Production Operations Management

Term Report

Submitted By:

INVENTORY MANAGEMENT
FARIHA SAMAD
AAZAR ALI SHAD inventory management in production logistics: a multi-agent
Knowledge-based
BBA-V
approach; Impact of Inventory Management on service flexibility and Performance
Research Article #1:

Knowledge-Based Inventory Management In

Production Logistics: A Multi-Agent Approach

C F Cheung*, W M Wang, and S K Kwok

Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University,
Kowloon, People’s Republic of China

Introduction
In order to make an enterprise more predictable and better prepared to deal with global
competition, it needs to have an accurate and highly dynamic inventory management model.
Traditional inventory management in production logistics is inadequate in managing inventories
with high fluctuation in demand and value. 1

To establish effective inventory management strategy, this research paper presents a

knowledge-based inventory management system (KBIMS) for active inventory replenishment
based on multi-agent dynamic forecasting and knowledge-based system technologies.

The knowledge-based inventory management system dynamically forecasts the fluctuation of

the demand and updates price of its raw materials. Using this method appropriate inventory
management strategies are derived to respond to the rapid market changes and the material
requirement.

At the end, the research paper also presents a case study of a manufacturing enterprise
wherein a prototype KBIMS has been built and successfully trial implemented.

Defining the Key Terms:

1
C F Cheung*, W M Wang, and S K Kwok
Let us begin by defining the key terms discussed in this research paper.

(i) Production Logistics: The term production logistics is used to describe logistic
processes within an industry. The purpose of production logistics is to ensure that
each machine and workstation is being fed with the right product in the right
quantity and quality at the right time. The concern is not the transportation itself,
but to streamline and control the flow through value-adding processes and eliminate
non–value-adding ones. Production logistics can be applied to existing as well as
new plants. Manufacturing in an existing plant is a constantly changing process.
Machines are exchanged and new ones added, which gives the opportunity to
improve the production logistics system accordingly. Production logistics provides
the means to achieve customer response and capital efficiency. 2

(ii) Knowledge Management: Succinctly put, KM is the process through which

organizations generate value from their intellectual and knowledge-based assets.
Most often, generating value from such assets involves codifying what employees,
partners and customers know, and sharing that information among employees,
departments and even with other companies in an effort to devise best practices. It's
important to note that the definition says nothing about technology; while KM is
often facilitated by IT, technology by itself is not KM.3

(iii) Knowledge-based systems: Knowledge based systems are artificial intelligent tools
working in a narrow domain to provide intelligent decisions with justification.
Knowledge is acquired and represented using various knowledge representation
techniques rules, frames and scripts.

2
Baziotopoulos (2008) An Investigation of Logistics Outsourcing Practices In the Greek Manufacturing Sector. PhD
thesis
3
“Knowledge management: Definitions and Solutions” by Meredith Levinsion ; www.cio.com
The basic advantages offered by such system are documentation of knowledge, intelligent
decision support, self learning, reasoning and explanation. Knowledge-based systems4 are
systems based on the methods and techniques of Artificial Intelligence. Their core components
are:

 knowledge base
 acquisition mechanisms
 inference mechanisms

(iv) Multi-Agent Technology Or Multi-Agent System (MAS): Multi-agent systems are a

new paradigm for understanding and building distributed systems, where it is
assumed that the computational components are autonomous: able to control their
own behaviour in the furtherance of their own goals.5

(v) Conventional Inventory Management Methods: What we mean here by

conventional inventory management methods is the conventional way of inventory
replenishment is based on classical forecast methods where safety stock is
determined on the basis of forecast error.

Why do we need a knowledge-based inventory management system?

A better control of inventory relies heavily on the know-how, experience, and quality of the
staff in deriving effective inventory replenishment strategies. This knowledge is difficult to
acquire and share; training is time consuming and costly, and staff turnover can negatively
impact operations. In addition, mistakes can be very costly if the inventory replenishment
decisions are made solely based on human subjective judgment.

4
Knowledge based system definition from FOLDOC.org

5
An Introduction to MultiAgent Systems - Second Edition by Michael Wooldridge
Knowledge Based Approach versus Conventional
Approach for Inventory Management

Conventional Inventory KBIMS

Management methods

The conventional approach starts by The proposed KBIMS approach provides

tracking the usage and vendor performance integrated tools that optimize inventory
Statistics to forecast demand. stocking levels and automate
purchasing to maintain favourable
customer service levels and inventory
balances.

