A neurofuzzy-evolutionary approach for product design

This study investigates a systematic approach to product design based on artificial intelligence. This investigation proposes the use of artificial intelligence techniques, including fuzzy theory, back propagation neural networks (BPN), and genetic algorithms (GA), along with morphological analysis to synthesize, evaluate and optimize product design. This study focuses on (1) how to model imprecise market information by applying fuzzy theory; (2) mapping relationships between design parameters and customer requirements using BPN; (3) synthesizing design alternatives by morphological analysis, and (4) realizing the synthesis in GA, using its searching capacity to obtain the optimal solution. Two case studies illustrate the practical value of the proposed methodology.