Optimizing Retail Location Selection Strategies Using Spatial Statistics and GIS: A Decision Support Framework for Supermarket Chains in Emerging Urban Centers

Abstract

Determining the most suitable location for a retail store is of great strategic importance, as it is known that choosing the wrong location can lead even the best retailer to failure. Spatial statistical methods play a critical role in analyzing urban settlements, revealing spatial relationships, and strengthening strategic decision-making processes. Objective outputs from spatial analyses enable newly established supermarket chains or small businesses to take more accurate strategic steps, particularly in terms of planning commercial activities and determining regional investment potential. This research proposes a GIS-based decision support framework that integrates multi-source demographic data with advanced spatial econometric approaches and is applicable at different urban scales. In this context, the study aims to determine the most suitable locations for new supermarket branches by integrating demographic data with GIS-based spatial statistical techniques, using the example of Sivas city center (T & uuml;rkiye), which has characteristics consistent with the concept of emerging urban centers. This methodology incorporates Global Moran's I, Anselin Local Moran's I, and Kernel Density Estimation (KDE) spatial statistical methods to measure the overall spatial pattern of market density, identify both clustering areas and statistically significant spatial outliers, and estimate the spatial density of geographically located point data, transforming this density into a continuous surface. Analyses using Global Moran's I identified clustering at market points in the city. Subsequently, Anselin Local Moran's I and Kernel Density analyses determined that the clusters were concentrated in central and large neighborhoods. Finally, a Population Density Map was generated for each neighborhood using population data, identifying areas with high population but few markets. All data obtained were overlaid using Overlay analysis to identify potential areas. The findings provide a robust decision support framework for marketing decision-makers, offering an objective guide for identifying high-potential investment areas while minimizing spatial risks. By bridging the gap between spatial econometrics and urban retail planning, this research contributes to the more systematic and sustainable management of commercial land use in developing urban contexts.