Total Hockey Optimizes Omnichannel Facility Locations

Omnichannel distribution, which blends brick-and-mortar retailing and e-commerce, is a key challenge for today’s supply chains. In this paper, we report on a study to design an omnichannel distribution system for Total Hockey, a growing U.S. sporting goods retailer in a competitive environment. Management strongly believes that e-commerce success will depend on high service levels characterized by one- or twoday delivery and initially thought that a new omnichannel warehouse located on the East Coast could support its expansion plans. To study the situation, we developed a profit-maximizing optimization model for locating omnichannel warehouses that supports both e-commerce and store shipments. The model uses estimates of e-commerce demand by metropolitan statistical area (MSA) across the United States, while incorporating management’s sales expectations regarding the value of high service levels, e-commerce sales lost to competitors’ stores, and reverse cannibalism from Total Hockey’s own retail stores. Multiple warehouse sizes allow modeling of nonlinear inventory costs. The facility-location optimization model allows exploration of multiple solutions and an assessment of the impact of higher service levels. The results of the study were contrary to management expectations and suggested a significant redesign of the distribution system. We report results for several analyses, implementation details, and managerial insights for omnichannel distribution.