Conjoint analysis has been a widely used method for measuring customer preferences since 1970s. This method is based on the idea that customers' decisions depend on all tangible and intangible product features. One of the fundamental steps in performing conjoint analysis is the construction of experimental designs. These designs are expected to be orthogonal and balanced in an ideal case. In practice, it is hard to construct optimal designs and thus constructing of near optimal and efficient designs is carried out. There are several ways to quantify the relative efficiency of experimental designs. The choice of measure will determine which types of experimental designs are favoured as well as the algorithms for choosing efficient designs. In this paper, we propose a simultaneous algorithm which combines two optimality criteria: standard criterion named by A-efficiency, and non-standard criterion, P-value. The algorithm was implemented as the procedure in MCON software for traditional conjoint analysis. The computational experiments were made and results were compared with results of SPSS® procedure. It was shown that, unlike the SPSS® procedure, MCON procedure could construct designs that contain an arbitrary chosen number of profiles. Furthermore, it was confirmed that the designs constructed using MCON procedure are not just highly efficient but also better balanced than those constructed by SPSS®.
Key words: Conjoint analysis, efficient experimental design, optimality criteria, approximate algorithm, MCON procedure.
Copyright © 2023 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0