirements.

Machinery Requirement Calculation


We take the example of injection moulding machines to illustrate machine require-
ments calculations. Moulding machine capacity required was determined based on
the mould given by the customer. For the car body component of the blue toy, daily
production requirement is 4000 units per day in a 22-h operating window, and thus
a takt time of 16 s (approximate) (Table C4.1). From the P-P matrix (Table C4.2),
the estimated cycle time for moulding car body is 40 s and the mould has two
cavities. So, 2 units per 40 seconds means an effective cycle time (ECT) of 20 s
which is about 20% more than the desired takt time of 16 s. However, the team
was confident of reducing the cycle time further along the way to meet the takt
time. Hence, one 150 T moulding machine was deemed to be adequate to meet
the demand of this component.
In a similar way, the machine requirements were calculated for all the moulded
components. A specific parameter to be kept in mind is the moulding machine
capacity (tonnes); larger machines are used for the bigger/heavier components.
After this, the parts were mapped to machines and a weekly schedule drawn up to
verify that all parts can be produced as per required quantities using the computed
machinery. Table C4.3 gives a snapshot of the first two days schedule for differ-
ent components. By the end of the week, all the components had been covered
for the week’s requirement. The best combination of machine configurations that
minimize investment was finally selected by the team. Two insights emerged from
this working.

1. Ten machines were adequate to meet the target; 150 T–—5 machines. 120 T—
3 machines and 80 T—2 machines. The layout could now be made for these
machines.
2. Up to a week’s inventory of moulded components may need to be stored in
racks, the number of racks to hold this needed to be calculated and space
provided in the layout.

Other process machinery requirements such as Tampo printing and paint booths
were also computed in a similar fashion.
The final processes were the assembly of the vehicle part of the toy, followed
by unit blister packing. Each pack contains one vehicle and one rider and these
packs are in turn packed in carton boxes for dispatch. While dismantling and
reassembling the sample toys, the team had been able to record approximate
assembly (fitting) time. The toy was designed for press fit by hand or by using
simple jigs and hand tools. In addition, there were design stickers to be hand
pasted onto the vehicle body. The team felt that since assembly and packing were
totally manual, the processes could easily be designed, implemented and fine-tuned
once the plant operations began. It was enough at this stage to just earmark the
276

Table C4.2 P-P matrix for “model blue toy”


Note: All values are in Seconds. Colors of the cell
Product - Process Matrix indicates paint color. Cell Merging implies combination
of the component
Fin.
Process ---> Moulding Decorations Component Assembly Assy. Packing

Parts
SN. Components

80T

no’s
180T
150T
120T

Ups in
Tampo
Tampo
Gluing
Snap Fit

Product Name
Blistering

Printing 1
Printing 2

Painting 1
Painting 2
Painting 3
Painting 4
Stickering
Stickering

Heat & Fix


Fix Vehicle

Tampo - BC

Glue/Sonic 1
Glue/Sonic 2
Box Forming

MH Tampo-1
MH Tampo-2
1 Car body 2 40 10 16 10 20 20
2 Chassis 4 40
3 Drive cab handrail 8 40 10
80 8 25 25
4 Outer Wheels

Vehicle
4 40
5 Keylock
6 Wheels hub 4 35 40
7 Hat 8 35 10 10
8 Head front 16 35 10 10 15 15

Model Blue Toy


9 Head rear 16 35 15 8
15

Rider
10 Body LHS 10 10 10 10 8 8
4 40 12
11 Body RHS 10 10 10 10 8
Case Study 4: Lean Design for New Product Manufacturing
Case Study 4: Lean Design for New Product Manufacturing 277

Table C4.3 Moulding machine requirements for ‘blue toy’


Sl. Moulding Day 1 Day 2
No. M/C ton Shift 1 Shift 2 Shift 3 Shift 1 Shift 2 Shift 3
1 150T Car body Car body Car body Car body Car body Car body
2 150T Chassis Chassis Chassis OW and Key OW & Key OW and Key
3 120T Drive cab Drive cab Wheels hub Wheels hub Wheels hub
handrail handrail/TC*
4 120T Blue toy body Blue toy body Blue toy body Tool change Water bomb Water
LH/RH LH/RH LH/RH and cushion bomb/cushion
5 80T Blue toy hat Blue toy Blue toy head Blue toy head Red toy hat Red toy
hat/TC front/TC rear/TC hat/TC
6 150T Car body Car body Car body
Car body Car body Car body
7 150T Chassis Chassis Chassis/TC
OWF, OWR OWF, OWR OWF, OWR
& Key & Key & Key
8 150T Llad/DecLH& Llad/DecLH& Llad/DecLH& Llad/DecLH& Llad/DecLH& Llad/DecLH&
RH/Hand RH/Hand RH/Hand RH/Hand RH/Hand RH/Hand
9 120T Whe Hub Whe Hub Whe Hub Red toy body Red toy body Red toy body
F/Whe Hub R F/Whe Hub R F/Whe Hub LH/RH LH/RH LH/RH
R/TC
10 80T Head F/R Head F/R/TC Long pin/short Rotate arm Rotate Fan blade/TC
pin/TC arm/TC

work area for assembly and packing operations as a block in the layout drawing.
Assembly would be done on tables and the number of tables; its orientation and
other aspects could be worked out during the pilot production and ramp up stages.

Process Flow Design—Flow and Pull


Now, the team was able to move on to the final step of freezing the entire process
flow design. The key question that needed to be addressed was the location of seg-
ments for continuous flow and supermarket pull. The process-wise considerations
that went into this decision are listed below:

• Constraints on investment, limited the number of moulding machines, which


meant multiple changeovers to provide all the parts needed for a toy. There was
therefore a need to produce and maintain a small batch of most parts before
doing the changeover.
• Painting is a manual operation and the cycle times are similar or adjustable
through line balancing and hence it would be easy to maintain smooth flow.
• Tampo printing, though a machine operation, is simple and cycle times easily
balanced. So, machines could be grouped into cells that can be allotted to one
or more toy variants in sequence.
• Assembly needs all the parts, i.e. entire kit for a toy to be readily available;
these parts come from multiple sources including moulding, painting, printing
and directly from stores as well. Hence, it was necessary to have a supermarket
from which compete kits can be issued to the assembly lines.
• Packing was a manual process that could be easily linked on line with assembly
lines and hence it was proposed to have continuous flow
278 Case Study 4: Lean Design for New Product Manufacturing

With these considerations, the rough future state VSM was drawn as shown in
Fig. C4.2.
Two supermarkets were defined in future state with the following rules:

Super market 1—Two days inventory of moulded components based on


weekly customer order schedule. As parts are pulled by the paint-
ing/printing/subassembly operations, the racks are refilled with the next item
in the schedule—this is a sequenced pull situation.
Super Market–2—Supermarket for kitting—all parts required for the toy are
placed one day prior so that assembly goes ahead as per schedule.

Since there is continuous flow from assembly onwards, the daily schedule is
planned at this “pacemaker”. Supermarket 2 triggers the production of upstream
processes.

Step 2—Freezing the Layout

Now that the process flow had been defined and the type and number of
machines/workstations for each operation fixed, the team then moved on to freez-
ing the layout design. Since this was a new facility, the team had the advantage
of starting with a “blank slate”—a built up but empty shed and the land around
it. The goal was to arrive at the best possible physical material flows through the
plant in accordance with the process flow design. This meant that the following
aspects needed to be kept in mind while making the layout:

• Minimum material transport distance,


• Minimum material handling—number of touches,
• Sufficient storage facility for peak requirement—raw material and finished
goods,
• Incorporate global customer specifications and requirements.

Inspect Physical Location (Built up Area/Land) on Site


The team arrived at the site to examine the site’s condition and noted the actual
dimensions, positions of doors, windows, and entry and exit points. The location
for provisions for electricity, water source and existing drainage were all marked
on the layout drawing. In addition, office block, land slope, and height with respect
to the shutters were all marked.

Compute Space Requirement


After making the calculation for all the facilities required, the space occupied by
the machines and workstations were calculated. To calculate the space the machine
foot print was taken (the machine size along with the working clearance). Since
most of the machines were still on order, the suppliers were contacted, and this
Weekly Schedule Daily Schedule

Painting Super Market 2


Super Market 1

Assembly and Secondary Packing


Moulding Printing Unit Packing (Carton Box)
Case Study 4: Lean Design for New Product Manufacturing

Sub-
Assemblies

Fig. C4.2 Rough future state VSM


279
280 Case Study 4: Lean Design for New Product Manufacturing

data was collected from them. The calculation for raw material stores and the
Finished Goods warehouse involved computing maximum stock levels for all the
major items and computing the space requirement considering effective use of
vertical space. The rack space for the two WIP supermarkets was also factored in
to the layout.

Finalize Layout Drawing


The design engineer who was part of the team was then tasked with preparing the
first cut of the layout using AutoCAD software. He took into account the following
inputs for this drawing:

1. Process flow design as per Step 1,


2. Physical shed measurements and details noted during site visit,
3. Special factors to be considered for adherence to factory laws and customer
requirements.

Ultimately, the layout underwent multiple iterations as successive brainstorming


sessions brought up alternate options that fine-tuned previously suggested layouts.
After eight revisions, the final agreed layout was then taken up by the plant man-
ager and project manager to plan the required civil works for the machines to be
installed as and when they arrive. Some of the specific considerations that went
into the layout finalization are:

• Dust and fume extractor set-up needed for painting booths—hence, the paint
booths needed to be close to the building wall to connect to the extractor
outside.
• One operator to manage two injection moulding machines influenced the
parallel placement of these machines,
• Multiple assembly lines to assemble a mix of toy variants feeding a common
secondary carton packing line,
• Cell grouping for the Tampo printing section.

The finalized layout with overall broad material flow is shown in Fig. C4.3.

Step 3: Physical Implementation of the Designed Process Flow


and Layout

Physical Placement of Machines


As the machines arrived, they were put in position in accordance with the final-
ized layout drawing and all the required auxiliaries—electrical, pneumatic and
hydraulic, water inlets and drains to enable operation of the same. The stores and
warehouse areas were cordoned off with fences and storage racks were put in
place. The assembly tables were also placed in the allocated location.
Case Study 4: Lean Design for New Product Manufacturing 281

Fig. C4.3 Final layout

Pilot Production—Manufacturing Design


Initially, the customer had asked for a sample of about 50 numbers of each toy
variant to be made, packed and sent to them for manufacturing approval. This was
the first time the product was actually being made at FS and the team used this
opportunity to observe and understand the Muda, Muri, and Mura in assembly and
packing. As we saw earlier, the team had decided to finalize the assembly and
packing process layout at this stage.
Operations such as stickering, wheel assembly, top assembly, and packing were
observed in detail. Some operations such as wheel and hub cap fixing were con-
verted into offline sub assembly operations which could feed all the assembly lines.
Each assembly workstation was fixed including positioning of jig/fixture, compo-
nent arrangement, tools, input and output placement and the operators were made
to practice during the trial production stage.
The biggest constraint was in the blister packing of the toy—fitting the toy into
the pack and tightening the “locks” was a strenuous job. Since the locks were
located behind the pack, the fingers had to do this job by feel, resulting in high
cycle times that varied from toy to toy.
As a remedy, a wooden fixture was developed where the pack is placed, toy
fitted in position with the locks facing up to the operator giving her the visibility to
complete the packing without any unwanted motion and strain. Cycle time could
thereby be reduced to match the assembly time resulting in a balanced line.
The final assembly and unit packing line was a waist height work table with
six workstations. The number of workstations in use depended on the toy variant;
282 Case Study 4: Lean Design for New Product Manufacturing

some vehicles had more complexity and utilized all the stations while others could
operate with three or four workstations. Each workstation was designed to accom-
modate all required components in plastic bins in front of the operator and the
extra inventories stored below the table to be used for replenishment of the bins.
All the people working in this plant were newly recruited as this was the first
such factory in the area. Women who comprised the majority of the workforce
were from nearby villages and were working in a manufacturing industry for the
first time. They all had to be trained on the operations such as painting, printing,
and assembly from ground-up.

Ramp Up Stage
Within two months of pilot production, the plant had reached a production level of
6000 toys/day. The process and plant had been designed for 15,000 toys/day and
the demand from the customer was picking up and there was an urgent need to
ramp up and standardize the operations. By this time, the factory team including
supervisors, engineers, planning, stores, and logistics executives was in place. They
were to be responsible for the day-to-day operations of the factory.
At this stage, we conducted a Lean workshop for the newly assembled factory
team, focusing on observing the running process to identify areas for improvement.
Three-member cross-functional teams were formed for the moulding, painting,
printing, and assembly processes. Key observations were shared as below:

• Process was designed with two days of supermarket inventories after moulding
but more than a week’s WIP was actually found in the racks spilling over onto
the floor while some of the items in the running schedule were in shortage
leading to incomplete assembly kits.
• The painting process was just about able to make parts for about 6000 toys per
day and this was therefore identified as the bottleneck to achieve the ramp up
target.
• WIP of parts seen in trays all over the printing area—while the scheduled
production was running in a hand-to-mouth situation.

Based on these observations, improvement projects were identified in all the four
major processes and were executed within the next couple of months.

