Application of Taguchi Method to Facility Location Selection Problem in a Retail Sector

Abstract

In this study, the site selection problem for a supermarket chain is examined in Tarsus, Mersin of Turkiye. We consider four key factors: the new branch's area, parking capacity, local population density, and nearby competitor count. We employ three methods: TOPSIS, VIKOR, and the Taguchi method. While TOPSIS and VIKOR methods traditionally involve ranking previously studied alternatives for facility location, the Taguchi method introduces a fresh perspective to site selection. With the Taguchi method, we expedite the site selection process by drawing upon data from existing facility locations, eliminating the need to extensively explore alternative sites. We also predict the anticipated profitability upon launching the selected alternative. This approach underscores the uniqueness and advantages of the Taguchi method when compared to traditional methodologies in the field. Our study begins by evaluating the performance of existing branches and subsequently employs the Taguchi method to streamline the selection of a new branch location, culminating in a comparison of the results obtained from the three methodologies. TOPSIS and VIKOR yield similar rankings for existing branches. In contrast, the Taguchi method, using these branches, determines criteria for the new branch based on an L9 orthogonal array. A variance analysis reveals competitors (40.2%) and population size (35.8%) as the most significant factors. This study highlights the distinct perspectives these methods offer in facility location selection and their role in decision-making.