Based on the specific replenishment It collects the inventory and market

controls, actual inventory usage, and the information continuously and
purchase order lead times with a pre- automatically.
defined quantity, an inventory
replenishment decision is made when the
inventory level is below safety stock.
This replenishment strategy not only is static A dynamic forecasting method is
and passive but also leads to excessive deployed to forecast
inventory and the inflexibility of coping with the dynamic trends of the inventory
urgent demand. requirements and market conditions.

Based on the forecasted results,

decision support is provided to suggest
advisory recommendations for the
inventory replenishment which includes
optimal order points, safety stock,
optimum order quantities, and
shipment schedule.

The staff can then refine the

suggestions given by the system.
Knowledge of the refinement is
assimilated to provide continuous
learning of the system.

Structure of the KBIMS

The architecture of the KBIMS consists of a multi-agent system, which is composed of
independent agents namely an inventory replenishment agent, a planning agent, and a series of
dynamic forecasting and data agents. It is built on a five-layer framework.

The Multi-Agent Approach for Knowledge Integration

As defined earlier the knowledge based inventory management is built using the tools of
artificial intelligence and multi agent technology. Agents are relatively small pieces of
autonomous software that act in various roles on behalf of a specific function or user. 6

These agents serve as assistants to the human staff, thus mapping human intelligence to
intelligent agents.

6
Chorafas, D. N. Agent Technology Handbook, 1998 (McGraw-Hill, New York).
Data Agent: a data agent is an autonomous search agent responsible for searching different
websites or databases. For instance, the data agents for metal materials are responsible for
searching and updating periodically for market prices of raw metal materials (e.g. copper and
aluminum). All data from multiple data agents are fed to the dynamic forecasting agents.

Dynamic Forecasting Agents: The dynamic forecasting method starts with the forecast
of the data by an autoregressive (AR) time-series model. Basically, the dynamic forecasting
agent can be divided into two parts, i.e. the learning phase and the application phase. The error
in forecasting is continuously monitored within the predefined tolerance. When the module
detects an excessive forecasting error, a feedback alarm signal is issued to alert the attention of
management to the need for the recalibration of the forecasting model of the dynamic
forecasting agent.
Decision Support Agent: The decision support agent serves the purposes of deriving
useful knowledge from the forecast data predicted by the dynamic forecasting agents, data
agents, and enterprise resources planning system. The derived knowledge is used to accomplish
the inventory replenishment decision, etc.
Potential Advantages Of KBIMS
It provides a more accurate and dynamic forecast for its inventory level. It also enables more
efficient connections with the up-to-date market information and rapidly changing market
conditions. KBIMS is able to capture and use the knowledge of the most experienced staff to
generate dynamic business rules for inventory management. Effective inventory replenishment
strategies in terms of optimal order points, order quantities, and safety stock can be generated
on the basis of an analysis of its inventory demand and the relevant market conditions, and the
production logistics activities are streamlined. This helps to maintain optimal stock levels, to
reduce inventory carrying costs, and to help to eliminate special or urgent reorder charges.
Purchase orders can also be generated automatically.

Application In Pakistani Industries:

The concept of knowledge management is almost non-existent or relatively new in Pakistani

industries. Implementation of a knowledge-based inventory management system in Pakistani
industries would therefore be a challenge, as the structural rigidities in Pakistani industries do
not enable change to take place easily, nor are the employees open to change. Moreover, the
technical infrastructure of the local industry is not developed enough. In this scenario, the
Multinational organizations should play there positive role in bringing this change to the
Pakistani industry.
Masood textile Mills, with its headquarters located in Faisalabad, is a leading knitwear exporter
in Pakistan. Over 85% of its exports are to USA and the rest are to EUROPE. MTM is proud to be
the only pakistani company to have installed a sophisticated computerized bar coding system
which allows it to track a product and all of its various components right from spinnning to the
packed product. This has dramatically improved their quality of production and reduced
rejection rates. MTM as launched a direct-to-store (DTS) business service which reduces lead
times and logistic management worries. (article in Appendix)

Research Article # 2
Impact Of Inventory Management On
Service Flexibility And Performance
Literature Review

The focus of this article is flexibility from competence within and outside firm. It will also
explore the relationships between inventory competence flexibility, service capabilities
flexibility, and performance. The results show that intra-firm inventory management flexibility,
which reflects internal competence, has a positive impact on service capabilities and
performance, while inventory coordination flexibility, which captures inter-firm competence,
has no significance on those elements. Also it explains that collaborative inventory
management flexibility is an important at least for mainland china.

