Constructing a hybrid Kansei engineering system based on multiple affective responses: Application to product form design

This study proposes an expert system, which is called the hybrid Kansei engineering system (HKES) based on multiple affective responses (MARs), to facilitate the development of product form design. An HKES is consists of two sub-systems, namely the forward Kansei engineering system (FKES) and the backward Kansei engineering system (BKES). The FKES is utilized to generate product alternatives and the BKES is utilized to predict affective response of new product designs. Although the idea of the HKES and similar hybrid systems have been applied in various research fields, such as product design, engineering design, and system optimization etc., most of existing methodologies are limited to search optimal design solutions using single-objective optimization (SOO), instead of multi-objective optimization (MOO). Hence the applicability of the HKES is limited when adapted to real-world applications, such as product form design discussed in this paper. To overcome this shortcoming, this study integrates the methodologies of support vector regression (SVR) and multi-objective genetic algorithm (MOGA) into the scheme of HEKS. The BKES was constructed by training SVR prediction model of every single affective response (SAR) using basic product form samples as input data and the affective response score gathered from the questionnaires as output values. The FKES generates optimal design alternatives based on MOGA-based searching method according to the MARs specified by a product designer as the system supervisor. A case study of mobile phone design was given to demonstrate the analysis results. The proposed HKES based on MARs can also be applied to a wide variety of product design problems, as well as other MOO problems involving with subjective human perceptions.