Improvement Projects

Project 1: Line Flow in Painting Process


At the initial process design stage, the process route for each component had been
defined in the P-P matrix based on available data from the other manufactur-
ing location. But in the pilot production phase, several parts underwent process
changes—for example, a part originally designed for printing was now undergo-
ing painting. The team also observed multiple handling of painted components
between one stage and the next as the allocation of paint booths as per process
flow had not yet been finalized.
Case Study 4: Lean Design for New Product Manufacturing 283

The team first established this line flow for all toy parts. For each part, there
are multiple painting operations which were allotted to adjacent booths to support
single-piece flow. All the operation cycle times were studied and booth allotment
done based on line balancing principles so that the parts can flow through all the
stages as shown in in Fig. C4.4.
The line flow was maintained for a month as per the booth allocation. However,
the team recorded that even at the maximum cycle time of 15 s, instead of an
output of 1680 parts, the actual output was only 1200–1300 parts per shift. Further
observation showed that each booth would frequently stop painting operations for
mask cleaning. The components were covered by masks so that only the surface
area to be painted is exposed. After every few components, the masks need to
be cleaned of the paint build-up to avoid quality issues. To minimize this loss,
spare masks were purchased and a mizusumashi (material feeder) was allotted for
material feeding and mask cleaning. Every 15 min, she exchanges all the used
masks for cleaned masks, brings them to the cleaning station, and cleans them.
With these improvements and extending painting into a two-shift operation, the
capacity went up to match the target requirements.

Project 2: Cell Formation—Printing and Sealing


The painting and printing sections were linked through a small rack containing
painted components. Quality cleared painted components were placed in small
trays and kept in this rack from where the concerned printing cell members pick
it up for their work. However, several trays of partially processed parts were also
observed on the floor which pointed to a lack of flow in the printing section.
The rider toy parts needed multiple printing operations, 2 stages of sonic sealing

Operation 1 Operation 2 Operation 3 Operation 4

Cycle 12 secs 28 secs 14 secs 31 secs


time

1
1
Booths
allocation 1 1

1 1

Effective
12 secs 14 secs 14 secs 15 secs
cycle time

Fig. C4.4 Booth allocation and flow balancing in painting process


284 Case Study 4: Lean Design for New Product Manufacturing

and a glue-based head fixing operation before the rider is complete and ready for
packing.
A product-process matrix was made for this section for the six toy variants and
two identical cells were formed based on similarity and quantities required. Each
cell comprised of Tampo printing, sonic sealing and glue fixing workstations. The
cells were handling the rider part of the toy. Each rider consisted of two body parts
(left hand side and right-hand side), two head parts (front and back) and a cap or
hat. The cells were run in flow as follows.

Cell 1: Body parts LHS and RHS run in parallel and joined together in sonic
sealing process.
Cell 2: Head front and back processed.

Head and body attached in second sonic sealing stage; the cap glued on in the
final station.
Since each cell would handle multiple variants within a day, the schedules
were made to ensure flow and completing the required quantities within the day.
Fig. C4.5 shows the cell layout.

Project 3: Mixed-Model SKU Assembly and Packing


The original process design envisaged the final carton box pack to contain the toy
variants in the approximate ratio of the monthly requirement. As per the original
P–Q table, this ratio was blue and red 2 each, grey and yellow 1 each, green and
orange 0.75 each. But as the plant operations gained momentum, it was realized
that customer orders vary widely in terms of final pack size. Each order would have
a different combination of the 6 toy variants to be packed into a single carton. The
carton sizes also varied as can be seen from the order position for a particular
week shown in Table C4.4.

Rider Glue
Station

Body+Head Cap
Sealing 2

Single Single
Sonic

Head Head
Printer 1 RHS Printer 1
Printer 2
Printer 2

Single
Single

Head
Head

Body
Rider Head Cell
Rider Body Cell
LHS
Sealing 1
Sonic

Head
Printer 1 Printer 2 Printer 4 Printer 3
Multi Head Multi Head Multi Head Multi Head

Fig. C4.5 Cellular layout at printing


Case Study 4: Lean Design for New Product Manufacturing 285

Table C4.4 Customer orders in a typical week


PO No Toys per carton Pack ratio Order size Line allotment No. of days run
(cartons) (No. of Lines)
1 6 Blue—2, 15,000 Blue -2, Red-2. 5
Red—2, Grey-1, Yellow
Grey—1, -1
Yellow—1
2 3 Green—3 3000 Green -1 6
3 4 Orange—2, 4500 Orange -1, 6
Yellow—2 Yellow -1

The assembly and unit packing lines were standardized for flow, and worksta-
tion arrangement and quality checks were such that a stable 1500 units were being
produced per day per line. The layout was made to provide for 10 such lines—5
on each side of a motorized belt conveyor. These lines were flexible and any line
could take up any toy within half a days’ notice. To meet the demand as shown in
Table C4.1, nine lines were allotted accordingly. Blue and red toys had a capacity
of 3000 toys per line (twice as much as the other variants) as they had less oper-
ations. Hence as per the schedule above, all three POs would be running at the
same time.
Each assembly line would deliver a blister packed toy onto the moving belt
conveyor and these unit packs were being placed into the carton box by a packing
team positioned at the end of the conveyor. However, while the assembly lines were
well set, the team observed a lot of stress and strain for the carton box packing
operators. They were placing the unit packs again on holding tables, where differ-
ent variants were getting mixed up and the conveyor had to be stopped repeatedly
to clear the backlog. Operators were moving around trying to grab the required
toy unit pack to be put into the carton box as per the packing list.
A separator table was created—smooth laminated top on which the unit packs
could be separated and placed without damage and this top was at an angle so that
unit packs slid down to the end with a slight push. Fig. C4.6 shows this separator
table.
Aeroplane wing layout was created at the carton packing zone at the end of the
conveyor to ensure smooth packing operation with minimum operator strain due
to movement, reduce number of touches of the unit packs and continuous flow
of packing, final quality clearance and move to the FG warehouse. One operator
stationed at the end of the conveyor picks up the unit pack and places it in the
allotted row in the separator table. Packing operators stand along the table with
the carton box positioned on a stand below the table; they pick up the required
unit packs and place them in the carton boxes, seal the completed box and place
it on the adjacent pallet. The modified assembly and packing layout is as shown
in Fig. C4.7.
286 Case Study 4: Lean Design for New Product Manufacturing

Fig. C4.6 Separator table at carton packing

Project 4: Standardization through Lean Scheduling Model


Once the complete physical material flow had been set in accordance with the
process design, the team started working on finalizing the complete planning sys-
tem. In designing the Lean information flows, pull-based scheduling were put in
place. The scheduling philosophy was nicely captured in the slogan, “No Kit No
Cut”, meaning if the assembly kit is incomplete the assembly lines would not run
putting pressure on the upstream processes such as moulding to make only what
is needed, and in the quantity needed so as to fulfil the kits demand. The planning
model is shown in Fig. C4.8.
The customer provides a three-month rolling demand forecast based on which a
monthly plan is made for SKU-wise quantities for the next month. This activity is
completed in the first week of the month and purchase orders/intimation placed to
suppliers. While imported and smaller materials are procured by the 25th, packing
materials which are bulky are procured in two lots—first by the 25th of the current
month and second by the 10th of the next month.
The month’s plan is converted to 4 weeks rolling production plan by the 25th,
i.e. one week before and this incorporates any changes by customer or backlogs
expected from current month. Every Wednesday, a firm production schedule for
the next week is made for the pacemaker (the assembly process). Information
on bought out parts, packing materials and raw materials is taken and only those
pack combinations for which entire “kit” of materials is available are scheduled. A
schedule given to assembly for D-date will translate to the painting schedule of
D-1 date and the moulding schedule for D-3 date.
All finished goods (FG) are quality approved online and transferred to the FG
warehouse and orders are dispatched from time to time as per customer schedule
confirmations. To facilitate monitoring of operations and throw up deviations in the
Case Study 4: Lean Design for New Product Manufacturing 287

Parts Kit
Assembly Line 1 Assembly Line 6

Assembly Line 2 Assembly Line 7


C
O
Assembly Line 3 N Assembly Line 8
V
Assembly Line 4 E
Assembly Line 9
Y
O
Assembly Line 5 R Assembly Line 10

Pallet 1
Pallet 1
Pallet 2
Pallet 2

TO FG WAREHOUSE
BY PALLET TRUCK

Fig. C4.7 Assembly and packing layout and material flow

process/plan for immediate action, several visual management tools were deployed
both on the Gemba and electronically. These include an hourly production plan vs
achieved board at each process, an assembly kit status board for each customer
PO and an electronic raw materials kit status in a spreadsheet used by the planner.

Outcomes of the Project

FS had to design, execute, and run a completely new factory with a product variant
that was completely new to them. This began with the initial process design and
layout finalization in September of 2017. While this was taking place, the company
was working on completing the site acquisition and legal registration process in
parallel. Civil works began from end of 2017 and machines started arriving from
the New Year. By April 2018, most of the machinery and facility was in place.
The next six months were spent in implementing the layout and ensuring ramp up
288 Case Study 4: Lean Design for New Product Manufacturing

Monthly Plan
5th of previous
month
Weekly Rolling
Plan – 4 weeks

By 25th of
previous month Suppliers Next week firm
schedule

FG to Customer

By 10th of current month

Moulding Assembly &


Painting
D-3 Packing
D-1 Kit
FG
D Day
2 Days
Buffer

Fig. C4.8 Final planning model

Oct ’18
Sep ’18 Synchronize-
12,000 units/day
Jul ’18 Standard
process –
Jun ’18 Process
finetuning 9,000
May ’18 Trial lot – – 4,000 units/day
1,000 units/day
Pilot units/day
production
and process
flow design

Fig. C4.9 Fast ramp up for from project execution to full production output

to the designed capability levels. Figure C4.9 shows the progression of this phase
and the overall outcomes of this Lean design and implementation project.
Following are some of the highlights of this project.

• Consistent output of 10,000 + units per day is being achieved from November
2018 and comfortably meeting the customer peak season requirement.
• World class visually managed facility meeting global customer standards.
• Pull-based planning system enabled and line flow operations ensure that WIP
is kept to a maximum of three days.
Case Study 4: Lean Design for New Product Manufacturing 289

• Quick ramp up, streamlined operations, and quality standards being met have
made the customer happy, and they have already given the go ahead for another
set of new products requiring further expansion.
Case Study 5: Lean at Gubba Cold
Storage

Background

Gubba, though an SME, happens to be India’s largest cold storage company in


terms of capacity. The company has multiple cold storage facilities (called plants)
spread out cross the outskirts of Hyderabad, India. With seed cultivating regions
and major pharmaceutical manufacturing in and around Hyderabad, the company’s
focus on seeds, pharmaceuticals, and food products seems to be well placed.
The earlier generation of Gubba family was involved in food products and
traded tamarind. With the opening of their first cold storage facility in 2000, Gubba
has been on a continuous growth trajectory led by the second-generation family
members, brothers Kiran and Prashanth. Elder brother Kiran takes care of the
finance, sales, HR and administrative functions and Prashanth handles the core
operations, technology, and projects.
Gubba has always been on the forefront of technology, processes and systems.
In their early years itself, they internally developed their own software application
for managing the entire operations such as, truck loading/unloading, storage and
invoicing. Having global agri-business giants including Bayer, Syngenta, and Pio-
neer Seeds as customers have driven Gubba to implement the best practices for
safety, inventory management, pest control, and plant operations and maintenance.
By 2012, the company was a leader in the cold storage industry and Kiran and
Prashanth were on the lookout for newer opportunities to grow. Is there anything
more they could do to add value to their customers while improving business and
profitability? It was at this point in time, that Prashanth had the occasion to meet
one of his friends Ashish, who was working in his own family-owned enterprise,
manufacturing deep freezers and bottle coolers. Ashish spoke to Prashanth about
how his factory had been transformed through a Lean initiative leading to a huge
increase in output, turnover, and profitability. This struck a chord with Prashanth,
who took the initiative to contact the same Lean expert and invite him to Gubba
for a discussion.
And soon after Gubba began its Lean journey under our guidance!

© The Editor(s) (if applicable) and The Author(s), under exclusive license 291
to Springer Nature Singapore Pte Ltd. 2023
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9
292 Case Study 5: Lean at Gubba Cold Storage

Operations

The seed industry is a seasonal industry with a significant time-gap between seed
production and sale to farmers who plant the seeds during specific times of the
year. Seeds have a tendency to germinate in a warm and humid atmosphere; hence,
the need to preserve these seeds in a low temperature, low humidity storage facility
in the interim period. The seed companies therefore often use contracted cold
storage facilities in order to store the seeds either closer to the markets or to
production facilities. The logistics are handled by the seed companies themselves;
after grading the seed in their plants, the seeds are packed in either jute bags (50 kg
per bag) or large jumbo bags (1–1.5 metric tonnes per bag), loaded onto trucks and
sent to Gubba. At Gubba, these bags are unloaded and stored in their cold storage
plants. Whenever the demand for seeds picks up, the seed companies communicate
the details of seed variety and quantity to Gubba and send their trucks to collect
these. Thus, Gubba’s services cover storage and preservation of the seeds.
Gubba operates three types of facilities—conventional storage, racked storage,
and a standalone frozen foods storage plant. In conventional storage, the jute bags
are manually unloaded, moved, and placed at the storage location within the ware-
house using elevators (chain lifts) to access other floors. Jumbo bags are handled
through forklifts. In racked storage, the bags are placed on pallets and the pallet
put into a rack using a combination of forklifts, pallet trucks, and sleeves on rails.
In recent years, the racked storage has been upgraded to handle pharmaceutical
products which have more stringent storage and handling norms. The frozen food
storage plant is a separate unit, and is not discussed in this case.
To avoid any discrepancy between what the customer has sent and what is
received at Gubba and vice versa, an assistant supervisor is deployed during truck
unloading and loading to count the bags and tick them off against the delivery chal-
lan (manifest or packing list). This cross-check is meant to verify the quantities
of seed varieties based on the following three parameters.

1. Seed Type: For example, paddy (rice), cotton, wheat.


2. Hybrid: Within a type, say paddy, there are multiple hybrids, the hybrid is a
specific variant developed by the company
3. Lot Number: The same hybrid can be produced at different locations/times and
are defined by specific lot numbers.

