Linking quality function deployment with conjoint study for new product development process

Conjoint analysis (CA) is a popular marketer's tool for new product design. Quality function deployment (QFD) is another approach, frequently used by engineers, for design of new product. Typically, in a conjoint study, the attributes and their levels are determined through focus group discussion or market survey. On many occasions, the researchers leave out some of the more critical features altogether or include attributes with unrealistic sets of levels resulting in infeasible product profiles. In QFD, on the other hand, the new product development team attempts to identify the technical characteristics (TCs) that should be improved or included to meet the customer requirements (CRs) by using a subjective relationship matrix between CRs and TCs. QFD is not used to determine the attributes and their levels. As a result, more often than not, QFD captures what product developers "think" would best satisfy customer needs. In this paper, we link QFD with conjoint and propose a framework for objectively determining the attribute levels using the QFD approach for subsequent use in a conjoint study. For this purpose we obtain the so-called relationship matrix in QFD in a particular way that facilitates achieving our objective. We formulate an integer-programming problem for maximising the weighted sum of improvements in the product, subject to budgetary constraint and minimum percentage improvement for each or some of the attributes. We apply the framework for a commercial vehicle design problem with hypothetical data.