Strategically locating impulse items in a store

This paper studies the problem of optimizing the placement of items in a retail store setting by combining two approaches, one from Facilities Location theory and the second based on probabilistic modeling of customer travel within the store. Specifically, it focuses on designing a store layout where item placement is done with the objective of maximizing the total profit earned from the sale of impulse items and where customers travel is probabilistic based on distance between successive items purchased. It begins by presenting a simple version with a grid layout and rectilinear distance metric that is solved to optimality. Then, a two stage heuristic algorithm is developed for the general problem that employs a simulation analysis to verify the quality of the solutions generated. The performance of this two-phase algorithm is empirically tested using three different approaches: benchmarking against available results, empirical testing, and with real-world data taken from a grocery store in the western region of New York. Results attest to the effectiveness of the solutions generated by algorithm and its ability to solve larger problems than have been reported heretofore in the literature.