Typical organization structure at each of the Gubba plants is shown in Fig. C5.1.
Contract labour gangs are employed to load and unload bags from the truck
in the case of conventional cold storages while material handling equipment is
deployed to handle palletized goods and jumbo bags.
Each plant can handle 2~3 trucks at a time and the assistant supervisors are
deployed there for monitoring the loading, storage and unloading operations.
The computer operator handles documentation and ERP entries from the office
located at the entrance to the plant. The plant supervisor handles customer inter-
actions with respect to material arrival, requirements and is responsible for overall
upkeep of the plant and maintaining the storage conditions in conjunction with the
Case Study 5: Lean at Gubba Cold Storage 293

Fig. C5.1 Plant organization


structure Plant Supervisor

Computer
Operator

Assistant Assistant Operators


Supervisor Supervisor (Forklift etc)

refrigeration team. Each plant or group of plants have a refrigeration unit for main-
taining plant temperature between 12 ~14 °C. The refrigeration unit is handled by
a technician who reports to a Technical Head overseeing all the Gubba plants.

The Lean Journey

We know that Lean espouses maximizing value addition which is defined as


the transformation of material into a product or information into a service. Hence,
in manufacturing industries the stores and warehouse are considered as support
functions and non-value adding as such. How then do you implement Lean in a
cold storage facility which is essentially a warehouse for storing material? We do
this by considering the core purpose of storage as value adding.
Lean philosophy talks about“Doing more and more with less and less”; more
output with lesser use of resources. Applying this philosophy, Gubba’s opera-
tional performance is measured in terms of storage capacity utilization and ability
to handle customer receipts and dispatches in the shortest possible time. From
the resource utilization angle, storage space, trucks and people were the main
resources. Considering these factors, the team set the following goals for Lean:

1. Maximum utilization of warehouse space—This has a direct impact on the busi-


ness profitability as the main cost of storage is refrigeration, which is directly
proportional to the extent of space refrigerated.
2. Minimum Truck Turnaround Time—inward and outward—This enhances cus-
tomer satisfaction as each truck is used for multiple trips on a given day.
3. Maximize people productivity—Personnel cost is a significant component for
both Gubba and the customers who are charged for this.

Lean was initiated towards the end of 2013 with an initial four-month burst of
improvement activities between December and March. Since seed arrivals and dis-
patches are at a peak from March–July, the Gubba team validated the results of
the Lean initiatives taken during this peak time. Having done that, the stabiliza-
tion phase kicked in and continued till the end of the year through development
294 Case Study 5: Lean at Gubba Cold Storage

and implementation of standard operating practices (SOPs). In 2015, a core


cross-functional team was formed to audit the adherence to these SOPs and high-
light deviations. Repeated deviations were analysed and solutions implemented to
improve these practices further. This was the path adopted to sustenance of Lean
practices.
Another four years later, Gubba reconnected with us to look at taking Lean
to the next level. In the interim period, a new customized ERP had been devel-
oped and put in use; this software incorporated critical points of the SOPs in
an error-proofing mode which automatically ensured compliance. The booster
Lean intervention of 2019 focused on the corporate functions including purchase,
accounts, sales, HR, and administration while relooking at the information flow
at the plants. More details on each phase of Lean are discussed in the following
sections.

Initiating Lean Through Focused Improvement Workshops

The plant operations team was entirely composed of young males aged below
30 years, most of them were high school drop-outs and a very few were gradu-
ates. However, their first and only job till date was with Gubba and hence already
were quite fixed in their outlook to work. For Lean to take hold it was therefore
essential to break the team’s complacency and charge them up. Initially, one plant
of each type (conventional/racked storage) was selected and teams were formed to
focus on the two primary goals—truck turnaround and space utilization. A series
of three-day focused improvement workshops were conducted towards this objec-
tive. During the workshops, cross-functional teams were formed to take up specific
improvement projects across the various plants. The team leader would be the
supervisor of the plant, team members would include an assistant supervisor, com-
puter operator from the same plant, third from another plant and a fourth member
from the corporate team (safety office, quality, systems, etc.). These workshops
are akin to Kaizen events; every morning we would conduct a training session;
the teams would then work on the gemba the rest of the day and get together
again at the end of the day to share their experiences. Prashanth involved himself
fully for all the workshop days, sending a clear message to the employees on the
seriousness with which Gubba is taking this Lean initiative.
On the first day, the Lean expert gave the teams a basic orientation on how
to observe the process under Lean paradigms in order to identify Muda (waste or
non-value added), Muri (strain), and Mura (inconsistencies/variations). The teams
then spent the day observing the process and came out with a list of observations
and opportunities for improvement therein. These were shared in the evening with
the rest of the teams and Prashanth, and actionable points were decided.
Actions involving change in practices or methods of doing work, some modifi-
cations in equipment/tools to help workflow and reduce strain were implemented
on the second day. A few strategic decisions were taken on how the process should
be run and the new ways were tried out.
Case Study 5: Lean at Gubba Cold Storage 295

On the third day, the same process was observed to verify and validate the
impact of changes made and results measured. A fortnightly action plan was
prepared to complete the balance action items and these were monitored till com-
pletion on a daily basis at the plant level and a weekly basis at the management
level.
We look at some of the key projects that were taken up.

Improve Space Utilization

For a cold storage facility, value addition comes through space while the biggest
cost is also incurred in keeping that space at a specific temperature. At the outset,
we drove one simple Lean paradigm repeatedly into the minds of the plant opera-
tions team “Cooling is meant for the seeds”. So, any space in the facility occupied
by other things or lying empty is non-value adding from this perspective. It meant
money is spent on refrigerating unwanted items or simply cooling empty space.
In the conventional (non-rack) storage, the seed containing jute bags are layered
akin to bricks; each layer has bags oriented at right angles to the adjoining layers. If
the first layer has bags placed length-wise, the next layer on top of it will have the
bags width-wise and so on. Each such stack, called a thappi in the local language,
is formed in a bay (a rectangular area marked by pillars and walls). Each storage
bay is numbered in alpha-numeric term, and the material movement pathways are
depicted with dotted arrows in the top view of a typical floor in Fig. C5.2. A
typical conventional storage plant has about five such floors.
The first thing the team observed was the uneven heights of seed bag thappis
in different storage bays, resulting in unutilized space above some of the thappis.

A1 A2
Elevator

B1 B2

C1 C2

Fig. C5.2 Top view of storage layout in a typical floor


296 Case Study 5: Lean at Gubba Cold Storage

One reason for this was to keep different lots or hybrids in separate thappis. While
this practice facilitated easy and direct access a given lot (during retrieval or stock
checking), the loss of cold storage space had a much bigger cost implication. A
decision was taken to store different lots or seed hybrids within a bay by sepa-
rating the last layer of one lot from the first layer of the next lot with a coloured
tarpaulin sheet. Each lot already had a stack card (similar to a store bin card) while
the bags were also marked with the lot number. This arrangement necessitated
extra internal shifting (discussed in the next section) which involved additional
labour cost but this expense was far less than the refrigeration cost of unutilized
space. Besides, the internal shifting work could be scheduled during idle times of
hamalis (contract labour). Thappis could now be stacked to the maximum height
and a level indicator was painted on each pillar serving as a guide to the hamalis
who manually stacked the bags.
In the racked storage, the team counted a total of 4400 pallet slots available per
cold storage plant. Additionally, the average weight of the loaded pallet was only
1.3 MT against the designed capacity of 1.5 MT implying 13% underutilization.
The main reason for this was the difficulty in stacking the regular bags on the pallet
for bulky (less dense) seeds like cotton. As the bags are not compact, the stacks
reach the maximum height within the slot before the weight limit is reached. As
the racked storage facility was dedicated to a few key large customers, Gubba was
able to convince them to send the seed in jumbo bags of 1.5 MT capacity which
could be easily accommodated in the racked slots. At the ground level of the rack,
there was no such weight restriction. These slots were therefore allotted to pallets
and bags weighing more than 1.5 MT.
Like the conventional storage, 20% of the racked slots were only partially filled,
to keep customer orders separate. A similar policy of segregating customer lots
with a separator and storing another lot in a different pallet on top of the existing
pallet was implemented.
Within a couple of months of implementing these improvements, Gubba had
increased the racked storage capacity from 1.3 to 1.5 MT/slot and overall space
utilization in the storage facility by 7 ~ 10%. The resultant reduction of refrigera-
tion cost per tonne directly boosted up the bottom line as customers pay for storage
on a per tonne basis.

Reduce Truck Turnaround Time

Trucks meant to drop or to pick up seed are arranged by the customers whose
plants are generally located within a two-hour radius of Gubba plants. The cus-
tomers would like to utilize the trucks for multiple trips in a day to optimize costs.
From this perspective, the sooner trucks are turned around at Gubba, the faster
they can get back to the customer plant for another trip.
A cross-functional team was formed to observe the end-to-end process of “Gate-
In” to “Gate-Out”; this process determines the overall time the truck is held up at
Gubba. The essential activites in truck operations are physically placing the bags
Case Study 5: Lean at Gubba Cold Storage 297

into the truck (if loading) or lifting bags out of the truck (if unloading). Every
other activity can be categorized as unnecessary Muda, and contributes to truck
waiting. Hence, the team’s observed the entire process under the paradigm of zero
waiting time of truck.
The team categorized their observations into three buckets and identified
and implemented solutions in each bucket separately.

Gate-In to Docking
Trucks arrive at gate and wait. At Gubba’s largest campus (Yellampet) consisting
of six plants, the security at the main gate first checks which plant the arriv-
ing truck has to be sent to. At the docking point of the plant, if other trucks were
already under process, then the newly arrived truck is sent to a parking area to wait
its turn. Finally, when the truck is able to dock, the driver hand’s over the delivery
challans (manifest) to the assistant supervisor who then goes into the office and
ask’s the computer operator to check the location(s) of the seeds to be loaded (for
dispatch) or empty location(s) where incoming seeds are to be unloaded. Through
this period, the truck would be waiting. The computer operator then generates an
inward or an outward memo sheet with the location(s) information and gives this
along with a manual tally sheet (for cross-checking the bag count) to the assistant
supervisor. The assistant supervisor proceeds to the docking point to commence
the process. Hamalis (contract labour) are called and allotted positions to start the
loading/unloading process. After observation and analysis, the team concluded that
the major reason for truck waiting was that this entire preparation process prior to
actual loading/unloading began after the truck has docked. And the root cause for
this is the uninformedarrivals of trucks from customer locations.
After a brainstorming session focused at finding solutions to this problem, facil-
itated by Prashanth, the team decided to implement a system of “Pre-Alerts” where
the plant supervisors would interact with their regular customers every evening to
get information about the trucks expected on the next day. The supervisors were
worried that implementing this radical new step would spoil their relationship with
the customers. Over the years, customers were used to just sending the trucks and
then pressurizing Gubba to quickly send them back. How would some of these
global and local giants react to a small organization like Gubba asking them for
prior information? To overcome this initial hesitancy, Prashanth and Kiran took
the lead in communicating about the new “Pre-Alerts” to all the customers. The
communication focused on how the customer would benefit through faster truck
turnaround just by providing this basic information in advance. Already customers
were sending a copy of their requirements by email. If truck arrival details are
also known, the supervisor could plan labour, material handling equipment and
advance internal shifting of seeds inside the plant.
Finally, a pre-alert form was introduced. The supervisor would speak to all
major customers daily and fill up the details in this form. Based on the truck ETA,
he would instruct assistant supervisors to prepare in advance. One of the activities
of preparation is internal shifting. In many instances, different hybrids and lots are
stored in the same bay, with bags stacked one above the other to maximize space
298 Case Study 5: Lean at Gubba Cold Storage

utilization. Internal shifting involved removing the lots on top in order to access the
required lots from below. In some cases, the hamalis would also remove and keep
the required seed bags in the anteroom to facilitate loading. Similarly, if incoming
material was expected the next day, internal shifting would clear suitable area for
the bags of incoming seed to be stored.
In the first few weeks, less than 25% of the trucks were pre-alerted. The team
patiently kept sensitizing the customers and some of the customers started noticing
the improvement in their truck turnaround for pre-alerted trucks. As the season
wore on, pre-alerts were up to 70% which meant at least 7 out of 10 arriving
truck details were already known. In the last couple of years, the pre-alert sheet
has been upgraded to a mobile application linked to the rest of the ERP. Plans are
afoot to have these apps installed at the customer logistics team system so that
information is fed once by the customers, and gets transmitted instantly to Gubba,
without errors.

Actual Loading and Unloading Process


The team now moved on to observe the actual loading and loading process from
the perspective of zero waiting. They recorded instances when no activity was
happening while the truck was docked on the bay. Further analysis yielded the
following major reasons for this waiting:

(1) Delay in either locating the material (for loading) or empty space (for stor-
ing) and the need for internal shifting to continue the job. The team could
address this issue through pre-alerts as we saw in the previous section.
(2) Short stops of the elevator as the hamalis inside the store area are unable to
offload the bags in-time during unloading.
(3) Gaps or empty platforms in the moving elevator when the hamalis are unable
to place bags on each platform in time during loading.

