The Design Process

Tom Mattson

Design is the heart of engineering work. In fact, ask a practicing engineer and s/he will most likely tell you that "Engineering is design!".

ABET defines engineering design as follows:

"Engineering design is the process of devising a system, component, or process to meet desired needs. It is a decision-making process (often iterative), in which the basic science and mathematics and engineering sciences are applied to convert resources optimally to meet a stated objective. Among the fundamental elements of the design process are the establishment of objectives and criteria, synthesis, analysis, construction, testing and evaluation. The engineering design component of a curriculum must include most of the following features: development of student creativity, use of open-ended problems, development and use of modern design theory and methodology, formulation of design problem statements and specification, consideration of alternative solutions, feasibility considerations, production processes, concurrent engineering design, and detailed system description. Further it is essential to include a variety of realistic constraints, such as economic factors, safety, reliability, aesthetics, ethics and social impact."

This definition is put into practice via a Design Process.

The Design Process is the engineer's version of the Scientific Method. It is broken into 10 steps, as follows:

The Design Process
1. Identification of a Need
2. Problem Definition
3. Information search
4. Constraints
5. Criteria
6. Alternative Solutions
7. Analysis
8. Decision
9. Specification
10. Communication

Although I have presented the Design Process as a list, it should not be thought that it is linear in nature. Very much to the contrary, Design is nonlinear and iterative. That is, one may, in the process of examining one's chosen solution (Specification--Step 9), determine that the solution is not feasible after all. In that case, one may have to go back to looking at some of the other candidate solutions (Alternative Solutions--Step 6). Furthermore, one could always find oneself going back to search for more information on a given problem (Information search--Step 3).

Stay tuned, and I'll go into detail on each of the 10 steps...

wolram

i agree with your listing TOM, when it comes down to
the nitty gritty its always 4 and 6 that cause
the most problems, throw in time limitations ,
prototyping ,thats when the strain sets in.

Integral

Let's talk about reliability. So many engineers are so confident in the Reliability of their design that they forget that it WILL fail and that it WILL need to be repaired. It seems that one of the last considerations is MAINTAINABILITY, few engineers even give this consideration, the fact is that down time is a very big issue in any significant manufacturing process, the easier a tool is to repair the shorter the down time.

You would be amazed at how many million $$ tools sit doing nothing while some tech is contorting his body, spending hours to reach and replace a $100 part. Meanwhile the production line is waiting. A bit of solid design work by an engineer who is aware that his baby will fail at some point in time can save a company mega dollars. That is not an exageration or pipe dream, that is life in the semiconductor industry.

Artman

This is very true of beginning engineers, such as students just coming out of school and entering the field. Another design criteria that is often missed besides maintaining is the ability to both bring the equipment into the space it needs to occupy and get it out should it someday fail and need to be replaced. I've seen lots of designers draw up boiler replacements or chiller replacements only to have to say, "It looks like it will fit great in the room, you even have maintenance space, but how do you get it in there (down the stairs and through the three foot wide door)?" The answer they give is usually, "I didn't even think about that."

russ_watters

Thats true, but a lot of times the issue is unavoidable. So as often as not, the answer is: "well, you have two choices, either punch a hole in that wall or rip off the roof."

Artman

True, but there are alternatives that could be explored such as modular design boilers and chillers that can be walked through existing doors.

I just want to know that the problem had been considered, even if the answer, as you said, is to put a hole in the wall or in the roof.

Tom Mattson

I'd like to say a little about 2 of the steps in the Design Process.

Some human need is identified, and engineers are approached to find a technical solution. This part of the process is typically handed to the enigineer by the customer, or via his boss. Engineers who begin their design work by executing this step themselves are called "inventors".

Keep the Definition Broad
This is where the engineer takes the human need identified in Step 1 and formulates a specific problem to be solved. The trick is to not define the problem in such a way as to suggest a particular solution to the exclusion of others. To tacitly pigeonhole the problem to just one class of solutions is to prematurely stifle the creative process that is Step 6 (generation of Alternative Solutions).

For instance, in my Introduction to Engineering Design (IED) class I gave the assignment to build a device that can launch an egg and put it in a bucket 60 feet away. I warned them about possible problem definitions that could violate the precept above. That is, I told them to guard against forming a preconceived notion of what the solution is to be before going through the process, by making a problem definition such as, "Build a catapult to launch an egg into a bucket 60 feet away." (The assignment said nothing of catapults).

Cure the Sickness, not the Symptom
"For many years residential subdivisions were designed so that the rainfall would drain away quickly, and expensive storm sewer systems were constructed to accomplish this task. Not only were the sewers expensive, but they also resulted in transporting the water problem downstream for someone else to handle. In recent years perceptive engineers have designed land developments so that the rainfall is temporarily collected in "holding pools" and released gradually over a longer period of time. This approach employs smaller, less expensive sewers and reduces the likelihood of flooding downstream. The real problem was not how to get rid of the rainfall as rapidly as possible, but how to control the water.

--"Introduction to Engineering Design and Problem Solving, 2ed; Eide, Jenison, Mashaw, and Northup; McGraw Hill, 1998.

Forming Good Problem Definitions
One way to formulate a good problem definition is to view it as a transition from State A (the current State) to State B (the hoped-for State after the problem is successfully solved).

State A-->State B

In my IED class, the problem definition could well be stated as:

Eggs out of bucket-->Eggs in bucket

That is sufficient to the task, and at the same time it does tacitly not lock the team into a solution.

Stay tuned for the next Steps. In the mean time, I welcome any questions, as well as any input from our practicing engineers here.

kawai

I am new here..
I am a year one student of the Electrical and electronic engineering,
currently have assigned to write a essay about this 10 steps engineering process.Besides my lecturer ask us to find the info from the internet and i have found this site..can the post master explain the other 8 steps?

my essay topic : Explain the engineering desgin process from start to end. Include important stages and use examples in your explaination if necessarry.

I wish to hear from u soon.
Thanks a lot

FredGarvin

I think you're painting with a pretty broad brush there Integral. I can't think anyone would have the audacity to think that thir designs will not break down over time. We all have experience with after-the-fact engineering and saying that it would have been better to do this and that. That's much easier than doing it up front and trying to think of every contingency and situation for a complicated piece of equipment. If you were able to sit down with the designers of a specific piece of equipment, I would bet it's a pretty even mix between "we hadn't thought of that" and "we did it this way because of x,y and z reasons." I would venture a guess that some constraints placed on the designs had something to do with the bad choices. In my experience, most "bad designs" had external constraints painting designers/engineers into a corner with how to accomplish something. Certain mistakes like no wrench clearance or assemblies that can't be disassembled without major destruction are unforgiveable.

We are constantly in a fight with our production guys. Something that takes us R&D guys 6 steps to accomplish needs to be done with 2 in production. If you give them 2, they ***** and want it in 1 step. There is no winning in a production environment simply because of the stress those guys are under. At the same time, there is definitely diminishing returns with how much time is spent on refining designs.

Now, let's talk about automotive designers that think it's ok to have to drop an entire transmission just so you can remove an oil pan (cough cough Ford cough cough).

curiousguy79

good points

Koonism

As an Industrial controls engineer/project manager..it all has to work first on the drafting table/cadd drawings. The logic schematics will prove the concept. The most important factor is How much work one puts into the project to assure all of the i's are dotted and the t's are crossed. There is no such avenue as a short cut. Short cuts and a lack of endeavor leads to failure. Luck is for gamblers not engineers.