Pickup and Delivery Problem with Synchronous and Asynchronous Transshipment for Q-Commerce Delivery

This study explores ways to improve delivery efficiency and reduce costs in the rapidly growing quick commerce (Q-commerce) market, which is expanding because of advancements in Internet of Things technology and the rise of contactless consumption. However, it faces significant challenges, such as high transportation costs and inefficient vehicle utilization. To address these challenges, a novel delivery system is proposed that simultaneously incorporates synchronous and asynchronous transshipment. The proposed system, the Pickup and Delivery Problem with two types of transshipment, is formulated as a mathematical optimization model. Since the problem is Nondeterministic Polynomial-time hard (NP-hard), implying that finding an optimal solution is computationally intensive as the problem scale increases, a two-phase heuristic algorithm is developed that combines adaptive large neighborhood search metaheuristic with transshipment scheduling. In Phase 1, adaptive large neighborhood search is employed to improve the initial Pickup and Delivery Problem solution. In Phase 2, the two transshipments are integrated into the improved solution. Experimental results from various scenarios show that the proposed two-phase algorithm effectively reduces Q-commerce delivery costs by approximately 10.97% compared with initial solutions. Of note, simultaneously considering synchronous and asynchronous transshipment resulted in an additional 0.5 percentage points (pp) improvement in the objective function value compared with using a single transshipment. These findings suggest that transshipment solutions can effectively reduce vehicle operating times and request travel distances, contributing to the future popularization of multi-echelon delivery systems.