This article explains that to achieve the effectiveness and efficiency of inventory management,
internal competence and external capabilities are very important. Competence is not visible to
customers; however, marketing and manufacturing strategies are visible. Also, Inventory
service capabilities which are externally focused can be viewed as linkages between firms'
marketing and manufacturing strategies. Flexible managerial competence, which is internally
focused, is an important precursor to inventory service capabilities.

The article has divided itself in Inventory management flexibility [competence perspective] and
Inventory service flexibility [capabilities perspective].

Inventory Management Flexibility [Competence Perspective]

Inventory management flexibility specifically considers volume flexibility alongside adjustment
capacity and process flexibility. Recently, Business to business infuse flexibility throughout its
organizations by developing inter and intra firm inventory management. Intra firm inventory
management is related to coordination between departments related to inventory i.e. finance,
marketing, manufacturing and engineering. On the other hand, inter firm inventory
management is related to firm coordination with its outside stakeholders like suppliers.
Therefore, management flexibility is defined as a systematic competence to manage and take
responsibility for two or more functions along the supply chain, whether within or outside the
firm.

Intra-firm Inventory Managerial Flexibility

Intra-firm inventory managerial flexibility can be achieved only when three conditions—joint
management, information sharing between cross-functional units, and a low level of internal
conflict— are satisfied.

In order to achieve integrity and traceability of Inventory, a company should have a very flexible
team to coordinate with the production function, customer needs, and overall internal value
chain activities, and all impacted areas should agree on the appropriate performance metrics
and their definitions.

Inter-firm inventory management flexibility

Inter-firm inventory management flexibility suggests that organizational structures allow
Managers and firms' inventory to respond swiftly to change by utilizing inter-firm systems and
techniques to influence the environment so that the firm becomes less vulnerable to
uncertainty.

There are three kinds of alignment which can make inventory management flexible. Following
are the alignments:

Information alignment refers to information sharing among channel participants. Continuous

information exchange regarding availability and delivery could reduce uncertainty for the
supply chain and create the opportunity for maximum flexibility.
Process alignment is realized when inventory management processes are synchronized within
or between trading partners.
"Plug and play" alignment is concerned with dynamically interchanging products and partners
in a fast-changing environment. In this context, inventory management needs to cater to the
inter-relationship of products and partners to reflect the flexible supply chain configuration.
Continuous replenishment (CR) and vendor-managed inventory (VMI) can be regarded as
effective ways of "plug and play".

Inventory Service Flexibility [Capabilities Perspective]

Inventory service flexibility provides an organization with the ability to keep inventory services
to an agreed service level in a predictable fashion with acceptable risk and cost, and this
capability can be tested and valued by customers.

There are three categories of inventory reflecting the level of availability to the customer: raw
material inventory, work-in process inventory, and finished goods inventory.

Hypothesis of Research
Following were the hypothesis of the research done in article which was conducted 206
companies and 126 valid cases:
HI: Intra-firm inventory management flexibility will be positively associated with firms'
performance.
H2: Inter-firm inventory management flexibility will be positively associated with firms'
performance.
H3: Intra-firm inventory management flexibility will positively impact on inventory service
flexibility.
H4: Inter-firm inventory management flexibility will positively impact inventory service
flexibility.
H5: Inventory service flexibility mediates between intra-firm inventory management flexibility
and a firm's performance.
H6: Inventory service flexibility mediates between inter-firm inventory management flexibility
and a firm's performance
Explanation and Conclusion
The empirical results do not support inter-firm inventory management flexibility positively
impacting service flexibility and performance in Mainland China, which indicates that the
positive relation between inter-firm inventory management and inventory service/
performance are not very strong among Mainland Chinese firms and therefore has no
significant impact on service flexibility and on upgrading the performance of organizations. It is
because firms in China inevitably focus on intra-inventory managerial flexibility rather than
inter-management flexibility which lead to the neglect of information sharing and cooperative
inventory managerial practices with their partners.
Thus, only internal competence in activities such as inventory control and warehousing is not
sufficient in achieving competitive advantage; the most important thing is to achieve internal
and external competence, which enables firms to effectively coordinate external partners'
activities and internal functions to flexibly respond to events and trends of present and
prospective customers.
Therefore, to effectively improve inventory performance, the implementation of information or
inventory systems should be accompanied by the implementation of inter-firm operational
practices. To be exact, intra-firm inventory flexibility emphasizes cross-function and internal
inventory management, which is associated with a firm's inventory service capability.
 APPENDIX

