Hybrid truck-drone systems for last-mile delivery logistics
Abstract
With growing consumer demand and expectations, companies are attempting to achieve cost-efficient and faster delivery operations. The integration of autonomous vehicles, such as aerial and ground drones, in the last-mile network design, could curtail many operational challenges and provide a competitive advantage. This dissertation deals with the problem of delivering orders to a set of customer locations using multiple drones that operate in conjunction with a single truck. Four variants of the problem are considered. The first variant takes advantage of the drone eet by parallelizing the delivery tasks via concurrently dispatching the drones from a truck parked at a focal point (ideal drone launch location) to the nearby customer locations. The key decisions to be optimized are the partitioning of delivery locations into small clusters, identifying a focal point per cluster, and routing the truck through all focal points such that the customer orders in each cluster are fulfilled either by a drone or truck. The first problem variant restricts each focal point to one of the customer locations, while the second addresses the same delivery problem when allowing truck viii stops to be anywhere in the delivery area (i.e., a customer or non-customer location). Mathematical programming models are developed to jointly optimize both the clustering and routing decisions. The third variant suggests allowing the usage of non-customer locations (referred to as exible sites) as truck stops for drone launch and recovery operations (LARO). This relaxes a common constraint in the literature restricting the drone LARO to customer locations. The proposed variant also accounts for three key decisions - (i) assignment of each customer location to a vehicle, (ii) routing of truck and UAVs, and (iii) scheduling drone LARO and truck operator activities at each stop, which are always not simultaneously considered in the literature. A mixed integer linear programming (MILP) model is formulated to jointly optimize the three decisions. Furthermore, to handle large problem instances, we develop an optimization-enabled two-phase search algorithm by hybridizing simulated annealing and variable neighborhood search. The fourth variant in this dissertation proposes utilizing a network of docking stations and repositioning of drones to enhance the efficacy of delivery operations. In particular, such stations are used for drone docking before and after delivery operations to avoid both loading all required drones to the truck at the depot and waiting of the truck on its route to recover drones. A MILP model is formulated to optimize the management of this setting of facilities and delivery operations. Finally, extensive computational experiments are conducted in this dissertation for the four variants to obtain several insights aiding the logistics practitioners in decision making.
Degree
Ph. D.