LECTURE PACKAGE 4: DESIGN PROCESS (PART 1)

Topics Scientific Method

covered: Engineering Design Process
1. Project requirements Project requirements
2. Assumptions and constraints Assumptions and constraints
3. Background research Background research
4. Balance of trade-offs Balance of trade-offs
5. Design and optimization Design and optimization
6. Methodology and experimentation Methodology and experimentation

1. DESIGN PROCESS

Design process is an approach for breaking down a large project into manageable chunks.

STEP 1: Define the problem

 What is the challenge?
 What needs to be improved?
 What is the need?
1. You can’t find a solution until you have a clear idea of the problem
2. Consider all potential parts and related causes
3. Set and prioritise goals
SINGLES OLUTION OPEN-ENDED
Suppose that you are asked to If you change the problem statement to
determine the maximum height of a read, “Design a device to launch a 1-kg
snowball given an initial velocity and snowball to a height of at least 160 m“, this
release height. This is an analysis analysis problem becomes a design
problem because it has only one problem.
answer.

STEP 2: Collect information

1. What questions do I have?
2. What observations can I make?
3. What solutions to the problem exist?
4. How can they be improved?
5. What materials are available?
1
6. What is my solution limited by?
Method Overall Advantages Challenges comments
Purpose
questionnaires, •quick and/or •anonymous & •wording can
surveys, easy inexpensive bias client's
checklists •non •many people & responses
threatening lots of data •are impersonal
& no full story
interviews •impressions •full range and •can take much Single person at
or experiences depth of info time & costly a time
•develops •can be hard to
relationship analyze and
compare
documentation •Historic •comprehensive •often takes retrospective
review information & historical info much time
•In-depth and •information •need to be quite
wide already exists clear about
research
observation •Study things •view operations •difficult to Prospective
closely •adaptable interpret seen non-interference
behaviours
•can influence
behaviours
focus groups •discussions •quick and •can be hard to qualitative group
•understand reliable analyze (unlike interview
issues impressions responses with a single
•efficient •need good person)
facilitator for
safety and
closure
case studies •understand or •fully depiction •time consuming qualitative
depict •powerful to collect, sample size few
experiences organize & even one rare
describe events
•represents
depth of
information,
rather than
breadth

STEP 3: Brainstorm and analyse ideas

1. What are different ways to solve the problem?
2. What materials are available?
3. How can I use the available materials?
4. What is my solution limited by?

THINGS TO NOTE
1. Begin to sketch, make, and study so you can start to understand how all the data and
information you’ve collected may impact your design.
2. Apply existing/newly acquired knowledge, skills, and/or strategies that one determines to
be useful for achieving goals
3. Collaborate to encourage ideas, opinions and contributions.

2
STEP 4: Develop solutions / build a model
1. What are the strengths and weaknesses of each solution?
2. Which solution is the most useful?
3. Which solution is the least complicated?
4. How will I develop my solution?

THINGS TO NOTE
1. Take your preliminary ideas and form multiple small-scale design solutions.
2. Breaks goals into actionable steps.
3. Apply existing/newly acquired knowledge, skills, and/or strategies that one determines to
be useful for achieving goals.

Example: ALUMINIUM CAN CRUSHER

Students are being asked to design a simple device to crush aluminium cans.
A student design team proposed four solutions to the problem.
They developed six criteria that are important to a successful design.
A rating factor is assigned to each solution, based on how well that solution satisfies
the given criterion.
The rating factor is on a scale of 0 to 10, with 10 representing a solution that satisfies
the given criterion the best.
The rating factor for each design alternative is assigned according to the consensus
of the design team.
Four solutions to this problem:

1. A spring-loaded crusher
2. A foot-operated device
3. A gravity-powered dead
weight crusher
4. An arm-powered lever arm
crusher
The rating factor R is assigned
according to the following scale:
Excellent (9-10)
Good (7-8)
Fair (5-6)
Poor (3-4)
Unsatisfactory (0-2)

The student team agrees that the most important criteria (or desirable attributes) of the
design and assigned weights are
1. Safety: 30 percent (30 points)
2. Ease of use: 20 percent (20 points)
3. Portability: 20 percent (20 points)
4. Durability and strength:10 percent (10 points)
5. Use of standard parts:10 percent (10 points)

3
6. Cost:10 percent (10 points)

Design 4 was chosen the best design largely due to the rating assigned for safety, criterion
l. The team felt that the chances of human injury were negligible for this design. Since
safety is the most important factor (30% of the total weight), the high safety rating for
design 4 gives it the highest overall score (9x30, or 270).

