| dc.description.abstract | As consumer habits evolve, so do the methods employed to deliver goods and services. The growth of online sales is driving up transport demand on an annual basis. The traditional B2B delivery model has shifted towards B2C or C2C delivery. This has created challenges for traditional transportation in meeting future demands. Consequently, to enhance the efficiency of delivery and reduce costs, the logistics industry of the future will increasingly rely on a diversified transportation system, with drones becoming one of the principal delivery methods. In light of the mounting demands placed upon the logistics industry, drone logistics operators must expand their drone fleets and enhance their service offerings. The utilization of flexible collective drones for deliveries can not only improve the quality of service and facilitate scheduling flexibility, but also circumvent the issue of single-service dependency. Several domestic operators are currently engaged in experimental operations in drone logistics. If collective drone planning is conducted in advance, it can address the scheduling challenges posed by increasing demand and improve overall service levels.
This study constructs an integrated drone cargo delivery model from the perspective of drone operators, focusing on customer reservations and considering both reservation demands and relevant drone delivery constraints. The research employs mathematical programming and space-time network flow techniques to minimize operational costs. The model is defined as an integer network flow problem with additional constraints, classified as NP-hard. However, when dealing with large-scale practical problems, it often becomes challenging to solve them within a limited time due to the scale. Consequently, this study combines Lagrangian relaxation with CPLEX and develops a heuristic algorithm to address the problem effectively. To evaluate the model′s practicality, example tests and sensitivity analysis of key parameters were conducted. The results indicate that the model and its heuristic solution perform well, significantly reducing operational costs while meeting reservation demands, and providing valuable insights for future drone scheduling planning for operators. | en_US |