A Methodology for Model Selection in Engineering Design

Engineering design consists of a series of stages during which a number of decisions need to be made by the designer. Since the information available to the designer is limited during initial design stages, to make these decisions and be able to proceed further in the design process, the designer needs to depict the nature, visualize the form, and predict the behavior of the product through the use of aids called models. These models guide these decisions, therefore, the designer needs to ensure the downstream validity of these decisions by constructing models with sufficient accuracy and resolution. Because higher quality and accuracy of information is most often accompanied by a higher cost for a model, determining a satisfactory level of goodness for a model is a fundamental and pervasive question in engineering. Hence, a key aspect of design model construction is deciding whether a model is appropriate for a particular design specification or evaluation, considering accuracy and cost factors. This paper presents an approach for design model construction using utility theory. Since model selection is a design decision, uncertainties in parameters and models are considered by evaluating the confidence in the selection of any model. A method for proceeding in the reverse manner to determine the required goodness of a model is also discussed. We present this research through application to a race car sway bar.

[1]  Kristin L. Wood,et al.  Estimating errors in concept selection , 1995 .

[2]  A. Sveshnikov,et al.  Problems in Probability Theory, Mathematical Statistics and Theory of Random Functions , 1979 .

[3]  R. L. Keeney,et al.  Decisions with Multiple Objectives: Preferences and Value Trade-Offs , 1977, IEEE Transactions on Systems, Man, and Cybernetics.

[4]  F. B. Vernadat,et al.  Decisions with Multiple Objectives: Preferences and Value Tradeoffs , 1994 .

[5]  Joseph Edward Shigley,et al.  Mechanical engineering design , 1972 .

[6]  Kristin L. Wood,et al.  THEORETICAL FOUNDATIONS FOR TUNING PARAMETER TOLERANCE DESIGN , 2000 .

[7]  William L. Oberkampf,et al.  Uncertainty quantification in computational structural dynamics : A new paradigm for model validation , 1998 .

[8]  Deborah L Thurston,et al.  Fuzzy Ratings and Utility Analysis in Preliminary Design Evaluation of Multiple Attributes , 1992 .

[9]  E. Rowland.  Theory of Games and Economic Behavior , 1946, Nature.

[10]  Deborah L Thurston,et al.  Optimization of design utility , 1991 .

[11]  Sven Erik Magnusson.  Uncertainty analysis : Identification , Quantification and Propagation This report is financed by BRANDFORSK , 1997 .

[12]  David Draper,et al.  Assessment and Propagation of Model Uncertainty , 2011 .

[13]  George A. Hazelrigg,et al.  On the role and use of mathematical models in engineering design , 1999 .

[14]  E. Antonsson,et al.  The Method of Imprecision Compared to Utility Theory for Design Selection Problems , 1993 .

[15]  Deborah L Thurston,et al.  Quantifying the house of quality for optimal product design , 1994 .

[16]  Kevin Otto,et al.  Measurement methods for product evaluation , 1995 .

[17]  Rajesh Radhakrishnan,et al.  A framework for sufficiency estimation of engineering design models , 2002 .

[18]  Karl T. Ulrich,et al.  Product Design and Development , 1995 .