STEP 5: Present your ideas to others for feedback

1. How should I best communicate my results?
2. Should I make a drawing and show people my prototype?
3. How should I ask for others’ feedback on my design?
THINGS TO NOTE
1. Seek input to gauge others’ understanding of the message.
2. Present your ideas to as many people as possible: friends, teachers, professionals, and
any others you trust to give insightful comments.
3. Documentations, audio and presentations

STEP 6: Improve your design

Did my design work the best that it could?
1. How could I make it better?
2. Is it practical?
3. Are the materials cheap and easy to find?
4. Does my solution create new problems, or the need for another new product?
5. Will others be able to use it equally as well?

THINGS TO NOTE
1. Reflect on all of your feedback and decide if or to what extent it should be incorporated.
2. It is often helpful to take solutions back through the Design Process to refine and clarify
them.
3. Identify alternative ideas/processes that are more effective than the ones previously
used/suggested.
4. Apply existing/newly acquired knowledge, skills, and/or strategies that one determines to
be useful for achieving goals
Example: The Wright brothers’ airplane did not fly perfectly the first time. They began a
program for building an airplane by first conducting tests with kites and then gliders. Before
attempting powered flight, they solved the essential problems of controlling a plane's motion

4
in rising, descending, and turning. They didn't construct a powered plane until after making
more than 700 successful glider flights. Design activity is therefore cyclic or iterative in
nature, whereas analysis problem solving is primarily sequential.

2. The Engineering Design Process and the Scientific Method

1.1 Engineering design

• Most engineering designs can be classified as inventions: devices or systems •

• Created by human effort and did not exist before or are improvements over existing
devices or systems.
• Inventions or designs do not suddenly appear from nowhere.
• Result of bringing together technologies to meet human needs or to solve problems.
• Design is the result of someone trying to do a task more quickly or efficiently.
• Design activity occurs over a period of time and requires a step-by-step methodology.
• What distinguishes design from other types of problem solving is the nature of both the
problem and the solution.
• Design problems are open ended in nature
• They have more than one correct solution.
• The result or solution to a design problem is a system that possesses specified
properties.
• Design problems are usually more vaguely defined than analysis problems.
o Example: Suppose that you are asked to determine the maximum height of a snowball
given an initial velocity and release height. This is an analysis problem because it
has only one answer.
o If you change the problem statement to read, "Design a device to launch a 1kg
snowball to a height of at least 20m," this analysis problem becomes a design
problem.
o The solution to the design problem is a system having specified properties (able to
launch a snowball 20m).
o Whereas the solution to the analysis problem consisted of the properties of a given
system (the height of the snowball).
o The solution to a design problem is therefore open ended, since there are many
possible devices that can launch a snowball to a given height.
• Solving design problems is often an iterative process.
• As the solution to a design problem evolves, you find yourself continually refining the
design.
• While implementing the solution to a design problem: the solution developed is unsafe,
too expensive or will not work.
• Back to the drawing board and modify the solution until it meets your requirements.
o For example, the Wright brothers' airplane did not fly perfectly the first time.
o Solution is subject to unforeseen complications and changes as it develops.
• The solution to a design problem does not suddenly appear in a vacuum.
• A good solution requires a methodology or process.
• There are probably as many processes of design as there are engineers.
• Therefore, this lesson does not present a rigid "cookbook" approach to design but
presents a general application of the five-step problem-solving methodology
associated with the design process.
• The process described here is general, and you can adapt it to the particular problem you
are trying to solve.

5
Engineering design process1

Two types of Design Process: (a) Scientific method and (b) Engineering method

Scientific method Engineering method

Scientists use the scientific method to make Engineers use the engineering design
testable explanations and predictions about process to create solutions to problems
the world.

Scientists ask questions and develop Engineers identify specific needs by

experiments asking “who needs what because why?”
scientists study how nature works, engineers create new things, such as
products, websites, environments, and
experiences
engineers and scientists have different objectives
Scientists perform experiments using the engineers follow the creativity-based
scientific method engineering design process.
Both processes can be broken down into a series of steps, as seen in the next diagrams.2

1
Ali, S.Z. (2019). https://www.quora.com/What-are-the-8-steps-of-the-engineering-design-process
2
https://www.sciencebuddies.org/science-fair-projects/engineering-design-process/engineering-
design-compare-scientific-method
6
1.2 The Scientific Method

The scientific method is a process for experimentation that is used to explore observations
and answer questions.

