Making Better Fulfillment Decisions on the Fly in an Online Retail Environment

Abstract

Relative to brick-and-mortar retailers, online retailers have the potential to offer more options to their customers, with respect to both inventory as well as delivery times. To do this entails the management of a distribution network with more decision options than a traditional retailer. The online retailer, not the customer, decides from where items will ship, by what shipping method, and how or whether multipleitem orders will be broken up into multiple shipments. What is the best way to fulfill each customer's order to minimize average outbound shipping cost? We partner with an online retailer to examine this question. We develop a heuristic that makes fulfillment decisions by minimizing the immediate outbound shipping cost plus an estimate of future expected outbound shipping costs. These estimates are derived from the dual values of a transportation linear program (LP). In our experiments on industry data, we capture 36% of the opportunity gap assuming clairvoyance, leading to reductions in outbound shipping costs on the order of 1%. These cost savings are achieved without any deterioration in customer service levels or any increase in holding costs. The transportation LP also serves as the basis for a metric that provides information on the quality of the inventory position. Based on initial successful piloting, our industrial partner has implemented the metric, as well as a version of the heuristic that it is applying to every fulfillment decision for each of its SKUs in North America.

Mitigating Spillover in Online Retailing via Replenishment

“…However, the parameters for this simulation were also utilized in a previous analysis that was performed for this same industrial partner in which the authors proposed operational changes. The improvement that the industrial partner experienced as a result of that analysis was about the same as was predicted (Acimovic and Graves 2015). To some extent, this validates the aspects of the simulation such as shipping costs and feasibility of shipping modes for different customers.…”

“…The first stream of literature relates to inventory policies in online retailing. Given a current distribution of inventory among FCs, several papers examine the problem of deciding how to fulfill a customer's order from this set of FCs (Acimovic and Graves 2015;Jasin and Sinha 2015;Lei et al 2016;Mahar et al 2009;Xu et al 2009). This current paper builds on that stream of work to examine the supply chain one step earlier: how to develop better forward-looking allocation policies that replenish inventory into the FCs.…”

“…The four delivery speeds are next day, second day, four day, and eight day. We group together geographical regions using k-means clustering, according to the methods described in Acimovic and Graves (2015). …”

“…In specifying the replenishment policy, we assumed a myopic fulfillment policy; one might determine how to recast the replenishment policy, allowing for one of the more sophisticated fulfillment policies mentioned in the literature review. Indeed, there is evidence that a smarter fulfillment policy can lead to significant cost reductions in comparison to a myopic rule (Acimovic and Graves 2015;Tiemessen et al 2013). Additionally, one would like to relax the assumption of a deterministic and common lead time to allow for uncertain and heterogeneous lead times.…”

Real-Time Optimization of Personalized Assortments

“…Both the AdWords and personalized assortment optimization problems contain the b-matching problem as a special case (Kalyanasundaram and Pruhs 2000). Mehta et al (2007) proposed an algorithm that achieves an optimal-competitive ratio for the AdWords problem by taking into account both the bid and the budget of the advertisers; see for a survey on online algorithms and Acimovic and Graves (2012) for another application in the context of inventory management.…”