Engineering Problem Solving course 1

[cbiber]

In the "boot camp" thread on this forum, there was a recent post to the effect that a course on "engineering problem solving" ought to be mandatory for freshmen. I'm an adjunct professor, and also a working engineer (though feeling underemployed at the moment), and I'd like to dig into the subject a little bit more.

What should be part of this course? In other words, what should be on the syllabus? What would the homework be like? Would there be a lab? or presentation? On what basis would a professor be able to grade the student's performance (or should there be a grade at all)? What are the key skills to be acquired? What should the student's learning goals be? What are some effective methods for teaching problem solving to those who might not have either experience or natural ability?

Thanks in advance for your input and fruitful discussion.

Cathy Biber

 

[rhodie]

I'm all for the "junkyard wars" approach:

Give students a pile of marginally useful components, present a problem, and let the free thinking and adaptation begin.

Grades would be determined upon the robustness of the solution, with consideration paid to safety and simplicity.

Then cover troubleshooting. Present a complex system that first needs to be understood; one that needs subtle adjustment or refinement. Then make the aforementioned marginally useful components available once again, and let them fix the problem.

Grades would be determined by the advent of a truly fixed system, with no aggregate problems created as a result of the solution.

This would take a lot of creativity and work (can't use the same problem again for 5 or so years!) on behalf of the instructor(s), but would have a tremendous impact on the students and their confidence.

 

[TheTick]

We actually had a probem solving course in my freshman year (at Marquette). All I can say is that problems need to be carefully written, or you end up with a bunch of "lawyer" solutions.

I may make you feel, but I can't make you think.

 

[PSE]

cbiber,

You could approach such a course in several ways, I would look at providing a variety of challenges. I would also do a team approach to performing the tasks presented.

Design- Present a need for a product that does X, the design needs to be completed within a defined time frame. Then provide time for a team presentation on the design an a peer "design review". This provides input from different points of view and gives you the opportunity to evaluate the student's skills of analysis and presentation.

Build- This could be a continuation of the first "project". Have the students actually build a design. They can then evaluate it on how long it took to fabricate, and how much it cost in materials etc.

Test- See how well the design worked and which one ended up providing the best overall performance/value.

Regards

 

[cbiber]

Thanks for your suggestions so far.

rhodie,
the "junkyard wars" remind me a bit too much of my miserable "Design Project" course sophomore year. It didn't teach me problem solving, it showed who already knew how to do it and who didn't (I didn't! But somehow I still managed to become an engineer -- I just don't happen to be building things from marginally useful components as part of my professional life!). But to follow up on your suggestions: What kind of educational component could enhance learning the "free thinking and adaptation" that is necessary to survive as an engineer? How does one "teach" various strategies for troubleshooting?

TheTick -- what was your course like, and do you feel it was effective? Can you give an example of "carefully written" and "lawyer solution"?

PSE -- what would the instructor be teaching the students while they are puzzling through their assignment? I am thinking along the lines of specific tools and tactics they can use to implement the strategy you suggest.

I'm coming from the angle that although I might have a feel for how I myself approach problem solving, but that not everyone who can successfully solve problems will have the same approach. There must be a variety of approaches available, one of which will suit your personality and preferences; one of the roles of an educator is to present the various options and let the student discover which one "clicks".

More ideas?

 

[TheTick]

Our freshman "design challenge" was to divide nto groups and invent devices that would remotely place ping-ong balls into buckets blocked by various obstacles. Not too bad a project, but most of the work (brain and hands) was done by 2 members of a 5 member team.

One team project was an industrial engineering 101 assignment. It was to streamline a process for hanging and folding shirts. Points were given for meeting certain criteria of quality as well as for time. Most of us opted for lower quality, more defects, and faster times, which yielded higher scores. The professor blew a gasket!

I think that team efforts short-change the 60+% of students who mosly ride along on such projects. Individual projects would be good. Perhaps a set of projects, so that a whole class doesn't copy one single result.

I may make you feel, but I can't make you think.

 

[BobPE]

TheTick:

I think those statistics you calculated carry through to the real worl as well...I am sure you noticed it too....LOL

All my classes involved problem solving...but not many included teaching us how to combine what we know for use in use in the real world...