Steps of the Scientific Method

Ask a Question: The scientific
method starts when you ask a
question about something that you
observe: How, What, When, Who,
Which, Why, or Where?
Do Background Research: Rather
than starting from scratch … use
library and Internet research to help
you find the best way to do things and
ensure that you don't repeat mistakes
from the past.
Construct a Hypothesis: A
hypothesis is an educated guess
about how things work. It is an
attempt to answer your question with
an explanation that can be tested. A
good hypothesis allows you to then
make a prediction: "If _____[I do this]
_____, then _____[this]_____ will
happen."
Analyse Your Data and Draw a
Conclusion: Once your experiment is
complete, you collect your
measurements and analyse them to
see if they support your hypothesis or
not. Scientists often find that their
predictions were not accurate and
their hypothesis was not supported,
and in such cases they will
communicate the results of their
experiment and then go back and
construct a new hypothesis and
prediction based on the information
they learned during their experiment.
This starts much of the process of the
scientific method over again. Even if
they find that their hypothesis was
supported, they may want to test it
again in a new way.
Communicate Your Results:
Scientists publish their final report in a
scientific journal or present their
results on a poster or during a talk at
a scientific meeting.

7
1.3 The Engineering Design Process

The engineering design process is a series of steps that engineers follow to come up with a
solution to a problem. Many times the solution involves designing a product (like a machine
or computer code) that meets certain criteria and/or accomplishes a certain task.

Steps of the Engineering Design

Process
Define the Problem. The engineering
design process starts when you ask the
following questions about problems that
you observe:
•What is the problem or need?
•Who has the problem or need?
•Why is it important to solve? [Who]
need(s) [what] because [why].
Do Background Research: Learn from
the experiences of others — this can help
you find out about existing solutions to
similar problems, and avoid mistakes that
were made in the past. So, for an
engineering design project, do
background research in two major areas:
•Users or customers
•Existing solutions
Specify Requirements: Design
requirements state the important
characteristics that your solution must
meet to succeed. One of the best ways to
identify the design requirements for your
solution is to analyse the concrete
example of a similar, existing product,
noting each of its key features.
Brainstorm Solutions: There are always
many good possibilities for solving design
problems. If you focus on just one before
looking at the alternatives, it is almost
certain that you are overlooking a better
solution. Good designers try to generate
as many possible solutions as they can.
Choose the Best Solution: Look at
whether each possible solution meets
your design requirements. Some
solutions probably meet more
requirements than others. Reject
solutions that do not meet the
requirements.
Develop the Solution: Development
involves the refinement and improvement
of a solution, and it continues throughout
the design process, often even after a
product ships to customers.
Build a Prototype: A prototype is an

8
 operating version of a solution. Often it is
made with different materials than the
final version, and generally it is not as
polished. Prototypes are a key step in the
development of a final solution, allowing
the designer to test how the solution will
work.
Test and Redesign: The design process
involves multiple iterations and redesigns
of your final solution. You will likely test
your solution, find new problems, make
changes, and test new solutions before
settling on a final design.
Communicate Results: Engineers
document their solutions thoroughly, so
that they can be manufactured and
supported

Steps of the Scientific Method Steps of the Engineering Design

Process
Ask a Question: The scientific method Define the Problem. The engineering
starts when you ask a question about design process starts when you ask the
something that you observe: How, What, following questions about problems that
When, Who, Which, Why, or Where? you observe:
•What is the problem or need?
•Who has the problem or need?
•Why is it important to solve? [Who]
need(s) [what] because [why].
Do Background Research: Rather than Do Background Research: Learn from
starting from scratch … use library and the experiences of others — this can help
Internet research to help you find the best you find out about existing solutions to
way to do things and ensure that you similar problems, and avoid mistakes that
don't repeat mistakes from the past. were made in the past. So, for an
engineering design project, do
background research in two major areas:
•Users or customers
•Existing solutions
Construct a Hypothesis: A hypothesis Specify Requirements: Design
is an educated guess about how things requirements state the important
work. It is an attempt to answer your characteristics that your solution must
question with an explanation that can be meet to succeed. One of the best ways to
tested. A good hypothesis allows you to identify the design requirements for your
then make a prediction: "If _____[I do solution is to analyse the concrete
this] _____, then _____[this]_____ will example of a similar, existing product,
happen." noting each of its key features.
Brainstorm Solutions: There are always
many good possibilities for solving design
problems. If you focus on just one before
looking at the alternatives, it is almost
certain that you are overlooking a better
solution. Good designers try to generate
as many possible solutions as they can.
9
 Choose the Best Solution: Look at
whether each possible solution meets
your design requirements. Some
solutions probably meet more
requirements than others. Reject
solutions that do not meet the
requirements.
Analyse Your Data and Draw a Develop the Solution: Development
Conclusion: Once your experiment is involves the refinement and improvement
complete, you collect your measurements of a solution, and it continues throughout
and analyse them to see if they support the design process, often even after a
your hypothesis or not. Scientists often product ships to customers.
find that their predictions were not Build a Prototype: A prototype is an
accurate and their hypothesis was not operating version of a solution. Often it is
supported, and in such cases they will made with different materials than the
communicate the results of their final version, and generally it is not as
experiment and then go back and polished. Prototypes are a key step in the
construct a new hypothesis and development of a final solution, allowing
prediction based on the information they the designer to test how the solution will
learned during their experiment. This work.
starts much of the process of the Test and Redesign: The design process
scientific method over again. Even if they involves multiple iterations and redesigns
find that their hypothesis was supported, of your final solution. You will likely test
they may want to test it again in a new your solution, find new problems, make
way. changes, and test new solutions before
settling on a final design.
Communicate Your Results: Scientists Communicate Results: Engineers
publish their final report in a scientific document their solutions thoroughly, so
journal or present their results on a poster that they can be manufactured and
or during a talk at a scientific meeting. supported

