A semantic recognition-based approach for car's concept design

A semantic recognition-based method for aiding the designer to do automotive concept design is addressed in this study. The human psychological responses to forms and colours of cars are quantified using membership functions. The effects of forms and colours on the image perceptions a person has of a car are analysed with the technique of analysis of variance (ANOVA). A consultative computer program for offering basic design information (form, colour, and design criteria) is constructed. With the aid of this program, five coloured car styles (arranged by size of membership functions) and the corresponding design criteria are suggested to the designer after he/she inputs the demanded image word/words for specifying the characteristic of a car in the concept design stage. In this way, not only the conceptual design is mastered by the designer more rapidly and the lead time of design stage is shortened, but the consumption psychology of consumers is grasped more precisely. And consequently, a product that more closely fits the demanded characteristics for most of the consumers can be designed in a shorter time.