To understand the short stops and gaps, we need to understand the layout of the
warehouse. Each warehouse has five storeys including a basement level. The truck
docks at ground floor. During unloading, the bags are lifted from the truck by
hamalis and carried through an anteroom to the motorized chain-link elevator
which operates at a fixed speed. Each bag is placed on a small platform attached to
the chain as it is moving and the same bag is removed at the allotted floor (level)
by another set of hamalis. Each hamali carries one bag at a time to the storage
point, places it in neat stacks (thappis) and walks back to the elevator to pick up
the another bag. The same process is done in reverse in the case of a truck being
loaded. Figure C5.3 depicts the elevator layout in the plant.
While the distance from the truck to the elevator is fixed, the distance from
the elevator to the storage location varies depending on the storage point (see
Figs. C5.2 and C5.3). Therefore, the number of hamalis needed for continuous
flow of material on the storage floor varies depending on the distance moved. The
short stops were mainly cases where the flow rate was out of sync. The elevator
moves at a fix speed and the seed bags placed on each elevator plate reaches the
Case Study 5: Lean at Gubba Cold Storage 299

Elevator

Level 3

Level 2

Level 1

Ante-room C
Ground

Truck

Basement

Fig. C5.3 Elevator layout

unloading floor platform within a few seconds. If a hamali is not available to take
off the bag, the elevator is stopped by the unloader and restarted only when a
hamali arrives to take off the bag.
A basic time study helped the team note the time taken to and from different
points in the floor. Based on this, the number of hamalis required to synchronize
with the elevator speed was calculated. This was verified by allotting the ’calcu-
lated’ number of hamalis for a few trucks and observing again to see if there are
short stoppages or empty plates (on the elevator). Once the team had verified the
calculations, a standard was fixed for number of hamalis to be allotted for three
different ranges of storage locations. This standard now serves as a guide to the
assistant supervisors enabling them to efficiently allot hamalis.

Documentation and Dispatch of Truck


To ensure no discrepancy between customer and their own material count, Gubba
had introduced tally sheets. The delivery challan (DC) brought by the truck driver
has the details of seed lots. The assistant supervisor would copy the lot numbers on
to the tally sheet and then maintain count against each lot received or dispatched
on this tally sheet. These details would be then added up and sent to the computer
operator who would then make an inward/outward memo which served as a gate
pass for the truck to leave Gubba premises.
The team observed that anywhere from 10 to 20 min were lost in this documen-
tation procedure. During this time, the truck continued to wait at the dock. The
process flow chart in Fig. C5.4 a depicts the step-by-step existing procedure.
The team brainstormed and implemented a revised process that, with a slight
modification in the software, made the computer operator’s role redundant thus
reducing document hand-offs (Fig. C5.4b).
Minimizing the documentation, need for multiple printouts and simplifying the
inward memo and gate pass helped the team reduce the truck waiting time.
300

(a) Earlier State


Truck
arrives at
Driver prepares
docking
truck for loading Supervisor He prepares
point
/ unloading hands over memo ( gate prints and Truck
Loading or unloading with documents to pass)by hands copy leaves
continuous count check using computer entering all the to driver docking
tally sheet. operator
Driver hands DC details in point
Supervisor takes
over documents Supervisor checks the software
tally sheet and
to assistant storage location assigns hamalis
supervisor

0 2 7 10 45 62 65

Draft (b) After Improvement


Truck arrives Driver prepares memo
at docking truck for loading confirmed
point / unloading in
software
Loading or unloading with based
Supervisor Supervisor continuous count check
Driver hands
prepares draft assigns hamalis using draft memo prints and Truck
over documents
to assistant memo that has as per location hands copy to leaves
supervisor storage distance driver docking
location template point
35 40
2 10 42
0 7

Fig. C5.4 Earlier and improved truck turnaround process


Case Study 5: Lean at Gubba Cold Storage
Case Study 5: Lean at Gubba Cold Storage 301

The Result
Through all these improvements, the average TAT for a 10-tonne truck was reduced
from 70–75 min to 40–45 min. The labour productivity went up as the idle
time of hamalis reduced with them being involved continuously in the actual
loading/unloading activity. Gubba was adding new facilities, and the "redundant"
computer operators were retrained to do the work of assistant supervisors at these
new warehouses. This helped Gubba reduce the cost of manning the plants by
avoiding new hiring.

Standardization

The pre-alert form was updated with the actual truck arrival time and exit time
from Gubba premises. With this information, the turnaround time of each truck
could be tracked. As the ERP was upgraded and mobile applications developed
it was even easier to capture this data without errors. At the gate the security
officer now records the in and out time on the mobile app. Similarly, the assistant
supervisor captures the dock in and out times. This data is presented in graphical
format against the specified standard of 3 min per tonne for actual loading or
unloading and five minutes per truck for documentation. Hence for a 10-tonne
truck, the standard TAT is 35 min, and for a 15 tonner, it is 50 min.
Performance of various Gubba plants is compared and monitored on this
parameter which has been incorporated as a KPI for the plant supervisors.

Stabilization Phase

In the peak season months following the initial round of improvements, the team
was hard-pressed to maintain the improvements achieved. While, space utiliza-
tion was ensured, these was still some firefighting to turnaround trucks within
stipulated times, especially on the days with a peak load of over 50 trucks a
day. Still the employees benefitted through a definite reduction in strain. While
in previous years, they would stay till the night to complete the day’s loading and
unloading; in this season, such occasions were few and far between. By end of the
season, Prashanth estimated that Gubba had sustained about 60% of the original
implemented improvements.
Post the peak season, the stabilization phase of Lean was launched through
framing and implementation of standard operating procedures (SOPs) that would
incorporate all the good practices implemented during the improvement phase.
As a preursor, a TPM facilitator was invited to conduct a couple of training
programs on autonomous maintenance for the refrigeration technicians and mate-
rial handling equipment team. Abnormalities were identified and corrected, CLRI
checklists developed and implemented and the existing Planned Maintenance
sheets fine-tuned.
302 Case Study 5: Lean at Gubba Cold Storage

Standard operating procedures (SOPs) were then written for the entire gamut
of operations at the plant including the interactions with corporate functions. Each
SOP was written by the core team comprising of operations manager, group super-
visor, quality manager under our guidance. This SOP would then be reviewed
by Prashanth, discussed again with the concerned functions, revised as required
and finally approved by Prashanth for roll out. The SOP incorporated all the
improved practices and additional checks and verifications to ensure these would
be followed. A sample SOP for material inward process is shown in Fig. C5.5.
The entire cycle of framing, reviewing, and freezing the SOPs took three
months. By December, the team was ready to roll out and a launch event was
organized. Prashanth and Kiran headlined this one-day event which began with
Prashanth explaining to the plant teams, the relevance of SOPs. The structure of an
SOP was then explained to the team. Two-member teams were then formed, with
each team tasked to validate the current status of a couple of SOPs at the gemba.
By lunch time, each team had identified the gaps in SOP adherence and the things
to be done for ensuring compliance to the SOPs. These included floor marking,
cleaning of cupboards, pallet segregation, visual markings on equipment, switches,
etc., in other words basic 1S, 2S, and 3S activities needed to be done. Some quick
fixes were taken up on the same day to demonstrate management intent and Kiran
concluded the day with a stirring motivational address to the team. By the end of
December, 5S was in full swing and Gubba plants were getting ready to adhere to
the SOPs in the upcoming season.
In February, we along with the core team conducted the first SOP audit. Gubba
team was familiar with customer audits and took them seriously. Hence, a typical
5S assessment wast weaked and re-branded as a SOP audit with the 5S scores
being replaced by a count of the number of SOP deviations. This first audit also
served to train the core team on the art of assessment as they were to take over
the Lean initiative from us. A monthly audit schedule was fixed for the first three
months coinciding with the 2015 peak season. A deviation register was created
in Google sheets and shared across the plants and with Prashanth and Kiran. The
deviations were captured with evidences such as a photo or video and the auditee
(plant supervisor) was expected to take corrective action and record the evidence
in the adjacent column of the sheet. At a glance, Prashanth was able to see plant-
wise deviations and how many had been corrected, enabling him to intervene with
the plant supervisors when needed.
A trend chart of deviations for each plant was also put in place for the monthly
reviews. Once a month, all the plant supervisors met at the corporate office for a
team lunch followed by a monthly plant performance review during which half an
hour was kept aside to discuss the SOP adherence status. The core team also started
looking at the data on repeat deviations within a plant or across plant locations. In
line with a typical Kaizen approach, repeat deviations were analysed, root cause(s)
identified and improvement actions implemented to prevent them from recurring.
Gubba also began implementing a reward and recognition scheme for employ-
ees wherein the best improvement ideas are appreciated in various forums like the
newsletter, the notice boards, and in various internal forums.
Process Process Sub process Sub Process
Number Number
10 Receipts (Inward) 10.1a Planning for racked storage
10.1b Planning for Conventional Storage
10.2 Gate Entry - In
10.3a Unloading for racked storage
10.3b Unloading for conventional storage
10.4 Gate Pass Generation & Gate Entry - Out
10.5 Gate Entry - Out
Sub Process SOP
Sub Process Number: 10.1a
Sub Process: Planning for racked storage
Activity Activity (Step) Document Responsibility
Case Study 5: Lean at Gubba Cold Storage

No. Ref.No.
1 From the pre-alert sheet filled up the previous day, call transporter 1 hour before the truck ETA to Pre-alert Supervisor
reconfirm its arrival time Sheet
2a If the truck is delayed, reschedule the same in the pre-alert sheet and inform the client accordingly.
Supervisor
2b If the truck is on time inform the Assistant Supervisor to prepare for unloading the same
3 Keep ready the required hamalis if PP bags are expected, also alert the pallet truck operator to be Assistant
ready for unloading. Supervisor
4 From PACT system, check the availability of suitable slots for storing the expected material. Assistant
Physically verify the condition of these slots as per Racking SOP and get the same cleaned and Supervisor
rectified if required.
5 In case truck gets further delayed, call up the client/transporter again and follow up. Supervisor

Fig. C5.5 SOP of material inward process


303
304 Case Study 5: Lean at Gubba Cold Storage

Leveraging Technology to Sustain Lean

The practice of SOP audits, corrective actions, looking for repeat deviations, and
trying to improve practices had become a routine in Gubba by 2016. In a couple
years, it became evident that a few recurring deviations needed different solutions
to permanently solve them. For example, the pre-alert sheet not being filled up or
stack card updation not happening at the storage points were issues that cropped
up in audit after audit across plants. Additionally, the existing mix of manual
formats, updating the data again in Google sheets, and then compiling MIS from
these sheets was proving to be cumbersome and error prone. In fact, a lot of time
was going into these activities rather than the final analysis, decision-making, and
execution of improvements.
Around this time, Prashanth started looking to improve or replace the existing
software with a newer ERP. Technology especially with respect to smart phones,
internet connectivity, and mobile applications had developed at an express pace in
the last few years and Prashanth spotted an opportunity to leverage this. Through
a referral, he got in touch with a local ERP developer who was using open-source
technology to build custom ERP applications. Prashanth tied up with this firm and
started work on a new ERP. Within a year, the basic version was up and running;
Prashanth then recruited a bright young functional consultant into the corporate
team to work on upgrading the basic version with add on functionalities. Most of
the SOPs were integrated into the software so that the required MIS was generated
automatically. More importantly, SOPs which were being repeatedly deviated from
were integrated in an error proofed manner into the software.
We saw earlier that a pre-alert sheet would often remain unfilled and it was
not clear whether supervisors were actually speaking to customers and taking pre-
alerts. This pre-alert format was now in a mobile app installed on the supervisor’s
phone. Similarly, the earlier gate entry SOP had the security filling up manual
registers for entry and exit of trucks. Now, they would fill it up in a mobile app
while physically verifying the truck making it a lot easier to ensure. Since both
the pre-alert and the actual truck arrival is on the ERP, trucks without pre-alert are
automatically highlighted, and if more such deviations are observed, the concerned
plant supervisor is counselled.
Another major deviation seen was the non-updating of stock location following
internal shifting. In the existing process, assistant supervisors note the changed
location on a scrap of paper when they are monitoring the shifting inside the
plant. They are then expected to fill up these details in a separate “internal shifting
register” kept at the plant office. The new locations were then to be updated in the
ERP software. Not doing this resulted in delays in loading at a later date when
the stock was not found at the location given in the software. The new mobile
enabled ERP solved this problem as well. Assistant supervisors can now update
the location at the time of shifting directly on their phones instead of using scraps
of paper. The rest of the processes were rendered redundant since the ERP is
automatically updated with the location details.
Case Study 5: Lean at Gubba Cold Storage 305

From these two examples, it is clear that most SOP deviations occur when the
SOP is cumbersome or hard to follow consistently. Using technology to simplify
and make it easy to follow ensures adherence in the long run while also alerting
immediately of any deviations, without the need for special audits. In fact, by
2019, Gubba had reduced the frequency of SOP audits and the type and content
of each audit was also restricted to the essentials.

