Using Fuzzy Non-linear Regression to Identify the Compensation Level among Customer Requirements in QFD

The QFD method is widely used in the product preliminary design and competitive analysis of different products. The overall customer satisfaction is usually set as the objective and principle of the design planning and evaluation of products. As an MADM problem, in generally, the calculation of the objective value relies on the specification of importance weights to accomplish trade-offs among different CRs. To date, many traditional methods are developed to identify the important weights of CRs, including AHP, ANP, FAHP, FANP, NLP and so on. Most of them take no consideration of the compensation level among CRs, which is an important foundation of trade-off strategy in MADM except importance weights of attributes. The neglect of that would leads to misjudge the products to be evaluated or mislocate resource among enterprise characteristics and eventually, conduct the company make wrong decisions. In this paper, we employ the method of imprecision (MoI) in QFD to surmount this inability, and establish a fuzzy non-linear regression model to identify the compensation level s among customer requirements. At last, an illustrative example is provided to demonstrate the application and performance of the modeling approach.