Rough set and PSO-based ANFIS approaches to modeling customer satisfaction for affective product design

28 Facing fierce competition in marketplaces, companies try to determine the optimal settings of design 29 attribute of new products from which the best customer satisfaction can be obtained. To determine 30 the settings, customer satisfaction models relating affective responses of customers to design attributes 31 have to be first developed. Adaptive neuro-fuzzy inference systems (ANFIS) was attempted in previous 32 research and shown to be an effective approach to address the fuzziness of survey data and nonlinearity 33 in modeling customer satisfaction for affective design. However, ANFIS is incapable of modeling the rela34 tionships that involve a number of inputs which may cause the failure of the training process of ANFIS 35 and lead to the ‘out of memory’ error. To overcome the limitation, in this paper, rough set (RS) and par36 ticle swarm optimization (PSO) based-ANFIS approaches are proposed to model customer satisfaction for 37 affective design and further improve the modeling accuracy. In the approaches, the RS theory is adopted 38 to extract significant design attributes as the inputs of ANFIS and PSO is employed to determine the 39 parameter settings of an ANFIS from which explicit customer satisfaction models with better modeling 40 accuracy can be generated. A case study of affective design of mobile phones is used to illustrate the pro41 posed approaches. The modeling results based on the proposed approaches are compared with those 42 based on ANFIS, fuzzy least-squares regression (FLSR), fuzzy regression (FR), and genetic 43 programming-based fuzzy regression (GP-FR). Results of the training and validation tests show that 44 the proposed approaches perform better than the others in terms of training and validation errors. 45 2015 Published by Elsevier Ltd. 46 47 48 49