A COMPARISON OF CASE-BASED REASONING AND REGRESSION ANALYSIS APPROACHES FOR COST UNCERTAINTY MODELING

This paper presents cost uncertainty modeling for a concept selection using case-based reasoning (CBR) and compares this method with the regression analysis approach. During the product development stage, a number of decisions must be made under uncertainty, including selection of an ideal product concept. The cost of a concept, i.e., the cost of the final product developed from a concept, is a key factor influencing the choice of an ideal concept. The CBR approach creates a knowledge base (or database) containing past cases, defines a new case, retrieves cases similar to the new case, and adapts the solution of the retrieved cases to the new case. This paper illustrates the proposed approach using automobiles as an example.