Store Layout Optimization with Endogenous Consumer Search

Abstract

Retailers employ various store layouts to drive traffic, conversions, and ultimately sales. At one extreme is the forced-path layout, which guides shoppers through a defined route, maximizing exposure to products and encouraging unplanned purchases but at the expense of convenience. At the other end, open layouts allow easy movement between aisles, enhancing convenience but reducing category exposure. In the middle, plenty of store architectures are possible. We develop a theoretical framework linking layout and category-aisle assignments with consumer search and purchasing. We model consumers who choose to enter aisles by inspecting the cost of doing so against the expected value they may find in it, and subsequently they may purchase the categories included in the aisle when their utility is high enough. The retailer’s layout and assignment optimization results in a multiple knapsack integer program, with the added challenge that the weights can be of any sign and that the number of knapsacks to open is the retailer’s decision. It turns out that this formulation is solvable under certain conditions (either exactly or with an FPTAS). In other cases, the problem is hard even to approximate, so there is no hope of obtaining solutions with guaranteed near-optimal performance. Key insights from the model suggest that race-track layouts are optimal when search costs have economies of scale, while open-space configurations work best with diseconomies of scale, where smaller aisles are preferred. The optimal aisle composition strategically pairs high-traffic anchor categories with impulse products, while exploratory categories are best displayed separately. Finally, we demonstrate the model’s application through a case study, discussing estimation challenges and practical implications.

Authors: Laura Wagner, Victor Martinez-de-Albeniz.