Optimization of Drone Delivery Routes for E-commerce in Urban Areas

Drone delivery is expected to improve operations by enhancing flexibility and reducing congestion effects induced by lastmile deliveries.With rising digitalization and urbanization, however, flight routing of drones is constantly grappling with the challenge of uncertain demand.In this study, an opportunity-constrained model that considered both the endurance and payload capacity of drones and stochastic customer demand was developed to solve the drone delivery planning problem for E-commerce in urban areas.Considering the challenging solvability of the model, historical customer demand data were leveraged to linearly transform opportunity constraints with unknown true distribution.In addition, a constructive heuristic algorithm was used to solve the established opportunity-constrained model, and an effective strategy was developed to generate the quality of initial flight paths.The experimental results based on various scales of classic cases demonstrate that an effective drone delivery route scheme can be generated by the proposed heuristic algorithm compared to traditional savings algorithm, allowing drones to complete tasks in shorter flight distances at the same payload rate, thus reducing overall costs.Compared to deterministic models, more robust flight path schemes for addressing continuously changing customer demands can be obtained.The model and insights in this study could be used as a reference for the application of drone delivery in urban areas to promote logistics efficiency.