Lean in the Latter Years

Over the years, Gubba had some success in Lean. Improvements made in the ini-
tial intervention have got institutionalized through the SOPs. Some processes have
been improved further under Lean paradigms of eliminating or minimizing Muda.
Error proofing and improved standards have been incorporated through technology.
But what Gubba had not been able to achieve as yet was the spirit of contin-
ual improvement. Lean thinking had not become a way of life for the employees
and initiatives to keep improving processes and standards was not a part of the
company’s DNA.
This reality struck home when in late 2019, Prashanth and Kiran decided to
relook at the processes and activities that had become a routine over the past five
years. Another round of focused improvement workshops was planned. But the
objective this time was more towards expense reduction through optimization of
people. Gubba was under some financial pressure due to market conditions and
felt that by reapplying Lean principles and reducing non-value-added activities,
there would be a reduced need for the people who perform these activities. We
were roped in once again to guide the teams in this renewal of the Lean program.
Rationalization of people has never been an objective of Lean and no wonder
then that enthusiasm of both the consultants and the team was markedly less than
in previous years. Relooking at plant operations now using the new ERP software
threw up a large number of non-value-adding duplications and clearly showed a
reduced need for supervisory attention. The earlier structure of one plant supervisor
with two assistant supervisors was continuing and it was now seen that a plant
could easily be managed with one assistant supervisor. Another reason for this
rationalization is the off-season period of more than six months during which the
incoming and outgoing transactions are only a fraction of that in the season. The
supervisory staff are on the company rolls and therefore a fixed cost which yield
no real benefit for this off-season period.
Another solution which came up for discussion involved having a common
team of supervisory staff at the cluster locations like Yellampet (with six plants)
and Medchal (with four plants). On a given day, the assistant supervisors could get
assigned to the trucks as they come in irrespective of the specific plant where they
have to load or unload. This idea though sound in principle did not find traction
with the plant teams who gave all sort of reasons for not attempting it. In earlier
years, the operations manager, a management graduate brought in from outside,
had embraced and driven Lean activities. After his departure, one of the brighter
306 Case Study 5: Lean at Gubba Cold Storage

supervisors was elevated to this position. During this intervention, he was also
unable to convince his colleagues and supervisors about the need for improvement.
It was clear that the employees, the same ones who embraced Lean with gusto in
2013–14 had become too well entrenched and complacent in the system.
A similar intervention targeting corporate office functions went somewhat better
as the new ERP had clearly thrown up the need to eliminate a lot of non-value-
adding paperwork and duplicating activities being done on Google sheets. In fact,
the new ERP itself was called into question on a few processes. For example,
the purchase process involved multiple approvals leading to an increased time
between indent and actual purchase. A why-why analysis threw up the reasons
and almost half the approvals were found to add no value and were done away
with. Customizing the technology to suit the operating procedures of the company,
rather than other way round, is very important for lean to sustain, as can be seen
in the case of Gubba.
The action points that emerged from these workshops between December 2019
and February 2020 were in progress when COVID broke out, throwing the whole
team off gear and Gubba has had to redefine the way of working over the last one
year to adapt to the “new normal”.

Conclusion

When Kiran and Prashanth went ahead with engaging us in 2013, Kiran was still
sceptical on whether Lean would really make any difference to their business. In
his words “We had already put in place the best systems and processes and felt
we were operating at the maximum possible level in terms of capacity”. He was
therefore pleasantly surprised by the response of his employees and the outcomes
of the initial Lean improvements.
The focused improvement workshops brought together diverse employees, pro-
moted teamwork and a sense of achievement which left everyone on a high. A
10% increase in storage capacity utilization is a huge gain for a cold storage facil-
ity. The ability to turnaround trucks at an industry leading benchmark of 3 min per
tonne sets the standard for others to try and emulate. Coupled with the productiv-
ity increase in supervisory staff and overall look and feel of a world class plant
through SOPs and visual management have definitely given Gubba a positive Lean
experience. In the words of Prashanth “Before Lean we thought we were already the
best and there is nothing left to improve. But Lean opened up a different paradigm
and this in turn gave rise to a whole new set of improvement opportunities and ways
of thinking”.
Prashanth, a system-driven and tech savvy person, had always been implement-
ing innovative practices and systems. With his growing belief in Lean, he was the
key driver in developing several permanent solutions to simplify SOPs and ensure
adherence through the ERP software during subsequent years.
On the flipside, the inability to maintain the momentum given by the first couple
of years of Lean and allowing complacency to set in among the employees has
Case Study 5: Lean at Gubba Cold Storage 307

meant that Gubba has still not realized the true and full potential of Lean. The
years ahead will tell us whether they can reignite the passion for improvement and
bring in a lasting cultural change or remain content with past achievements.
Case Study 6: Applying Lean
to Problem Solving
in the Pharmaceutical Sector

Background

A decade ago, the Indian pharmaceutical industry had just begun its growth tra-
jectory with the opening up of the global market. Large Indian pharmaceutical
companies were contracted to supply to global majors which in turn drove the
expansion of the local supply chain as they developed vendors to supply the var-
ious bulk drugs and intermediates required by them. These vendors were mostly
SMEs. As the demand rose, these SMEs often struggled to meet the customer
delivery time expectations. For most part, the SME bulk drug units stayed with the
tried and tested formula; make one or two products for one of the large pharma
companies with a fixed set-up employing the minimum people needed to run the
operations. Once the product was trialled and approved, there were no further
changes. The process was standardized and continued that way as long as the con-
cerned product was in demand. Margins and earnings were fixed and a few, if any,
thought of growth or dynamic expansion.
At this time, GTZ, a not-for-profit organization of the German government,
was working on a scheme to help SMEs improve their long-term sustainability
and growth prospects. Mr Chandrashekhar Reddy, who led GTZ in India, was
open to new ideas related to process improvements and technology that could help
SMEs work towards the twin objectives of sustainability and growth. A referral
led to a meeting with Ganesh Mahadevan at the GTZ office, where the idea of a
targeted Lean intervention at select SMEs was born. GTZ invited SME owners to
a seminar titled “Performance Improvement through Lean” which was attended by
more than thirty entrepreneurs.
Two of these entrepreneurs came forward and expressed interest in applying
Lean to improve their operations. Over the next couple of months, they managed
to rope in a third SME and formed a mini-cluster to implement Lean at their
units. All three were young second generation entrepreneurs who had finished
their studies in the USA, worked briefly at other large organizations there and
returned to India to join their respective family businesses. The stint abroad had

© The Editor(s) (if applicable) and The Author(s), under exclusive license 309
to Springer Nature Singapore Pte Ltd. 2023
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9
310 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

Table C6.1 Key differences between discrete and process manufacturing


Parameter Discrete Manufacturing Process Manufacturing
Product Unique, easily identifiable Undifferentiated, often not visible
during the process
Measurement Discrete, one unit of product Weight/volume can change from one
remains through-out the process process step to the next
Operations Not continuous, each can operate Continuous, all the steps are generally
independently at its own rate linked
Key components Assembly of parts, Bill of Mixing of ingredients,
Materials recipes/formulations
Outcome Reversible, parts can be reworked Irreversible, often chemical change
as change is only physical

given them a basic awareness of Lean as well as the openness to try something
new which had not been attempted in bulk drugs sector before.
The manufacture of bulk drugs typically involves material synthesis in reactors,
separation of the liquids in centrifuges and drying of the wet product mass in
dryers before the final dry product is packed. The liquid by products from the
centrifuges/filters may undergo distillation for solvent recovery. The remnant liquid
is finally sent for effluent treatment. The effluent may be either completely treated
and discharged from the factory itself or partially treated and sent to common
treatment facilities.
At that time, there were very few documented cases of Lean implementation in
this industry. So, with a sense of adventure the bulk drug SME mini-cluster kicked
off a six-month long pilot Lean initiative exploring the use of Lean concepts,
tools, and techniques in resolving their constraints, getting better in operations,
and enabling them to grow.

Applicability of Lean in Bulk Drugs

With its origins in TPS, most of the Lean tools and techniques like SMED,
workstation design, line balancing, and error proofing were initially developed
to improve flow of material in the machining centres, press shops and assembly
lines. However, a process industry such as bulk drugs is quite different in several
key parameters as we see in Table C6.1.
Most importantly, material flow is by design an integral part of process man-
ufacturing. All the steps are already linked (usually) and material flows from one
step to another; in this sense, the bulk drug manufacturing process can be called
Lean by nature. So, was there any scope for applying Lean here to obtain sig-
nificant gains? We will now see how Lean was deployed to improve processes
here through a few specific. examples from all the three SMEs.
Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 311

Setting Up the Cluster

Two factors were key to this cluster.

• The industry owners knew each other. Two of them, Vamsi and Chaitanya, were
cousins while the third member Srinivasan was a friend
• Each of their factories was making a unique and different product for a different
customer, so there was no competition.

We planned to have focused improvement workshops at each of the three factories


wherein team members from the other factories would also join the cross-
functional teams to learn and apply Lean concepts and tools. They were expected
to then use the learnings to improve their own processes. We would review the
progress in all three factories post each workshop. This cross-factory team forma-
tion and learning was quite unique considering it was cross- organization as well
and actually contributed to the success of the implementation in two specific ways.

1. Process observations by the “outsiders” were rational, as there was no emotional


or intellectual bias towards the process
2. Solutions for improvement were enriched by the different set of experiences
and practices being followed in their respective factories.

The initiative began with the usual current state assessment and drawing up of a
unit-specific Lean roadmap. This exercise was conducted independently at each
of the three factories involving the key functional in-charges of that factory. The
concerned owners were was also involved in all the workshops and reviews in their
factories.

Current State Assessment

The first step in a typical Lean implementation is to assess the current state with
respect to the business goals and opportunities and identify the constraints. An
improvement roadmap of series of projects is then framed for logical and sequen-
tial elimination of these constraints. The example of Porus Drugs (SME #1) shows
how such a roadmap is made for this industry.
Porus was producing and delivering about 28 TPM (tonnes per month) of Prod-
uct X for a European customer until now. For the coming year, the customer
signalled an intent to increase the purchased volume to 40 TPM. This served as
the primary target for the Lean intervention at Porus with the management goal
being to increase output quickly without having to necessarily invest in new infras-
tructure. A value stream map (VSM) was developed to assess the current state for
product X which is manufactured through a series of processes.
In a discrete environment, VSM is made using data from practical observation.
Cycle times and changeover times are recorded by observing the process when
312 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

it is on. Most of the cycle times are in seconds or minutes and thus observation,
measurement, and recording is practically feasible. In the pharmaceutical industry,
the reaction cycle times are often of the order of several hours and the process is
not visible. However, it is mandatory to maintain a detailed batch record and in this
case the process data from the records of recently completed batches was compiled
to prepare the VSM. The steps for making current state VSM are described below.

Step 1: Set target output based on projected customer requirement. For Porus,
this was 40 TPM of finished product X.
Step 2: Convert target to takt time. Calculation for product X is shown below.
Available time: 30 days x 24 hrs = 720 hours per month
Customer demand: 40 TPM ÷ 500 kgs per batch = 80 batches per month
Takt time = Available time ÷ customer demand = 720 ÷ 80 = 9 hours per
batch
Hence, the plant is expected to deliver one batch of finished product every 9
hours and every process in the value stream has to produce at this or a faster
rate.
Step 3: Compute batch equivalent so as to have a common output measure for
all the processes. This is again a unique feature of chemical industry. Unlike
discrete manufacturing, the output to input ratio varies from process to process
depending on the chemical reaction. For Product X, final product batch size
was 500 kgs of finished product, but batch equivalent varied for each process.
Step 4: Compile process-wise data from the batch records and input the same
into VSM format. To identify any Mura (process inconsistencies), not just the
average cycle time but the minimum and maximum cycle times and the standard
deviation were all computed and incorporated into the VSM.
Step 5: Identify constraint or bottleneck(s) processes which need to be
improved. Basically, any process whose cycle time is more than the takt time is
a bottleneck. However, processes with too much variation can also impact the
lead time, and therefore should be targets for improvement.

The VSM data for Product X was captured by using these five steps as shown in
Table C6.2.
The bottleneck processes are those with average cycle times more than tar-
get takt time of 9 hours and it can be seen that there is also a large variation from
batch to batch. The core Lean concept of value-add and waste was incorporated in
this VSM as well. Activities within the process where a change in material prop-
erty or characteristic takes place are value-adding and the times taken for these
activities were summed up as the value-adding time for the process.
The process value-adding ratio is the value adding time divided by the total
cycle time, and gave an indication of the scope for improvement within the process.
Looking at the bottleneck processes, the team saw that reaction already had a high
VA ratio of 85% but with significant cycle time variations. Drying on the other
hand had a VA ratio of <40%. This helped the team decide on the appropriate Lean
tool to be used for process improvement in each specific case.
Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 313

Table C6.2 VSM data for Product X


Current Target
Customer requirement (Tons/month) 28 40
Batch size kgs 500 500
Demand rate Batches per day 1.9 2.7
Takt time hours per batch 13 9.0
Process Available equipment Cycle Value Change Effective VA
time Added Over Cycle ratio
Time time Time (%)
Numbers Batch (hours
equivalent per
batch)
Reaction 5 5 Average 67 57 4 13.4 85
Min 57 46 0
Max 87 73 10
Neutch filter 1 1 Average 11 4 10.6 33
Min 6 1 0
Max 18 6 0
Dissolution 1 1 Average 6 2 5.8 33
Min 4 1 0
Max 12 3 0
Carbon 3 1.5 Average 18 12 11.7 69
Min 12 8 0
Max 27 20 0
Acidification 3 3 Average 16 6 5.3 37
Min 10 4 0
Max 26 11 0
drying 4 2 Average 20 8 10.0 39
(Part-II) (350 Min 16 4 0
kgs per drier)
Max 41 21 0
Methanol 1 1 Average 11.8 3.8 12 32
purification Min 7.5 3 0
Max 34.75 6 0
Filtration 3 3 Average 21 12 7.0 55
Min 11 4 0
Max 34 21 0
drying 3 2.07 Average 21.4 8.44 10.3 39
part—III Min 10.5 2 0
(350 kgs per
drier) Max 30 9 0
314 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

Fig. C6.1 Current state VSM for product Pavest

A similar VSM (this time pictorial) made for KRR Drugs (SME#2) for its main
product, Pavest as shown in Fig. C6.1. Here, we see that reaction, distillation, and
drying are bottlenecks to achieving the takt time of 450 min per batch.
Once the VSM was done, the team did a detailed walk-through of the shop
floor to identify improvement opportunities with respect to the working conditions,
safety and environment as well as upkeep of the plant and machinery. A Lean
roadmap was then made listing out the performance goals and the improvement
projects to be taken up for achieving these.