TO BE NOTED

1. Keep in mind that although the steps are listed in sequential order, you will likely return to
previous steps multiple times throughout a project.
2. It is often necessary to revisit stages or steps in order to improve that aspect of a project.
3. In real life, the distinction between science and engineering is not always clear. Scientists
often do some engineering work, and engineers frequently apply scientific principles,
including the scientific method.

10
3. Project Requirements

What is the aim of the project?

 These are often poorly defined or over-defined, or sometimes even contradicting, due to
the nature of new / innovative ideas with little known background. In all engineering
projects, trade-offs typically exist (discussed below) and the optimal solution that satisfies
the project requirements are often a fine balance of these.

 An important skill is to focus down on the main / most important requirements (i.e. the
wood from the trees) and to concentrate on those. Often, these project requirements are
not drawn up by engineers and so can be very difficult to achieve, entirely. Aesthetics is
almost always a requirement. Sometimes, project requirements have social and
environmental impacts that will test an engineer's ethics and morals such as weapons
and missiles, and chemical waste.
o Example: A relevant example is the design of a rope-less hoist for South African
deep-level mines, done here at Wits University. The cable required to hoist a mine-
cage used in the deep mines weighs in excess of 14 tonnes! This requires an
enormous electric motor to lift and lower the cage, typically weighing only a fraction
of the weight of the cable. To save the huge costs in electricity, a ropeless hoist was
designed: a special linear electric motor that used permanent magnets and copper
coils along the full-length of the shaft. The benefits were notable: massive costs
savings in electricity, significantly smaller in size, had inherent braking mechanism
that could safely arrest the cage should it fall down the shaft due to a technical
failure. Despite these benefits, not a single mine worker wanted to step inside the
new cage, knowing that no rope was connected to it!
o Nuclear power stations and cellphone towers, often fall into the "not in my backyard"
category of failures, despite their proven technologies.

4. Assumptions and Constraints

Within what limits?

 In all engineering problems, constraints are ether explicitly given or exist inherently due to
the nature of the project. Financial and or time constraints are often the most prevalent,
11
 especially in the product-development industry. And often, these constraints change,
resulting in scope creep.
o Example: An interesting example of an engineering project where new constraints
became apparent after the project's commencement, leading to its ultimate demise,
is the Airbus A380 - recognised as one of the most sophisticated engineering
projects of the decade. Designed to move large numbers of passengers at the lowest
fuel cost, it was initially very successful with airlines. However, the small number of
airports with sufficiently strong runways that could tolerate its weight, limited the
number of destinations to which such an aircraft could fly to, resulting in its use in a
hub-to-hub business model - successfully used by airlines for decades. Yet, modern-
day passengers preferred a point-to-point service, which only the smaller jets could
provide. Ultimately, production of the mammoth aircraft will cease in 2021.

 Similarly, very often, not enough information is provided in a project specification and the
engineer would therefore have to assume things. Additionally, assumptions are made to
simplify complex problems.
o Example: to test the robustness of aircraft cockpit windows and jet engines to the
reality of bird strikes, a chicken gun was developed, used to fire dead chickens at
speeds of over 600km/h, at the components under test. This was achieved using a
compressed air canon, much like a giant air rifle. However, due to the inconvenient
shape of a chicken, engineers had to initially assume a spherical chicken. This
significantly simplified the design of the canon since a large tube could be used as
the barrel of the gun. In practice, the non-uniform chicken carcases are loaded into
cylindrical ice-cream tubs that fit well into the barrel.

