An intelligent decision theoretic approach to producibility optimization in conceptual design

Product optimization involves selecting design, manufacturing, and support attributes that can produce the best system. Producibility or manufacturability is the term often used to describe the relative ease of manufacturing a product. In complex systems, productibility optimization is a very difficult process, particularly when the values of many attributes are restricted by constraints. One challenge is to develop more universal producibility metrics for the conceptual design phase when design information is limited and drawings are nondimensional. This paper develops a new method for producibility optimization in conceptual design based on a combination of both decision theoretic and expert system techniques. Decision theoretic techniques provide the means to model the design for producibility problem in a manner that can deal with risk, uncertainty, and user (or corporate) preferences, and can effectively integrate diverse factors to provide a measure of the overall worth of a design. The particular decision theoretic approach employed is based on multi-attribute utility theory. An illustrative example of the methodology is applied to the conceptual design of a structural composite part.