Implementation

There were two main phases of this pilot Lean initiative as shown in Fig. C6.2.
A cross section of the key improvement projects done across the three factories
are explained in further detail below.
Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 315

IMPROVEMENT PROJECTS

I. Enable Flow II. Support Flow

To eliminate bottlenecks so that To ensure that the established flow is


material flows without stopping sustained by providing requisite
through various stages of manufacture resources – equipment availability,
from Raw Material to Finished Goods utilities, ETP, laboratory, good working
warehouse. conditions

Fig. C6.2 Phases of improvement projects

Phase 1—Enable Flow

In this phase, each SME worked on all the bottleneck processes identified in the
VSM. The focus was firmly on Muda, Muri and Mura reduction through adoption
of Lean tools such as VA and NVA analysis, line balancing, problem solving, and
improving work methods. An example of each type of project is explained in the
following sections to give the reader an overall perspective of Lean implementation
in this industry.

Reducing Muda of Batch Wait Time

Problem Definition
To understand the concept of Muda as it applies to a process plant, we shall take
the case of product Pavest. The VSM shows a constraint of high cycle time in the
distillation stage of the process; the cycle time is 585 min per batch as against
takt time of 450 min per batch. A cross-functional team was formed to work on
reducing the cycle time to below 450 min which would enable the plant to meet
demand of 3.2 batches per day.

Observation and Analysis


The product is processed through three stages, each stage being carried out suc-
cessively in Reactor No’s 114, 115 and 116 with the key distillation process taking
place in stage 2. The current sub-process with the activity-wise cycle time break
up is shown in Fig. C6.3.
Reaction takes place in Stage I under brine-based cooling condition after which
the separated organic layer is first transferred to Reactor 115. The aqueous layer
is then extracted with solvent and also transferred next to Reactor 115. Only
after both the layers are transferred, the next process, high temperature distillation
begins.
316 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

Column

Reactor GLR 114 Reactor SSR 115 Reactor SSR 116

Process time= 360 min + Solvent


Methanol Charging + 2
Transfer = 60 min part distillation – 360 Transfers – 60 min
Total = 420 min min
Waiting time – 45 min Water addition +
io layer Maintenance – 120 min Maintenance = 270
Aqueous Transfers – 60 min min
Organic layer Total = 585 min Total = 330 min

2 stage transfer Mass


Mass
-1st organic
-- 2nd aqueous

Centrifuge

Fig. C6.3 Process stages of Pavest

Detailed process observation was carried out for the distillation process wherein
four consecutive batches data was taken over two days. For each batch, hourly
distillation quantities were monitored along with the relevant process parameters
(see Table C6.3).
The activity-wise cycle time break-up and the why-why analysis of distillation
batch data presented the following insights:

• Excess water found in organic layer transferred from previous stage in batches
1 and 4
• Steam pressure fluctuation leading to slower distillation rate in batch 3

Table C6.3 Distillation monitoring data


Distillation monitoring board
Batch No. 1st hour 2nd hour 3rd hour 4th hour 5th hour 6th hour Total time Total
(min) quantity
(l)
Target 350 225 225 150 100 0 300 1050
1 480* 220 225 127 93 315 1145
2 345 235 234 122 90 21 310 1047
3 345 215 200 135 80 90 410 1065
4 340 221 225 150 120 153 400 1209
*Values deviating significantly from Target are marked in italics, and analyzed further
Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 317

• The cycle time of next process in Reactor 116 was only 330 min which meant
a 120-min cushion against the takt time of 450 min.

Countermeasures Implemented
Eliminate, Combine, Simplify and Rearrange (ECSR) principles were used to
improve the process. The key solutions implemented were:

• The reactor level indicator was made visible to the operator and water flow
control valve shifted next to the level indicator to simplify and enable filling
of correct quantity of water in Stage 1 thus avoid excess water in the organic
layer
• The waiting time between transfer of organic and aqueous layers was elim-
inated by starting the distillation immediately after getting first Layer from
previous stage
• A Kaizen was done to reduce distillation time by preheating of Reactor 115
by hot water. Hot water already available from another process was routed
through an existing overhead tank to Reactor 115 before each transfer. This
small modification using existing resources was implemented within two days
and resulted in significant reduction and consistency in distillation time.
• The last process step in Reactor 115 involves maintaining the still mass for
120 min post distillation. Since the next stage in Reactor 116 had a lower cycle
time, this step was rearranged such that half the maintenance activity could be
done in Reactor 116 by transferring the mass one hour after distillation.

Result
The highest cycle time was brought down to 420 min and the three stages were
well balanced as can be seen from Fig. C6.4. As a result, an average of 3.3 batches
could be produced per day which is 30% higher than earlier and enough to meet
customer requirements for Pavest.

Reducing Muri (Eliminating Dust in the Milling Room at KRR Drugs)

Problem Definition
The dry milling process inludes grinding, weighing and packing the bulk drug.
As per the VSM, the time taken for dry milling of Pavest was well within the
takt time. However, during the shop-floor walkthrough at the time of current state
assessment, it was observed that the operation was generating of a lot of fine
dust and powder. This was not only leading to material loss at the final stage of
processing but also affecting the working environment and health of the operators
in the room. The operator would be fully covered with product dust by the end of
his shift; in fact, there was so much dust that the camera lens itself was fogged
while attempting to take a photograph of the current state. Please see Fig. C6.5.
318 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

Column

Reactor GLR 114 Reactor SSR 115 Reactor SSR 116

Process time= 360 min + Solvent


Methanol Charging + 2
Transfer = 60 min part distillation – 300 Transfers – 60 min
Total = 420 min min Maintenance – 60 min
Waiting time – 0 min Water addition +
io layer Maintenance = 60 min Maintenance = 270
Aqueous Transfers – 60 min min
Organic layer Total = 420 min Total = 390 min

2 stage transfer Mass


Mass
-1st organic
-- 2nd aqueous

Centrifuge

Fig. C6.4 Three stages after improvement

Breathing
difficulty

Contamination

Spillage during
transfer

Leakage from
plastic joint

Fig. C6.5 Grinding and weighing—current state


Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 319

Table C6.4 Root causes and solutions to reduce dust spillage


Observation Solution implemented
Weighing and packing was done manually in By providing the weighing scale directly
the next step and involved manual transfer of underneath the miller, manual weighing
finished product from one bag to another. activity and transfer of fine powder from one
Material spillage as well as generation of fine bag to another was eliminated
dust observed during the transfer
Material spillage observed while feeding the All valve leakages and generation of fine dust
miller, leakage of fine dust during operation during milling was eliminated by sealing the
from sides of the dome, and while collecting miller and providing venting arrangement for
the material into the bags releasing trapped air
Physical strain of the operators and helpers To reduce the strain and make the feeding
while lifting the bags weighing 25 kgs each to hopper easily accessible the milling stand
charge into the hopper height was increased

An improvement project was therefore taken up with a goal to ensure “zero


powder loss” by eliminating all leaks and spills of material at this stage.

Observation and Analysis


The operation was observed in detail (see Fig. C6.5) and a why-why analysis done
to identify root causes of spillage and leakage. Key observations and implemented
solutions are summarized in Table C6.4.

Result
Zero dust working environment created where the operator could work comfort-
ably without even feeling the need to wear a protective mask. Product loss due to
spillage and leakage was eliminated. The following Fig. C6.6 shows the condition
of dry milling area after the above-mentioned improvements were implemented.

Reducing Mura (Smoothening Inconsistency in Reaction Time


at Porus)

Problem Definition
The VSM for Product X (Table C6.2) shows reaction to be the bottleneck processes
that constrains the manufacturing unit’s capability to meet customer demand. Five
reactors were available for carrying out this single stage reaction process and all
of them were utilized fully. Analysis of batch data from the VSM showed that
the average cycle time was 57 h which translated to an Effective Cycle Time of
13.4 h per batch and this was well above the takt time of 9 h per batch. A cross-
functional team was formed to work on this problem and bring down the reaction
time to below 45 h per batch in order to match takt time.

Observation and Analysis


Data showed that there was a wide fluctuation in cycle times from batch to batch
with the actual value-adding (reaction) time ranging from 46 to 73 h. Since five
320 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

New feeding stand

Can breathe easily

Dome with vent to collect fine Flow control plate; can


powder in bags. No leakage pre-fill powder and avoid
leakage from filling point

Online weighing.
Eliminates activity of
transfer

Fig. C6.6 Improved condition in dry milling

reactors were used for carrying out the same reaction, the differential diagnosis
technique was used to analyse the problem and arrive at the root cause(s) of this
variation. Differential diagnosis is a backward or deductive approach for solving
chronic problems. This method involves a combination of direct process obser-
vations and data collection to answer a series of questions under the following
categories:

1. What is the problem,


2. Where or in which place does it occur,
3. When does it occur, and
4. How much (quantity or frequency).

In each question, both the presence and absence of the condition is noted.
Through this, the causes of the problem are narrowed down and what remains
at the end will yield the root cause through a why-why examination.
In this case, the data showed that all five reactors exhibited significant variation
in reaction times which meant that there was a common cause for the inconsis-
tency. Each reactor had in some batches, been able to complete the reaction in
a time below 50 h. By comparing these low reaction times to the higher times,
the root cause could be identified. Whenever the reaction temperature of 145–
147 °C was consistently maintained, cycle time was between 46–50 h irrespective
of which reactor was doing the process. In cases when cycle time rose to much
higher levels, it was observed that the temperature dropped below 145 C due to
fluctuation in the pressure of the steam used for heating. It normally took at least
Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 321

Reaction Time for Reactor 14


70

65

60
Hours

55

50

45

40

35
1 2 3 4 5 6 7 8 9
Batch No. Before After

Fig. C6.7 Reactor time variation

two hours for the temperature to come back to the required range in such instances.
Since there was no alert mechanism, any such drop in temperature would be noted
during the operator’s routine half hourly observation round. Whenever this hap-
pened, the operator would manually adjust the valve and try to get more steam to
the reactor to raise the temperature. This adjustment time also varied due to trial
and error and operator experience.

Countermeasures Implemented
An improved method of steam control was put in place which included valve
position marking and an alert mechanism for pressure deviation. A process chart
was prepared and visual management system for monitoring of parameters put in
place. To ensure that there was no other possibility of variation, recalibration of
all temperature sensors and display units and cleaning of thermo-wells that house
the sensors was carried out.

Result
Variation in time fell by 50% across all the reactors (Please see Fig. C6.7) and
average reaction time came down to 44 h from earlier level of 57 h. As a result of
this improvement in conjunction with reduction of other bottleneck process cycle
times, the output of the product increased from 28 to 33 TPM.

Phase II—Stabilization

The improvement workshops targeting the 3 M’s helped the teams from all three
factories to apply the core Lean concepts of flow and waste elimination and
improve their critical processes. It also highlighted the importance of the support
processes such as quality laboratory, utilities maintenance, and stores, etc. It was
322 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

time to move to the second phase of the Lean journey, stabilizing the gains made
through streamlining of the support processes.

Reducing Laboratory Testing Time


One of the observations at AR Life Sciences (SME#3), pertained to the completion
of reaction stage. Their major product Q underwent eight reaction stages and at
each stage the completion of the reaction was known only through lab analysis
and verification of a sample drawn from the reactor. When the material had spent
the expected amount of time in the reactor, a sample was drawn and taken to
the laboratory for analysis. The lab chemist would confirm the completion of the
reaction and inform the production shift supervisor who would then transfer the
reacted mass to the next stage. This process (take sample - >go to QA - > test -
> confirm results - > start transfer) took 30–35 min on an average. So, we asked
the team a fundamental question “What happens in the reactor during this time?”.
Since the test result essentially gives the status of reaction half an hour ago, it
meant that reaction was already complete at that time. So, the material was actually
waiting for the result which was a Muda.
It took some persuasion of the team to come around to this Lean perspective
but once they got it, they quickly got onto the target of reducing the time for result.
Of the cycle, about 20 min was spent in the lab and the balance for collecting the
sample and bringing it in. At the sampling end, the spoon design was changed to
make it easy to pick up the sample in one shot thereby halving the time required.
At the testing stage, the team observed and shot a video of the entire activity from
the time the sample is brought in till the result is communicated to the production
floor. The major observations were:

• Actual test time only about 50% of total time spent,


• Chemist has to make multiple movements inside the lab during testing pro-
cess—the testing had multiple steps which needed various equipment including
an oven, titrator, wash basin for rinsing and a spot test set-up,
• Searching for glassware and chemicals during testing took some effort and time,
• Physical strain—bending for taking out certain reagents and glasswares, and
• Unused items were found occupying space in the shelves.

The observations highlighted the need to implement the core 5S principles in


the lab. Within the next couple of days, the team worked on the following;

1. Segregated unwanted items and disposed off to appropriate locations thereby


freeing up 50% of the shelf space,
2. The team made a simple product-process matrix for the lab tests after cate-
gorizing the tests as runners (frequently done), repeaters, and strangers. For
example, reaction tests that happen daily are runners while tests on certain
imported incoming raw materials happening once in three months would be
strangers.
Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 323

Cells were formed with test equipment, reagents, glassware, and test formats
for the runners arranged in one place as compared to the earlier item-type-wise
arrangement. All the items and locations were then marked by labelling and
indexing.
The change in orientation from item-type-wise to test specific arrangements
brought down the movements for the chemist. This coupled with search free
arrangement resulted in the testing time dropping to 12 min from 20 min. Overall,
the material waiting time in the reactor was reduced by 40% while the lab chemists
enjoyed strain free working conditions.
A 5S checklist was instituted which included the following.

• Red tag area defined and unwanted items to be shifted there periodically.
• Daily lab cleaning SOP framed and implemented.
• Disposal protocols including timelines and methods for rinsed solvents, broken
glassware, and old records were defined.
• Reordering of solvents, reagents, and other chemicals based on visual stock
level checks.

