A higher satisfaction product customization method for different customer groups

Product customization in response to the widely varying perceptual expectations of different customers is an effective means to obtain customer satisfaction. Rapidly optimizing products to meet the personalized needs of different customers while controlling costs and improving design efficiency remains a challenge. In the present paper, a higher satisfaction product customization model was proposed with aim of efficiently reducing the number of target research customers and rapidly generating customer-oriented product design. In this model, a clustering algorithm based on emotional preference and migratory behavior (EPMC) was combined with a coupled model of interactive genetic algorithm with hesitancy-based interval individual fitness (IGA-HIIF) and Kansei Engineering (KE) method. This is the first implementation of EPMC in the customer research field, and the output content is optimized, which makes it capable of quickly dividing the multi-dimensional customer data space constructed in this study and accurately identifying different types of representative customers (RCs). And then, coupled the IGA-HIIF and KE model was adopted to establish a customer-oriented product evolution design system (PEDS). Finally, the most satisfactory products were auto-generated by PEDS with the participation of the RCs. The proposed method was applied to a case of social robot design. The result verifies that the approach could reduce the number of target research customers effectively without reducing diversity, allow direct customer involvement in the design process, and accurately extracts customers' personalized emotional implicit information and preference differences.