A Reverse Logistics Network Model for Handling E-commerce Returns

Abstract

E-commerce supply chains are becoming more complex due to increasing global sales, and product returns from these sales are alarmingly high, highlighting the importance of effective return management. This paper proposes a reverse logistics network model to optimize return management. The proposed model applies ward-like hierarchical clustering with geographical constraints to detect return tendencies and utilizes mixed integer linear programming to optimize the network. The decision variables of the model include selection of Initial Collection Centers (ICCs), allocation of customer markets to ICCs, and optimal return volumes to be sent to each fulfillment center and recycling center from ICCs. The validity of the proposed model is established through a case study conducted in the consumer electrical and electronics sector of an e-commerce firm, providing 39.9% cost savings on average compared to the current Reverse Logistics (RL) network operation. This study contributes to the literature by integrating industry 4.0 technologies into the assessment of RL and facility planning with network optimization. The proposed RL network model serves as an operational planning tool, providing directions to e-commerce firms on optimizing RL networks and utilizing partner networks with integrated decision making for product returns.