Autonomous Maintenance for Utilities


The value-adding processes in any chemical plant are dependent on the consistent
availability of quality utilities such as steam, chilled, or compressed air or vacuum.
To ensure the process delivers consistent value and throughput, we have to make
sure there are no disruptions in the quality and quantity of these utilities.
In the stabilization phase of three months, the teams engaged in the initial
basic steps of autonomous maintenance focusing on the utilities. And these are
extensively driven through a network of pumps and motors and pipelines carrying
them to the customer processes. AR Life Sciences, like most other SME units, had
a basic maintenance team consisting of one mechanical fitter and an electrician
who both reported directly to the production manager. Utility operators were few
and were mainly for the boiler and chilling plants.
A maintenance expert consultant was therefore deployed to guide the plant
teams in abnormality identification, rectification, and certain good practices that
can prevent abnormalities. Cleaning, Oiling, Tightening, and Inspection (COTI)
sheets were developed for the all the critical equipment. Visual markings were
done on pipelines, pumps, and motors to facilitate quick and simple adherence to
the COTI routines.

Reorganizing the Stores


If autonomous maintenance was expected to keep the equipment up and running,
there was an equal need to ensure availability of the required spare parts for routine
time-based and condition-based replacements. Also, in the case of an unexpected
breakdown, having the spare in hand made all the difference between a quick
repair and restart versus hours or even days of downtime. One look at the stores
324 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

at Porus Drugs highlighted why the maintenance personnel spent significant time
searching for the required part and sometimes end up not finding what they need.
The team’s observations pointed to the need to implement 5S in the store. The
store was full of materials with no aisles to move inside. Material was found
lying in mixed-up conditions in most of the racks and on the floor; in some cases,
unusable items were also mixed up with the good ones. The stores assistant found
it difficult to reach items on the top most shelf of the racks for which he needed to
use ladder. All this meant that it took a lot of time to search for and get required
items.
The team initiated 5S on a war footing. Within the week, they had segregated
all scrap and expired items, and kept them separately to be disposed off. This
enabled them to discard two damaged racks and change the rack arrangement to
create aisle space and accessibility. They then rearranged all useful items neatly
on the shelves and racks as per type (e.g. bearings, motors, switches, etc.) and area
of use. Item address locators were pasted on each rack to create a search free store
as shown in Fig. C6.8.
Now, any item could be located in a minute without searching for it. Almost
20% of stores area was freed up and ease of access to all items provided. The
visibility provided for a better control on inventory and ensuring reordering of
items as and when required.

Sustaining Lean

They say that success is the biggest motivator; if that be, all the three SMEs ended
up with very good reasons to sustain the Lean path. Table C6.5 is the summary of
the outcomes reached at the three factories by the end of the pilot intervention.

Before 5S After 5S

Fig. C6.8 5S—Before and after in stores


Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector 325

Table C6.5 Summary of improvements achieved by the bulk drug pharma SME cluster
SME# Parameter Before After Improvement
1 Production volume (MT) 28 35 25% gain
Throughput time (days) 10.4 4.5 57% reduction in WIP
Yield (kgs) 510 540 6% savings in material loss
2 Throughput time (days) 32 20 38% reduction
Production volume (MT) 300 480 Potential of 60% increase
3 Production volume (MT) 8 12 Potential of 50% increase

A quick look at where these organizations are a decade later gives some pointer
on how early stage Lean intervention has shaped their thinking and growth. Porus
has grown tenfold in turnover with large new facilities set-up at different cities.
In fact, Porus is no longer an SME. AR Life Sciences was at a pilot phase of
developing a niche low volume but high value product (refer Table C6.5 for the
quantities) at the time of Lean intervention. They have up scaled to two midsized
manufacturing units one of which is US FDA approved and a third facility is under
construction. KRR Drugs did not really focus on their pharmaceutical business in
later years. Instead, they have grown their fabrication unit which now supplies
equipment to pharmaceutical factories in India and abroad. Vamsi, the owner, took
an additional step of implementing Lean in his foundry unit and ensuring his
fabrication unit team also follow core Lean concepts.

Conclusions

This experience with bulk drug manufacturing clarifies some important differences
in terms of how Lean is to be applied in a process industry as compared to a
discrete manufacturing unit.

• Muda is observed by understanding the change that has happened to the mate-
rial in terms of its chemistry and this change is often not visible or obvious
visually. A discrete process lends to physical observation of the seven wastes
which is easy,
• Reducing Mura leads to significant benefits in these industries as the processes
are reaction based and depend on multiple process parameters and raw material
variations. For this, the collection of accurate data is very important. In discrete
manufacturing, process variations are in manual operations can be observed
visually and acted upon
• In process manufacturing, effective value addition depends on utilities like
boilers, chillers and compressors being able to help maintain the required
conditions. The focus is therefore on ensuring these utilities perform to their
optimum. Whereas in discrete manufacturing, the focus is operator’s work
methods in conjunction with machine.
326 Case Study 6: Applying Lean to Problem Solving in the Pharmaceutical Sector

In conclusion, it may be said that the core philosophy and paradigm of Lean
remains equally relevant to a process industry as it is to the discrete manufac-
turing. The concepts need to be applied keeping in mind the inherent nature
of process manufacturing by slightly tweaking the methods of observation and
increased focus on data collection. The benefits of implementing Lean are equally
significant as could be seen from the three case examples described in this case
study.
Case Study 7: GSV Industries—A Case
study on Lean Implementation

Background and Context

GSV industries (GSV), located at Coimbatore, India, primarily manufactures two


types of products—stator and rotor stampings and motor guards. They cater to the
needs of the pump industry with their customers including industry leaders such
as Aqua pumps, CRI pumps, PSG Industrial Institute, Texmo Industries, etc.
Each year, the pump manufacturers have been constantly putting pressure on
their vendors to cut costs resulting in margin pressures for companies such as GSV.
The managing partner of GSV therefore felt the need to enhance the operational
efficiencies of the unit. As a member of one of India’s biggest industry associations
for small scale industries, Coimbatore District Small Scale Industries Association
(CODISSIA), he attended an association seminar on the benefits of Lean manu-
facturing where other members who had earlier implemented Lean shared their
experiences. This was a precursor to a Lean cluster implementation program pro-
moted by the Government of India (GoI) to help SMEs on the path to sustainable
growth. GSV joined the cluster to implement Lean manufacturing under the guid-
ance of the selected (by GoI) Lean Management Consultant. The program was
envisaged to be an 18-month long journey in implementing key concepts of Lean.

Projects Identification and Selection

The journey commenced with a diagnostic exercise during which the processes at
GSV were observed and studied under Lean paradigms of zero waste and zero
defects. The factory consisted of two sheds, one for manufacture of motor guards
and the other for stampings. The core team consisted of the two section-in-charges
and the managing partner all of whom joined the Lean consultant during this
diagnostic exercise. Key observations from this exercise and the improvement areas
identified are discussed further in this section.

© The Editor(s) (if applicable) and The Author(s), under exclusive license 327
to Springer Nature Singapore Pte Ltd. 2023
G. Mahadevan and K. C. Chejarla, Lean Management for Small and Medium Sized
Enterprises, Management for Professionals,
https://doi.org/10.1007/978-981-19-4340-9
328 Case Study 7: GSV Industries—A Case study on Lean Implementation

Stamping

Significant investment had been made in two large stamping presses which are
used for pump stator and rotor parts. The stamped pieces produced by the machine
are then assembled manually in the form of stacks. The Overall Equipment Effec-
tiveness (OEE) was computed to be around 45% only with the factory working
for 12 h (including overtime) to complete the daily production target. The equip-
ment availability was the major loss contributing to lower OEE and the following
observations were made in respect to these losses.

Changeover Time
There were two types of changeovers (viz., batch and product) observed in the
press. The raw material is in the form of steel strip coils. As one coil gets con-
sumed, the end bit is cut off and the next coil loaded and strip fed into the die.
This batch changeover from completion of one coil to the starting of the next took
5 to 10 minutes and was done several times in a day. Technically, this loading and
unloading is a part of the coil processing cycle time; however, it was deliberately
categorized as a changeover to have the team focus on it.
Further, the stamping die itself needs to be changed to run another product
variant, and this took about 90–120 min. There was no proper marked location for
dies and tools resulting in long searches for them at the time of changeovers.

Breakdowns
Frequent breakdowns were stopping the presses, the mean-time-to-repair was more
than two days as an external service technician had to be called to solve the
problems.

Stamping Projects

While the customer demand was expected to remain the same, the management
was focused on minimizing the overtime expense which was eating into the already
low margins. Increasing OEE would lead to a direct and proportionate increase in
output and the following improvement projects were identified to work on this.

• Reduction in changeover time: With at least six coil changeovers per day and
three die changeovers per week in stamping, the press availability eroded by
more than 30%.
• Reduction in breakdowns by establishment of system of autonomous mainte-
nance (AM): With a large amount of time in stamping lost due to breakdowns,
establishment of an AM system is expected to enhance the OEE of the stamping
lines.
Case Study 7: GSV Industries—A Case study on Lean Implementation 329

Motor Guard Section Observations

The motor guard manufacturing process has multiple steps involving man and
machine and gemba walk was done to observe Muda (waste) and Muri (strain).
The observations are listed here.

Inventory
Batch production system was being followed in motor guard section. Each process
was planned independently and there was significant overproduction leading to
piling of inventory between various processes. WIP inventories of various sizes
of motor guards had been kept deliberately before the jhali (grill punching) stage
(please see Fig. C7.1 for value stream map of motor guard production process) in
order to expedite delivery as and when order schedule was received.

Material Transportation and Handling


The team observed multiple handling of components between each process leading
to increased strain (Muri) to the workers. The component generally drops to the
floor after the process is completed and has to be picked again and placed in
baskets/bins before being moved to the inventory area or next process resulting in
significant operator strain and material handling.
Following the gemba walk, the overall process flow in motor guard manufac-
turing was mapped through a value stream map (VSM) as shown in Fig. C7.1.
As can be seen the value adding ratio (VAR) is only 0.01% mainly due to the
high WIP between processes leading to a high throughput time.

Motor Guard Section Projects

Given the low VAR, establishing flow was considered important in the imple-
mentation phase. There was a large WIP build-up between various motor guard
production processes because of independent batch processing by different work-
stations. With cycle time of the various processes being extremely low (usually in
seconds) and the lead time for delivery being at least a week, establishing flow in
motor guard line was expected to enhance the throughput of the line and improve
On Time In F startull (OTIF) delivery substantially.

Roadmap Creation

A core team was formed to work on the identified improvement projects, and a
stage-wise project roadmap was created. The projects were implemented in Stages
II, III, and IV of the implementation journey and the improved processes were
standardized in the last Stage V.
330

Monthly /
Weekly
Production Engineering Contract MONTHLY
Weekly plan Vendor P.O Plan Info Review Order receipt TARGET
Supplier Customer 50000 nos
No. Working
High Frequent Hours 200 hrs
changeover stoppages Multiple
High WIP Production
time material
Weekly between different Takt Time
Manager handling 14.40 secs
processes

Daily daily
daily daily daily daily Shipment

RM Circle Cup Grill Screw hole


120 hrs preparation 32 hrs cutting 32 hrs formation 16 hrs Turning 8 hrs formation 8 hrs drilling 16 hrs Packing 40 hrs
2 1 1 1 1 1 2
C/T per piece C/T per piece C/T per piece C/T per piece C/T per piece C/T per piece C/T per piece
= 20 sec = 10 sec = 15 sec = 23 sec = 5 sec = 5 sec = 2 sec
AT = 8 hrs AT = 8 hrs AT = 8 hrs AT = 8 hrs AT = 8 hrs AT = 8 hrs AT = 8 hrs

37912/1110600
VAR = = 0.01%

432000 115200 115200 57600 28800 28800 57600 144000 = 979200


20 10 15 23 5 5 2 = 80

Fig. C7.1 Value stream map of motor guard line


Case Study 7: GSV Industries—A Case study on Lean Implementation
Case Study 7: GSV Industries—A Case study on Lean Implementation 331

Table C7.1 Phase-wise implementation roadmap


Activity Phase I Phase II Phase III Phase IV Phase V
Improvement project
Milestone completion date Mar-16 Jul-16 Oct-16 Feb-17 Jul-17
I. Motor guard line: Set up one flow line with balanced cycle times and
2400 nos 3200 nos
improve flow to run in single piece flow
increase throughput Reduction in WIP from 2–7 days thereby reducing
and reduce lead time 1.5 days 1 day
strain
Change the entire plant layout as per the line concept
II. Reduce material validated for one flow line
handling and Define standard WIP where needed and mark space
transportation through only for the fixed quantity
flow layout Implement 1S, 2S and 3S at every work station and
throughout the factory
Reduce die changeover time for stamping from 120
30–45 min
min to <60 min
III.Stamping line:
Improve the OEE of stamping from current levels of
improve stamping 60% 70%
less than 45%
machine OEE
Implement basic Autonomous and Planned
Maintenance practices to minimize breakdowns
5S audits and self-sustenance model 32% 40% 68% 80%
Visual management systems
Training on Kaizen
IV.Synchronization and
Reporting system for Kaizen
sustenance of Lean
Indicative target for no. of Kaizens
Structure and action plan for handover to unit and
long term sustenance

Project Team

The core project team comprised of:

• Mr. Vijayarangharajan—Managing Partner


• Mrs. Beoula—Stamping section in-charge
• Mr. Das—Motor guard section in-charge

Project Schedule

The phase wise road map with targets for each project was finalized in conjunction
with the managing partner and is shown in Table C7.1.

