A fuzzy stepwise regression for developing customer preference models in new product development

Fuzzy regression methods have commonly been used to develop consumer preferences models which correlate the engineering characteristics with consumer preferences regarding a new product; the consumer preferences models provide a platform whereby product developers can decide the engineering characteristics in order to satisfy consumer preferences prior to developing the products. Recent research demonstrates that these fuzzy regression methods are able to partially address the characteristics of customer preferences including fuzziness and nonlinearities. However, these researches have a common limitation in that they do not investigate the appropriate polynomial structure which includes significant regressors that consist only of significant engineering characteristics or significant nonlinear terms. The inclusion of insignificant regressors is not an effective approach when developing the models. However, the exclusion of significant regressors may affect the generalization capability of the consumer preference models. In this paper, a novel fuzzy modelling method, namely fuzzy stepwise regression (F-SR) is proposed, in order to develop a customer preference model which is structured with an appropriate polynomial in which only significant regressors are included. Based on the polynomial structure, the fuzzy coefficients are determined using the fuzzy least square regression in order to address the uncertainties caused by the fuzziness and randomness of consumer preferences. The effectiveness of the F-SR is evaluated based on a case study of a tea maker design, where customer preferences regarding tea makers are nonlinear and the observations of the consumer preferences are uncertainties which are caused by the fuzziness and randomness of the quantitative evaluations. Results show that better generalization capabilities with more effective incorporation of engineering characteristics can be obtained compared with the fuzzy regression methods commonly-used for new product development. Also smaller scale consumer preference models with fewer engineering characteristics can be obtained. Hence, a simpler and more effective product development platform can be provided through the consumer preference models.