Energy Efficient UAV-Based Last-Mile Delivery: A Tactical-Operational Model With Shared Depots and Non-Linear Energy Consumption

In this paper, we have investigated a drone delivery problem to address the tactical decisions arising in last-mile applications where the connection with operational plans is taken into account. The problem deals with the tactical selection of a subset of FCs to launch and retrieve the drones, and the fleet sizing decisions on the optimal number of drones to be employed. We have incorporated the non-linear and load-dependent energy consumption function into the definition of a load-indexed layered network, leading to the definition of a MILP that can be efficiently solved for instances with 50 and 75 customers. There are several fruitful directions for future research. The use of shared depots implies for the drones the freedom to choose different FCs for departure and arrival. Anyway, a drawback may exist in the considered scenario, since we should have enough drones in each FC for the next period. The extension of the present model to the multi-period location routing case, where the location decisions are taken once and the routing plans are addressed within each period, is an interesting issue for future research. Moreover, the design of heuristic and self-adaptive approaches to alleviate the computational burden for larger instances deserves further attention, as well as the extension of the present model to en-route drone charging.