Implementation

Increasing the Stamping Press OEE

Die Changeover Time


As we saw earlier, there is a coil changeover several times a day and a die
changeover almost every alternate day. The team first focused on die changeover
time reduction.
A die changeover normally happens every alternate day on each press. A cross-
functional team was formed to observe the die changeover from product PE80
(single chute) to PE63 (double chute). The stamped parts are automatically dis-
charged into chutes where they get auto stacked continuously. Small stacks are
removed at the exit of the chute by an operator, and kept aside for weighment and
332 Case Study 7: GSV Industries—A Case study on Lean Implementation

manual assembly process. The chute size is specific to the product and each die
cavity is linked to one chute. The total changeover time recorded for this particu-
lar changeover was 102 minutes and involved two operators. The team first noted
that many of the changeover activities were being carried out after the press was
stopped (internal) even though they could be done while it was running (external).
These included:

(a) Movement of the crane from its existing location to the machine,
(b) Searching for the tools needed for changeover,
(c) Bringing the chutes needed for next product near the machine,
(d) Moving the removed chutes to its designated location,
(e) Bringing the new set of dies from its existing location,
(f) Moving the removed die to its designated location.

They then analyzed the observations on the activities that are done post machine
stoppage (internal changeover):
30 mins were spent in removing the output (stators and rotors) of the production
run from the chute. Only after this, could the chutes be replaced with the set of
chutes for the next product.
Gear changing and setting adjustments were also found necessary in some of
the changeovers.
In order to locate the chute plate, a steel ruler was used to measure the distance
from the left-side of the bed. It was observed that the operator made a mistake in
noting the reading which resulted in redoing the chute plate fixing.
Taking cues from the SMED concept, the team worked out an improvement
action plan and implemented the following:

• All the internal activities (a to f) listed above were converted to external and
carried out before (a, b, c, and e) the press is stopped for changeover or after
(d, and f) the press is restarted.
• Tie rods were used to remove the rotor material from the chutes instead of by
hand. Distance piece was used to correctly align the chute plate for every die
changeover instead of using the steel ruler for measuring the distance from the
left-side of the bed. The product number is inscribed in the distance piece itself
to avoid using of wrong distance piece. Pneumatic gun used for loosening and
tightening of bolts.
• Internal activities were done in parallel by different operators as per the respon-
sibilities assigned. All the available operators were utilized during the course
of the changeover as shown in the activity allocation Table C7.2.

The above improvements resulted in reduction in die changeover time from an


average 90 min to 45 min. This meant that press availability increased by 4%
resulting in an equivalent increase in output.
Case Study 7: GSV Industries—A Case study on Lean Implementation 333

Table C7.2 Changeover activity allocation


S.No. Activity Dependent activity Manpower Time (min)
1 Die bolt removal – 4 persons 1
2 Die removal 1 2 persons 2
3 Material removal from chute 1 4 persons 2
(stator + rotor)
4 Chute removal (4 bolts per stator + 3 4 persons 4
3 bolts per rotor)
5 Gear changing 1 1 person 12
6 Chute fixing 4 4 persons 7
7 Die positioning 6 2 persons 7
8 Die bolt fixing 5, 7 4 persons 1
9 Coil feeding and setting 8 2 persons 1
Time 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
in
min→
Worker 1* 2 4 6 7 8 9
1
Worker 1 2 4 6 7 8 9
2
Worker 1 5 8
3
Worker 1 3 4 6 8
4
Worker 3 4 6
5
Lady 3
Worker
1
Lady 3
Worker
2

Improve the Availability of Stamping Press Through Autonomous


Maintenance
During the diagnostic stage, the team found that there was no recorded data on
the machine breakdowns and initiated data collection of machine breakdowns and
short stops. The data showed that:

i. Regular breakdowns were occurring in the press and the mean-time-to-repair


was high because of need to call in external service technician.
ii. On an average three to four short stoppages per hour were observed on the
stamping machine.
334 Case Study 7: GSV Industries—A Case study on Lean Implementation

The Lean consulting team brought in their TPM specialist to conduct a workshop
on initial cleaning and abnormality identification. The operator was then given the
responsibility to regularly identify all abnormalities in his machine, correct those
which he could immediately do so and escalate the rest to the supervisor. After
this initial restoration of the stamping press, the next two steps of AM were put in
place.

i. Cleaning of machines was introduced as a regular daily activity. The operator


was provided 15 min at the end of the shift for machine cleaning.
ii. A CLRI (Clean, Lubricate, Re-tighten, and Inspect) checklist was developed
and implemented for the press and was regularly followed by the operators.

Operators started recording the abnormalities in a register kept in the supervisor’s


office and depending on the need, the external technician would be called in to
correct the identified issues before an actual breakdown occurs.
The effect of the above countermeasures was an improvement in the OEE of
stamping press seen in Fig. C7.2.
The actual production trend is shown in Table C7.3; the output is now com-
pleted in regular shift timings of 8 working hours instead of overtime in 12 h per
day, taken earlier.

Stamping OEE
80.00%

68.29%
70.00%

60.00%

50.00% 44.95%

40.00% 37.09% 44.41%


34.11%
30.16%
30.00%

20.00% 25.99%
22.13%
10.00%

0.00%
Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16 Jan-17 Feb-17 Mar-17

Fig. C7.2 Improvement in stamping OEE


Case Study 7: GSV Industries—A Case study on Lean Implementation 335

Table C7.3 Stamping production—output trend


Before implementation During implementation Standardization
Period Tons Period Tons Period Tons
Feb, 2016 74 Oct, 2016 89 Apr, 2017 107
Mar, 2016 76 Nov, 2016 120 May, 2017 127
Apr, 2016 92 Dec, 2016 131

Improvement in Throughput of Motor Guard Line

The second shed at GSV factory houses the motor guard manufacturing line. With
an increasing demand from the pump OEMs, there was pressure on GSV to adhere
to the stringent and ever-tightening delivery schedules. The current throughput
time for a batch of motor guards varied anywhere from 2 to 30 days, most often
resulting in delivery delays. Thus, the team decided to focus on improving the
throughput at this line.
The process flow for motor guard manufacture is shown in Fig. C7.3:
The team identified the existing layout as the prime cause for the shop floor
being filled up with stacks of WIP, leading to multiple material handling and
transport. They also observed that the operators themselves were moving the
materials intermittently keeping their machines idle.
The original machine layout was designed on conventional batch produc-
tion concepts. Machines were laid out functionally and no two machines were
linked together leading to WIP build-up between processes. Each machine oper-
ated independently of the others and it’s was collected in large cages as seen in
Fig. C7.4.
Cages of such components were then moved to the centre of the working area
from where the next process would again draw the material as and when needed.
When the cages became full the material was placed in heaps on the floor as

1. SHEET MARKING 2. HAND SHEARING 3. CIRCLE CUTTING

6. COVER JHALLI 5. COVER TRIMMING 4. COVER DRAWING

7. SCREW HOLES 8. REDOXIDE PAINT 9. COVER PACKING

Fig. C7.3 Motor guard manufacturing process


336 Case Study 7: GSV Industries—A Case study on Lean Implementation

Fig. C7.4 Material handling in cages

well. Inspite of all the WIP, some machines were still found waiting for input
materials. The team walked through the value stream (See Fig. C7.1) and measured
the material movement between operations as shown in Table C7.4.
There was no clear tagging of WIP, or identifying the next process it is supposed
to undergo. While a weekly production plan was made, it was not strictly adhered
to. Components taken up for production were often left incomplete and a new order
would be taken up succumbing to customer’s pressure of immediate deliveries.
As the team walked through, there was also a realization on the extent of criss-
cross material movement between operations and this came out clearly in their
material flow (“sphagetti”) diagram depicted below (Fig. C7.5).
With this visualization, it was evident that the current layout and batch produc-
tion method was hampering the smooth flow of motor guards through the value
stream thereby increasing WIP and the throughput time.

Table C7.4 Material


Process Step Distance Travelled (m)
movement between processes
Cutting to Circling 4.8
Circling to Hydraulic 15
Hydraulic to Lathe(Trimming) 4
Lathe to Jhalli 5.2
Jhalli to Screw Hole 1.5
Screw Hole To Painting 17
Total 47.5 meters
Case Study 7: GSV Industries—A Case study on Lean Implementation 337

Fig. C7.5 Material flow in motor guard manufacturing

Improvement Plan
The machines were reorganized based on sequence of operations for different com-
ponents to form a cell. Two such cells were formed as shown in the layout drawing
(Fig. C7.6). The dotted lines indicate the two cells while the solid line traces the
material flow in the new layout. The cells were designed based on the motor guard
size as larger motor guards needed to be made on the higher capacity hydraulic
presses.
In the new layout, the machines were spaced so as to facilitate an operator to
directly pick up the component from the outbox of the preceding process. Wher-
ever the machines could not be brought close enough, a simple inclined chute was
fabricated to enable the components to flow from one workstation to another. The
WIP between the workstations is controlled by length of this chute; if the chute
becomes full, the supplier process stops. Figure C7.7 shows one such chute.
Since the sheet marking (1) and circle cutting (3) operations were much faster,
they were delinked from the rest of the processes and a fixed standard inventory of
338 Case Study 7: GSV Industries—A Case study on Lean Implementation

Fig. C7.6 Cellular layout for two products

cut sheet circles maintained for use by all the cells. The weekly customer require-
ment was the basis for fixing the standard inventory levels. Two-day requirement
of cut circles of the sizes planned were maintained and replenished on a daily
basis. The production of these upstream processes was now driven by a pull-based
model where the replenishment of cut sheet circles was based on withdrawals by
the cells.
Based on the cycle time, some of the components on the jhalli formation (6)
and screw hole drilling (7) was done by a single operator in order to balance the
line. The machines were aligned to facilitate multi machine operation by a single
operator as shown in Fig. C7.8.
Case Study 7: GSV Industries—A Case study on Lean Implementation 339

Fig. C7.7 Chute controlling the WIP between processes

These two machines in the picture are aligned at 90 0


to each other to enable ease of access to a single operator

Fig. C7.8 Machine alignment to enable single worker operation

Another observation by the team working on this project was the multiple han-
dling of scrap being done by the machine operators. At each machine, the punched
or milled scrap was falling onto the floor behind the machine. Operators would
spend time before close of work in collecting this scrap and putting it into bins in
the designated area outside the shop floor. With the new chute system in place and
flow replacing the earlier batch manufacturing, most of the cages became redun-
dant. Some of these cages were now placed to directly collect the scrap falling
from the machine and moved out by a pallet truck when they became full.
340 Case Study 7: GSV Industries—A Case study on Lean Implementation

Table C7.5 Motor guard


Before implementation After implementation
production trend
Period Quantity Period Quantity
Jan, 2016 39,768 Mar, 2017 45,530
Feb, 2016 40,955 Apr, 2017 49,776
Total 80,723 95,306

Outcomes
With the establishment of flow in motor guard line, the Lean team could demon-
strate an increase in the output from about 900 per shift to 1800–2000 units per
shift with the same number of operators and machines. On an overall basis, allow-
ing for other factors such as actual customer orders, material availability and
schedules, the year-on-year comparison of output during the peak season shows
an actual 18% gain as seen in Table C7.5. At the same time, the new layout uses
only 50% of the space used by the earlier layout.

Standardization and Sustenance Plan

Following the improvement phase of the Lean journey, the last six months of
the roadmap schedule were devoted to ensuring standardization of processes and
building a structure for long-term sustenance of Lean.

Standardization Through 5S
This began with an awareness workshop for all the supervisors of the plant high-
lighting the importance of 5S and how it facilitates the sustenance of improvements
made under Lean implementation. The unit was divided into various zones and
each zone was assigned to a zone leader responsible for all Lean activities within
their span of control. To start off, a red tag area was identified in the plant in
order to accommodate all unwanted/unused tools, etc. The unit head was assigned
the responsibility for periodically reviewing the items of the red tag area and take
appropriate decisions for disposal or reuse.
2S activities were planned and carried out by each zone so as to have a system-
atic arrangement facilitating their process flow. The machine cleaning introduced
under AM was now merged into the shine activity with a fixed daily schedule to
carry out cleaning activities in their respective areas.
Two months into 5S, the Lean consultant carried out the first 5S audit to mea-
sure of the level of implementation of 5S in the plant and to create zone wise
competition as a motivating factor. By the end of the six months, the 5S score of
the plant had improved from 30% in Feb 2017 to 52% in July 2017.
Case Study 7: GSV Industries—A Case study on Lean Implementation 341

Table C7.6 Lean organization


S. No Particular Responsibility Role Frequency
1 5S Audit (cross audit) Ms. Beoula/Mr. Das Auditor Monthly
2 Lean implementation Ms. Beoula/Mr. Das Lean coordinator Weekly
3 Lean review Mr. Vijayarangarajan Proprietor Monthly

Organization Structure for Sustenance


In order the carry forward the improvements made under the various projects so
far, a responsibility matrix was created assigning responsibilities to the Lean cham-
pions of the unit. The matrix contained the name, role, and the frequency of review.
GSV being a small-sized organization had a very limited managerial bandwidth.
The people available to take the onus for continuity of Lean and 5S were few. This
limited the scope for assigning too many responsibilities and a basic structure was
created within the limitations as shown in Table C7.6.

Conclusion

Even a unit as small as GSV could gain significantly from the implementation and
sustenance of core Lean concepts. The output of the stamping section rose by 30
MT per month over the previous season as a result of an increase from 75,000
to 85,000 strokes per day in each stamping press. This resulted in an increase in
saleable output to the tune of Rs. 21.9 million with the same resources. An increase
of 5000 motor guards per month was also seen which translated to a revenue
growth of about Rs.3.6 million per year. All this was achieved with the existing
resources of people, machinery, and facilities and this resulted in a significant
contribution to the profitability of the unit.
In the 18 months, the team had worked hands-on and understood how each
of the Lean tools could be customized for their operations. This enhanced the
confidence of the members of the team to bring in necessary changes to stay
competitive in the market.

You might also like