The Application of Kansei Engineering System for Vehicle Interior Color Design

Product innovation design is a classic problem in industry. Traditionally in the manufacture of vehicle, the focus has always been maker-centric design. The next decade is likely to witness a considerate rise in human-center-design. Especially, previous studies in vehicle interior have not dealt with the meaning of color. This paper tries to build an expert system, which is named as Kansei engineering system with qualitative and quantitative measurements which is used for vehicle interior color design. The Kansei engineering system (KES) has two kinds of operation, called Kansei engineering extraction system and Kansei engineering evaluation system. The former applies optimization to output the suitable design, and the latter is used to do semantic prediction. However, the similar system has already been existed in different fields, but the KES in vehicle interior color research is still limited. In optimization step, this study utilized genetic algorithm-backpropagation neural network to do design selection which might replace supervisors' work. Evaluations were built by training BPNN, GA-BPNN, SVM and GA-SVM to prediction the image ranking. A case about vehicle interior was provided to complete the computation. Results showed that these methods could be inspired to designers facing huge workloads not only automatic color design but other product elements as well. It is evidently clear from the findings that artificial intelligence might replace designer to make decision in conceptual or other design processes.