AN APPROACH TO DECISION-BASED DESIGN
Author: Wassenaar, Henk Jan
Supervisor: Chen, Wei
Institution: Graduate College, University of Illinois Chicago
This research is motivated by the need for developing a rigorous Decision-Based Design (DBD) framework for engineering design and the need for developing an approach to demand modeling in product design. The importance of using a single-criterion approach (profit) for design alternative selection in DBD is presented by examining the limitations of multi-criteria approaches, including the use of Pareto frontier in the presence of uncertainty. An approach to Decision-Based Design is developed as an enhancement to Hazelrigg’s DBD framework; the procedure for implementing the DBD approach is provided. The technique of Discrete Choice Analysis (DCA) is employed for constructing a product demand model, which is crucial for assessing the impact of engineering decisions on profit. To enrich the use of DCA for demand modeling of complex engineering systems, the consideration of a hierarchy of product attributes is introduced to cascade customer desires to key customer attributes that relate to the design engineer’s analysis. In addition, approaches are developed to improve the predictive capability of the demand model by more accurately assessing the general shape of the customer utility function and integrating latent variables to capture customers’ perception and attitude. This dissertation represents the first attempt of using DCA for demand modeling to assist product design decision-making. The proposed DBD approach facilitates rigorous design alternative selection, incorporates the choice behavior of individual customers, and enables the integration of business modeling and engineering analysis. The developed approaches are demonstrated by using case studies of an electric motor design and a real (passenger) vehicle engine design. The latter case study is developed in collaboration with the market research firm of J.D. Power and Associates and the Ford Motor Company.