Order consolidation for the last-mile split delivery problem with green labels

Green labels can be utilized to steer customers towards selecting eco-friendly delivery options, such as extending the due time of shipments, which provides excellent opportunities for order consolidation to address the last-mile split delivery problem. Order consolidation involves postponing certain shipments at delivery stations and combining them with other incoming shipments, resulting in fewer trips to deliver the consolidated shipments to customers. To address the tradeoff of whether to offer green label choices to customers, we propose a two-stage stochastic programming model. This model integrates the first-stage decisions of presenting green label choices and the second-stage decisions of generating order consolidation schemes, taking into account uncertain green label choices (labels A, B, and C). The complexity of the deterministic equivalent formulation of the model increases significantly with the number of scenarios, which motivates us to develop a scenario-based order consolidation algorithm. By decomposing the problem based on scenarios, the proposed algorithm first generates a pool of order consolidation schemes for each customer with different types of labels. Then, for each scenario, it combines the solutions for different customers while considering inventory capacity. Numerical experiments have been conducted to demonstrate that the average advantage of the order consolidation approach with green labels is 2.35% compared to the approach without green labels. Additionally, managerial insights for the practical application of green labels are provided.