Design optimisation for customers' KANSEI requirement: application of interactive reduct evolutional computation to industrial design with curves

Abstract

We developed a design methodology for evolving on user's personal preference using Interactive Reduct Evolutional Computation (IREC). In this method, a user interacts with a computer system with IREC, which generates design samples. A user evaluates and scores each sample depending on his/her psychological preferences. The system estimates the design attributes that the user pays more attention to (preferred features) with reduct in rough set theory and generates new design samples. This interaction is continued until the sample converges to a satisfactory design. The objective of the IREC-based interactive design system is to help customers to externalise their implicit KANSEI (image of product or psychological feeling) into shape. To demonstrate its usefulness, it is applied to an automobile sideboard shape design. 3D computer model of automobile body is generated by the output of IREC. IREC is enhanced to support global feature as design attributes. Since IREC-based design system is applied in early stage of design global feature such as convexity and softness are important. The effectiveness of the method is demonstrated by the experiments carried out by non-designers. Commonality of preferred features among subjects is discussed with records of IREC's operation.