Complex data structures in product design: a sequential approach to elicit customer perceptions

Abstract

The present paper proposes a new methodology to integrate expert judgement in new product development decision process. In particular, within the garment industry product evaluation data come mainly from judges or consumer panels. Treatment of aggregate data is difficult as some measures could seem to be contradictory. To deal with this issue the present paper proposes the application of a sequential fitting (SEFIT) approach to exploit information from the whole set of data. SEFIT methods, proposed originally by (Mirkin, 1990) attempt to explain the variability in the initial data (commonly defined by a sum of squares) through an additive decomposition terms in the model. In this case, data from expert’s evaluation of a set of garment products, concerning six predetermined fashion themes (judge perception), are treated to determine the importance level of each criterion.