Inventory Optimization for Fulfillment Integration in Omnichannel Retailing

Abstract

With e-commerce growing at a rapid pace compared to traditional retail, many brick-and-mortar firms are supporting their online growth through an integrated omnichannel approach. Such integration can lead to reduction in cost that can be achieved through efficient inventory management. A retailer with a network of physical stores and fulfillment centers facing two demands (online and in-store) has to make important, interlinked decisions - how much inventory to keep at each location and where to fulfill each online order from, as online demand can be fulfilled from any location. We consider order-up-to policies for a general multi-period model with multiple locations and zero lead time, and online orders fulfilled multiple times in each period. We first focus on the case where fulfillment decisions are made at the end of each period, which allows separate focus on the inventory decision. We develop a simple, scalable heuristic for the multi-location problem based on analysis from the two-store case, and prove its asymptotic near-optimality for large number of omnichannel stores under certain conditions. We extend this to the case where fulfillment is done multiple times within a period and combine it with a simple, threshold-based fulfillment policy which reserves inventory at stores for future in-store demand. With the help of a realistic numerical study based on a fictitious retail network embedded in mainland USA, we show that the combined heuristic outperforms a myopic, decentralized planning strategy under a variety of problem parameters, especially when there is an adequate mix of online and in-store demands. Extensions to positive lead times are discussed.