Application of Inventory Management in Procter and Gamble

Business Launch competition

Case Study
Replenishmen Transit Replenishmen Minimu Inventory Per Container Container Hours per
t Points Times t Frequency m Cover Month Monthly trips per month
(To and inventor (Months) Cases Inventory month
Back y Cover Require Capacity
(Hours) (days) d
)
Karachi 0 N/A 12 0.4 15,177 N/A N/A N/A
0.1333333 2.26653479 1.20881855
Hyderabed 4 2 days 4 3 1,765 2 0.3022 6
0.2666666 2.26653479 13.2970041
Multan 22 4 days 8 7 1,765 2 0.6044 2
0.49272495 6.89814936
Lahore 28 5 days 15 0.5 8,118 5 0.2464 8
1.25918599 17.6286039
Faisalabad 28 5 days 15 0.5 3,177 6 0.6296 4
2.26653479 36.2645566
Gujranwala 32 5 days 15 0.5 1,765 2 1.1333 8
1.88877899
Rawalpindi 32 5 days 15 0.5 2,118 4 0.9444 30.2204639
2.83316849 45.3306958
Islamabad 32 5 days 15 0.5 1,412 1 1.4166 5
5.27681053 150.848292
1 4

Annual Inventory 423,554 Cases

Forecast Error 20 %
338,843 -
Inventory Cover (Annual) 508,265 Cases
Daily Inventory 2034 Cases
Inventory at CDC (15 Days 30,510 Cases
 Cover)

National Inventory Cover is 15 Days

Warehousing Cost PKR Per Pallet Per

(CDC) 150 Week
PKR Per Case Per
1.67 Week
50,850 PKR Per Week
2,644,20
0 PKR Per Annum
0.78 PKR Per Unit

Container hours required per month 151

Availaible Container hours per
container 15 per day
420 per month
Required Container trips per month 6 Conservative Estimate
Safest Case Scenerio 1 dedicated container for
non dedicated container
1 for
Fixed Costs 100,000 for dedicated container
Variable costs (dedicated container) 400,000
Variable costs (non-dedicated
container) 105,000
Total Transportation Costs 605,000 per month
7,260,000 per annum
Transportation Cost 17.14 per case
2.14 PKR Per Unit

Volume Forecasting
(Units)
Cumulative for Eight
Target Cities
No. of
Volume (%) Stores
Top End Retail 37 375
Walk-in Stores 44 2500
Kiryana Stores 19 16000
Top End Walk-in Kiryana
Year 1 Retail Stores Stores Total
Jan 24,000 28,541 12,324 64,865
Feb 61,720 73,397 31,694 166,811
Mar 96,000 114,162 49,297 259,459
Apr 112,000 133,189 57,514 302,703
May 120,000 142,703 61,622 324,324
Jun 120,000 142,703 61,622 324,324
Jul 120,000 142,703 61,622 324,324
Aug 120,000 142,703 61,622 324,324
Sep 120,000 142,703 61,622 324,324
Oct 120,000 142,703 61,622 324,324
Nov 120,000 142,703 61,622 324,324
Dec 120,000 142,703 61,622 324,324
Total Units Sold 1,253,720 1,490,910 643,802 3,388,432
423,554 Cases

Volume (Cases) Forecasts for Each City (Per Annum Per City for Year 1)
Per Month
Karachi 182,128 0.43 15,177
Hyderabed 21,178 0.05 1,765
Multan 21,178 0.05 1,765
Lahore 97,417 0.23 8,118
Faisalabad 38,120 0.09 3,177
Gujranwala 21,178 0.05 1,765
Rawalpindi 25,413 0.06 2,118
Islamabad 16,942 0.04 1,412
1.00
 Per Unit
(PKR)
Manufacturing Cost 75.00
Taxes and Duties 18.75
Freight Cost 7.50
Warehousing Cost 0.78
Transportation 2.14
FPLC 104.17
 Contribution Sheet

Fariha Samad
 Worked on the article “Knowledge-Based inventory management in production
logistics.”
 Worked on MTM case study
 Compiled, edited and formatted the final report

Aazar Ali Shad

 Worked on article “Impact of Inventory Management on service flexibility and
Performance.”
 Worked on “Procter and Gamble Brand Launch-Case Study”
 Got the final report printed