Dynamic vehicle-cargo matching strategy considering cargo-pooling

ObjectiveTo improve the efficiency of resource allocation on network freight platforms and reduce transportation costs, a vehicle-cargo matching strategy is proposed for intra-city short haul. This strategy is based on a comprehensive consideration of dynamic vehicle-cargo demands and the matching success rate.MethodTo achieve effective cargo combination and meet the real-time requirement of first-come-first-matched cargo, the cargo-pooling was adopted to design a cargo combination scheme. A bilateral matching evaluation index for vehicle-cargo combinations was then established. The matching dimensions included vehicle type, loading, time, routing, and transportation cost. The bilateral comprehensive matching degree was obtained through weighting. On this basis, a greedy-based dynamic vehicle-cargo matching algorithm was designed; and a dynamic vehicle-cargo matching model with sliding time window was established. Finally, a freight scenario analysis was carried out in Beijing to evaluate the model's effectiveness in optimizing resource allocation. The applicability of matching results with different time window widths and vehicle-to-cargo resource ratios was also analyzed.ResultCompared with traditional non-combination matching models, the proposed strategy improves the loading factor by approximately 20% in various transportation scenarios. The matching success rate is improved by approximately 10%. It saves nearly half of vehicle resources. Meanwhile, the significant advantages are achieved in freight profit for both vehicle providers and cargo owners. The proposed strategy is versatile and adaptable to various situations. Both loading factors and resource utilization rate hold steady with varying vehicle-to-cargo resource ratios. The matching success rate consistently exceeds 95% with flexible time window adjustment, demonstrating good dynamic demand response.ConclusionTo achieve system optimization in a practical application, the proper time window can be identified based on various vehicle-cargo resource ratios.