5 Fundamentals of Engineering Design
WHAT IS ENGINEERING DESIGN?
We have already established that engineers require many technical and non-technical skills in order to do engineering. The “doing” of engineering, however, requires the application of these engineering skills in order to create a device or system. Engineering Design can then be described as the act of combining all of your engineering skills and knowledge to create a solution to a problem, be it a physical object, a process, or a system.
This process tends to be very systematic, and highly iterative in order to minimize risk and optimize the problem-solving process. There are many different approaches to methodologies, but all of them aim to create the same thing, a solution to the problem. These frameworks are also extremely useful as problems in the real world tend to be very open-ended, while having constraints based on time, budget, safety, and so on.
While there may be many ways to approach different problems, one of the roles of an engineer is to find optimal solutions, and implement them where possible. To find optimal solutions, it is very important to fully define the problem by exploring the problem from different perspectives for any given situation.. These perspectives include factors such as wants of a user, or environmental concerns. Additional perspectives that might be considered are described in the box below [1].
- Facts vs Assumptions – When trying to fully grasp the situation at hand, it is important to distinguish between what is a fact and what is an assumption. Facts are can be hard constraints, and thus helps to constitute restrictions or requirements on the project, while assumptions are a bit more versatile.
- Symptoms vs Contributing Factors – Treating the symptoms of problems often leads to a temporary solution at best. When trying to come up with an optimal solution, the root causes of the problem should be addressed, so that the issue may be truly resolved.
- Stakeholders – All projects will involve stakeholders in some way. Their wants tend to be very important, as they have to agree to, and sign off on all potential solutions. On the other hand, their desires may not completely align with the needs of end users, or even the suggestions of engineers. An optimal solution aims to keep all individuals satisfied, with emphasis on the core audience, which are the people actually using the design/system being developed.
- Users and Tasks – Technical requirements of a project can sometimes cloud the judgement of the engineers designing a solution. A user-centered approach is often times better, as human operators are usually necessary for any solution, even if just for supervision and maintenance.
- Environmental Effects – This factor is very important, and also very broad. Minimizing environmental effects falls under a much larger topic of sustainability. Newer devices or technologies might be very hard to predict, and thus can not be completely safeguarded against, but it is still the job of the engineers to prevent as much harm as possible.
- Possible Solutions – Many solutions should be identified and considered when trying to solve a problem. There are often many ways to approach a problem, and thus you must always consider the fact that there may be a different, or even better approach, than the one you are currently trying. The most optimal of these solutions must be selected, and the method of selection must be logical and thorough. If it comes down to it, all of these decisions need to be defensible, even in a court of law, as engineers have obligations to uphold public safety.
SYSTEMATIC DECISION MAKING
Engineers often work in groups, be it a small team, or in a larger organization. To come up with solutions, you may have heard of methods to generate multiple ideas in the past, such as brainstorming, or even perhaps through mind mapping. How to narrow down these ideas is often the less talked about portion. To convince others that a solution is the most optimal, and to narrow down the ideas logically and systematically, we must employ some other tools.
To begin comparing any two things, you must first establish the basis on which they will be assessed. For example, in the design of a physical device, these factors may include manufacturing cost, portability, functionality, and so on. When comparing problem solutions, it is often beneficial to have a relatively comprehensive list of all important factors on which the two solutions must incorporate.
The most common method that you might even use in everyday life is a simple pros and cons list. When you have a good understanding of the problem, and the variables surrounding it, you can list possible solutions along with their pros and cons. For basic situations, simply choosing the solution whose pros and cons best meet all the judgment criteria/constraints, makes it a fairly optimal solution.
In more complex scenarios, there may be multiple solutions with similar advantages and disadvantages. There may even be solutions with comparable quantities of advantages, but in totally different aspects of operation or design. This is where more complex comparison schemes are necessary, such as pairwise comparison. For pairwise comparison to be possible, we must establish a baseline solution against which we can evaluate other solutions. This baseline can be a device or system currently in use, or one of the possible solution candidates. Additionally, judgment criteria must be decided on before evaluation can begin, as these are the factors we will be evaluating for each solution.
Once the baseline is selected, a table is constructed with the possible solutions (options) as the rows, and the judgment criteria as the columns. The reference is set as the first row in the table, and all other potential solutions are scored based on how it compares to the reference for each judgement criteria.
- Better than the reference: +1
- Same as the reference: 0
- Worse than the reference: -1
| Solutions | Manufacturing Cost | Portability | Functionality | Total |
| Option 1 | REFERENCE | 0 | ||
| Option 2 | +1 | 0 | 0 | +1 |
| Option 3 | +1 | -1 | 0 | 0 |
In this example, there are 3 Solutions and 3 judgement criteria. Option 1 was chosen as the reference, and is assigned a score of 0 as a baseline. All other options are compared against this baseline and scored accordingly. The comparative option evaluation is assessed by simply summing their respective rows. Option 2 had better manufacturing costs (+1), but the same portability (0) and functionality (0) as our reference, resulting in a final score of +1. Option 3 on the other hand, had better manufacturing costs (+1), but worse portability (-1), with the functionality remaining the same (0) as the reference, resulting in a neutral final score of 0. Based on these comparison criteria, Option 2 seems like the optimal solution, with options 1 and 3 being similar in effectiveness.