BobPE

 

[PSE]

cbiber,

Your role would be to end up as a "consultant" or project manager. The goal is to let the students teach each other as you observe. You become a resource for questions that the students find themselves struggling with. You can also step in to make sure that all team members participate. The Tick points out a very good pitfall to be aware of.

Tools used could be FMEA's for reviewing the design. A spreadsheet for determining/recording the "costs". A final report done by the students reviewing the project, their chosen approach, why they chose that approach, and results obtained.

It is good to see someone out there who realizes that education is not so much as throwing the information out there and hoping some sticks as it is to identify and remove the barriers to a students own capability to learn.

Regards

 

[TheTick]

Honestly, I didn't mean to imply that the other 60% of my compadres in my workgroups were freeloaders. The structure of the projects just didn't work with groups so large. Tasks can only be effectively divided so far.


This is where I am likely to make a few enemies.
I think the way that committee-work is taught in U.S. schools is misguided. Committees work well for generating ideas and peer review, but a committee is not a team. A team has a coherent structure and a leader.

I have seen many projects fail because of excessive committee thinking. These are the projects where everybody gets their own little pet feature, resulting in a product with poor cohesion and poor performance.

One skill that needs to be taught is the ability to accept a project leader's decision and proceed to make every effort to make that direction succeed. I've seen some designers sabotage decent designs because the direction chosen was not the one they contributed to the brainstorm session a week before.

I may make you feel, but I can't make you think.

 

[josephv]

Hello Cathy,

You may want to teach "problem solving techniques", for example:

1) Understanding and identifying the parameters of the problem (e.g. what are the boundary conditions?)

2) Making the right assumptions (e.g. are temperature effects negligible when solving a stress analysis problem, sometimes they are, sometimes they are not)

3) Using Engineering Software to help solve a problem, but understanding that software has it's limitations

4) An engineer must know if a calculation is in the right "ball park" (e.g. we don't want to design a 2 ton tennis ball, because we forgot to carry the one)

5) Testing a solution

6) How to solve very difficult problems. By using the "man from mars approach" or Reductio Ad Absurdum.

7) The importance of making a cost effective solution for a customer.

8) Brainstorming

9) Communication styles. The engineers will work with difficult customers at times, an engineer must understand different communication styles.

10) You knew this was coming! An engineering student must be able to present a solution in front of the class and write a report about it.

Cheers,

Joseph

 

[plusser]

There is one thing I would think be an excellent topic from a problem solving course, obsolescence.

The problem is that while domestic equipment gets cheaper and the product lifespan gets shorter, the length of time a component may be in production may be months rather than years.

This is a problem for those involved in producing equipment for the "Professional Market", that may have an extended life expectancy of 5 years in production and 25+ years in service. When you look at B52 bombers, they were designed in the fifties and are still in use today.

In many cases, emulation of a microprocessor may be possible, but when the re-qualification of saftey critical software is involved, the costs will start to sprial, so effective choices will need to be examined.

This would help the attendee to understand the potential pitfalls of designing using the latest technology with no proven roadmap for technology upgrades in the future.

 

[rhodie]

plusser

Great point about the B-52s. You have to point out, however, that the H models (or whatever letter the series has progressed to) that fly today are not the same airframes built in the 50's.

What you are saying has a lot of merit, though.
Designed obsolescence is a primary consideration in many industries. You can buy a new washing machine for only US$350, but you can bet that it was designed to last only 5-7 years. The OEM tooling for replacement parts are destroyed after this time, therefore you end up buying another $350 worth of washing machine.

Sometimes safety limits the designed service life of a product. You don't hear of too many 30 year old buried gasoline storage tanks still in use. "Forced" obsolescence can be a good thing.

 

[sc]

Back to problem solving at Uni. My experience was relatively positive as we were often fed the theory and then taught to apply it via solving problems, hence we were always problem solving.

Seriously though, in my experience the problem soving was restricted to the mech/elec streams with the civil/mining surviving on a diet of theory, calculations, assignments, etc. What little problem solving we did was in final year and was based around solving design/construction problems by knowing what theory to combine with the practical aspects of engineering.

So I would suggest that in my case the civil course needed to get their students some real world problems, not just straight design issues. I don't think a specific subject/class is the way to go, I think problem solving needs to be combined with the design theory.

reagrds
sc