5. Background Research

Don't re-invent the wheel!

 Good engineering often requires the development, refinement, or adaption of other
existing solutions, to satisfy the requirements. This is generally due to financial and
development-time constraints.
o Example: Very often in cars: only subtle changes (generally aesthetics) are made to
each new model; to do a total redesign would take far too long and cost far too much!

 Sometimes, this is done to avoid the exorbitant costs of re-certification, etc.

o Example: A topical example is of the Boeing 737 Max, where the entire family of 737
models is fundamentally based on the original (and only certified) 1967 Boeing 737-
100 model. Boeing insists that only minor aerodynamic and structural improvements
have been made over the decades; this to avoid the stringent requirements and
massive costs of re-certifying an airframe. However, the newer, larger, heavier engines
installed on the latest model had pushed the original design envelope too far, requiring
software interventions to prevent possible accidents on take-off. And this, ironically,
lead to the two fatal accidents that are currently in the spotlight.

 Extensive work has been done by previous engineers in all fields, and this work is well
documented; one just needs to find it by looking in the right place! Earlier work provides
future recommendations which can be used to one's benefit, either to enhance a
design, or avoid a similar mistake. Even failed engineering projects are beneficial: to
provide valuable advice to the next engineer or engineering project.

 One should remember that anyone can produce a YouTube video and or Wikipedia page,
but only in-the-field experts can produce conference papers and journal articles,
and hence these are considered credible sources. Academic institutions generally have
12
 access to journal libraries; one such common one available to Wits students is
ScienceDirect - one example of many!

6. Balance of Trade-Offs

Finding that balance

 In multi-variable engineering problems, often variables work against each other and a
benefit from one variable results in a disadvantage of another.
o Examples of trade-offs include: complexity vs reliability, size vs cost, mass vs
speed, etc.
o Example: a bus and a sports-car both solve the problem of moving people, both have
four wheels, seats, and an engine, and both cost about the same price. However, each
are good solutions to vastly different problems. In the end, size-vs-speed was the
trade-off made.

 A simple approach to dealing with multi-variable problems is to control one variable

at a time, whilst the others are held constant, to try and develop simple mathematical
relationships; or to see if the solution improves / gets worse.
o HINT: Often, the simplest solution is the best solution!
7. Design and Optimization

The BEST versus GOOD-ENOUGH

 Very often, due to constraints on development time and costs, a good-enough solution
that meets requirements is often requested. Sometimes this infringes on the quality,
reliability, and efficiency boundaries, which only become apparent at a later stage (in
some ways students' marks are similar). This sometimes leads to a patch, especially
common in software products, or a re-call of hardware (such as in cars - very expensive!).
In academic research, one often has the luxury to focus on finding the best / optimum
solution, ignoring development costs.

 The design process is somewhat iterative, incremental changes are made to the models
and or prototypes, followed by testing, until the requirements are satisfied. There are
several mathematical algorithms that can be used to quickly derive optimised
values, linear programming is one simple example.

 Often, achieving the perfect balance is not possible, and the amount of time required to
achieve near-perfection increases significantly.
o HINT: A useful guide: 90% of the project takes 10% of the time and the last 10% of the
project takes 90% of the time!

 Once a mathematical solution has been determined, a physical prototype is developed

for testing.

8. Methodology and Experimentation

What are we going to test and how?

 Once the key project requirements have been simplified and the engineering problem
identified, an experimental methodology is formulated.

13
 In engineering, by deriving mathematical models that represent real observations,
designs can be developed and optimised mathematically. However, sometimes these
mathematical models can only be derived from empirical data - test results of
multivariable systems that are difficult to model, such as fluid flow in a pipe, which
depends on friction, turbulence, viscosity, etc.

 An engineering methodology is plan of what and how one is going to test a

process or a prototype. Very often, the next step will depend on the results of a
previous step and a flowchart (a form of decision-making map) is a useful diagram to
represent this.

 Bear in mind that, often, the actual experimentation steps may have their own constraints,
such as availability, cost, accuracy, etc.
o Example: An example where an experimental process was not available or was
certainly very limited, was the design and development of the Space Shuttle. Since
very little information was known about space at the time and only a few spacecrafts
had actually been put into orbit, the Space Shuttle was predominantly developed
based on mathematical models. In publications after the first few missions, the
comparisons between actual measurements in space and the predicted models,
generally correlate quite well (well done to the engineers involved!) but in some cases,
there are quite large discrepancies too!
o HINT: It is best to start with something simple that works, before fiddling and
optimising it.

14