Over the past few years Bahill has worked with several major aerospace companies. Forthese companies he presented seminars on Systems Engineering, worked on productdevelopment teams and helped them describe their Systems Engineering Process. At eachcompany, he asked for examples of tradeoff analyses that had been done there, that hecould take them back to the University of Arizona and present to his students.
He has not yet found one.
A similar lack of tradeoff studies was found by Szidarovszky working with engineers in Hungary, Brazil and China. The suspicion is that systems engineers inindustry do not do quantitative tradeoff analyses.
When an engineer performs a tradeoff analysis and submits the results to his or herboss, the boss often says "No, that is not the right answer." In response tothis the engineer might change weights, values, scoring functions or tradeoff functions.Part of the reason that engineers do not do tradeoff studies may be that they have seenapplications where similar analyses using similar techniques have given differentpreferred alternatives. Tradeoff analyses fall into the field of decision making underrisk or uncertainty, often called utility theory or multicriterion decision making.
In this workshop the techniques of multicriterion decision making will be applied tothe system design process. It will be determined which techniques are most appropriate forwhich system engineering problems. Multicriterion decision theory is an old field withlots of researchers, lots of papers, lots of books and lots of practitioners, especiallyin the field of public policy, particularly in Europe. However sophisticatedmulticriterion techniques are not included in systems engineering textbooks and they arenot used by systems engineers.
In this workshop eight techniques for multicriterion decision making will be studied.
First, The mathematical basis of each of these eight techniques will be explained indetail.
Second it will be proven that each technique as used satisfies Ron Howard's list ofdesiderata that are needed for a technique to be a valid expected utility maximizationnormative rule.
Third, different techniques are based on different ways of expressing the preferencesof the decision makers, and in most cases these preferences are not consistent with eachother.
We will present methods that allow us to produce consistent preference orders byminimally changing those obtained from the decision makers.
Thus discrepancies between the results obtained by different techniques will beeliminated by the minimal alteration of the preference orders given by the decision makersand by adding requirements.
Fourth, discrepancies between recommendations of the techniques and the actualdecisions of the human decision makers will be explained.
Fifth, when the alternative design evaluations produce a nonconvex distribution, normalgradient search techniques will not work and bizarre recommendations will be produced bymost techniques.
Therefore, this workshop will show what types of system designs yield nonconvexalternative distributions and what types yield convex distributions.
And finally, we will test our hypothesis that when all the critical requirements havebeen captured, then the many tradeoff techniques will converge and produce one preferredalternative.
Intended Audience: Systems engineers who are supposed to do tradeoffstudies.
Class Size: Limited to 40 participants. Instructor:Terry Bahill was born in Washington, PA, on January 31, 1946. Hereceived a B.S. in electrical engineering from the University of Arizona, Tucson, in 1967,an M.S. in electrical engineering from San Jose State University, in 1970, and a Ph.D. inelectrical engineering and computer science from the University of California, Berkeley,in 1975. He served as a Lieutenant in the U.S. Navy, and as an Assistant and AssociateProfessor in the Departments of Electrical and Biomedical Engineering at Carnegie MellonUniversity, and Neurology at the University of Pittsburgh.
Since 1984 has been a Professor of Systems and Industrial Engineering at the Universityof Arizona in Tucson. His research interests include systems engineering theory, modelingphysiological systems, eye-head-arm coordination, the science of baseball, the systemdesign process, validating expert systems, and quality function deployment.
He has published over 150 papers and has lectured in a dozen countries. His researchhas appeared in Scientific American, The American Scientist, The Sporting News, The NewYork Times, Sports Illustrated, Newsweek, Popular Mechanics, Science85, Highlights forChildren, Rolling Stone, and on the NBC Nightly News. He is the author of Bioengineering: Biomedical, Medical, and Clinical Engineering, Prentice-Hall, 1981, KeepYour Eye on the Ball: The Science and Folklore of Baseball (with Bob Watts), W.H. Freeman,1990, Verifying and Validating Personal Computer-Based Expert Systems, Prentice-Hall,1991, Linear Systems Theory, (with F. Szidarovszky), CRC Press, second edition, 1998,Engineering Modeling and Design, (with Bill Chapman and Wayne Wymore), CRC Press, 1992,and Metrics and Case Studies for Evaluating Engineering Designs, (with Jay Moody, BillChapman and David Van Voorhees) Prentice Hall, 1997.
For the IEEE Systems, Man, and Cybernetics Society he has served as vice president, asassociate editor, as Program Chairman for the 1985 conference in Tucson, and as ConferenceCo-chairman for the 1988 conference in Beijing and Shenyang, China.
He served as an associate editor for the Computer Society's magazine, IEEE Expert. Heis serving as an associate editor for INCOSE's Systems Engineering Journal.
He holds U.S. patent number 5,118,102 for the Bat Chooser, a system that computes theIdeal Bat Weight for individual batters.
He is a Registered Professional Engineer and a member of Tau Beta Pi, Sigma Xi and PsiChi. He is the Editor of the CRC Press Series on Systems Engineering. He is a Fellow ofThe International Council on Systems Engineering (INCOSE) and of The Institute ofElectrical and Electronics Engineers.