COMPUTATIONAL DECISION MAKING
Decision making can require a more nuanced evaluation or comparison of criteria. For example, the importance of each criteria may not be equal, and the degree to which a solution meets any specific criteria may be defined on a relative scale of performance (not just yes or no). If this is the case, the method of “computational decision making” using tables or mathematical formulae can be utilized to capture these differences, and help evaluate the optimal design.
Setting Relative Criteria Weighting. The first step is to assign your design criteria percentages. These percentages should be in accordance with their importance to the final solution, and they should all add up to 100%. For example, if you were trying to decide on a computer to purchase, some important criteria with their respective weightings could be as follows:
- Cost (35%)
- Processor Speed (35%)
- Battery Life (20%)
- Weight (10%)
It is important to note that these weighted are not necessarily the same for everyone. Another person person selecting a computer might not be as sensitive to cost, and feel that the processor speed is more important. Another person might also consider other features, such as screen size or screen type.
Criteria Performance Rating. To identify the degree to which a specific solution satisfies a criteria, a rating table can be created to assign “points” for a specific level of performance for each criteria. For example, in your assessment, you may identify that a computer that costs less than $200 would be the ideal scenario. Thus, on a scale of 0 to 10, with 10 representing the best score possible, a computer that cost less than $200 would be assigned a performance score of 10. A computer that costs $600 would be less optimal, and could be assigned a score of 8. And so on. This type of rating evaluation is then created for each of the criteria. An example of a completed set of criterion rating scales for all criteria is provided in the table below. Note, that similar to the setting of the relative criteria weighting exercise, these scores can be assigned differently from individual to individual or project to project, based on the specific circumstances or need.
| Rating | Cost
($) |
Processor Speed
(GHz) |
Battery Life
(Whr) |
Weight
(lb) |
| 10 | < 200 | > 5 | > 60 | < 2 |
| 8 | 600 | 4 | 50 | 2.5 |
| 6 | 1200 | 3 | 40 | 3 |
| 4 | 1800 | 2 | 30 | 3.5 |
| 2 | 2400 | 1 | 20 | 4 |
| 0 | 3000 | < 1 | < 20 | > 5 |
Selecting the Optimal Solution: With relative criteria weightings and criteria performance rating establish, a weighted calculation can be performed for each solution (in this case Computer Options) can be performed to identify the optimal solution. In this case, the optimal solution will be the one that has the highest weighted score. Using the information in the tables above, a computer (Option 1) with a price of $2,400, a processor speed of 3 Ghz, a battery life of 50 Whr and a weigh of 2.5 lb would result in a total weighted score for all criteria of (0.35)(2)+0.35(6)+(0.2)(8)+(0.1)(8)= 5.2. This analysis could be carried out for 4 computer options as shown in the table below.
| Criteria | Criteria Weighting | Option 1 | Option 2 | Option 3 | Option 4 | ||||
| (wij)(rij) | fj | (wij)(rij) | fj | (wij)(rij) | fj | (wij)(rij) | fj | ||
| Cost | 35% | (0.35)(2) | 0.7 | (0.35)(8) | 2.8 | (0.35)(6) | 2.1 | (0.35)(8) | 2.8 |
| Processor
Speed |
35% | (0.35)(6) | 2.1 | (0.35)(4) | 1.4 | (0.35)(8) | 2.8 | (0.35)(4) | 1.4 |
| Battery
Life |
20% | (0.2)(8) | 1.6 | (0.2)(6) | 1.2 | (0.2)(8) | 1.6 | (0.2)(0) | 0 |
| Weight | 10% | (0.1)(8) | 0.8 | (0.1)(8) | 0.8 | (0.1)(2) | 0.2 | (0.1)(10) | 1.0 |
| Total | 100% | 5.2 | 6.2 | 6.7 | 5.2 | ||||
Based on the decision matrix, assuming the criteria weighting and ratings are fully representative, Option 3 is the optimal choice with a weighted score of 6.7/10, and Option 2 is the second choice with a weighted score of 6.2/10. The decision matrix could be tested for its sensitivity by changing the criteria weights or ratings. It is also important for you to reflect on the outcome of the matrix, as you ultimately decide on the solution. For example, in the end, you may decide to select Option 2 because the weight of Option 3 might be too heavy and an you would prefer something lighter, despite the outcome.
REFERENCES
[1] Andrews, G.C. et. al, Introduction to Professional Engineering in Canada, 2019.
