Using fuzzy non-linear regression to identify the degree of compensation among customer requirements in QFD

• The degree of compensation among customer requirements in QFD is discussed for the first time.
• The overall customer satisfaction is derived with the trade-off strategy consisting of the relative importance weights of CRs and the degree of compensation among them.
• We develop a fuzzy non-linear regression method to identify the degree of compensation among CRs.
• An example for customer satisfaction analysis of motor car is studied to illustrate the application of the proposed approach.

As an effective customer-driven approach, the quality function deployment (QFD) takes numbers of customer requirements (CRs) into account in the process of the initial product design and the competitive analysis. It is a traditional multi-attribute decision making problem, and the trade-off strategy among CRs which is interpreted as decision parameters, is crucial for resulting the overall customer satisfaction. Although the general trade-off strategies concern about the importance weights of CRs, which are specified with a variety of methods, they ignore the influence of the degree of compensation among them. In this paper, we embed the degree of compensation among CRs into QFD, which is expressed as a symmetric triangular fuzzy number, and develop a fuzzy non-linear regression model using the minimum fuzziness criterion to identify it. Furthermore, an illustrative example is provided to demonstrate the application and the performance of the modeling approach. It can be verified from the experimental results that the overall customer satisfaction as well as the prioritization of products are affected by the degree of compensation among CRs. Meanwhile, against to the products in example, the overall customer satisfaction obtained with the traditional weighted-sum method is confirmed to be underestimated.