Truck and Drone Collaborative Delivery Model Based on Integrated Battery-Swapping Capability

To address the limited flight endurance and inefficient recharging of drones in existing truck and drone collaborative delivery modes in rural areas, this paper proposes a new collaborative delivery model based on integrated battery-swapping capability, which is specifically designed to overcome these operational limitations. The new collaborative delivery model based on integrated battery-swapping functionality integrates the battery-swapping function of drones into existing rural infrastructure that is already available in rural areas. Under this model, there is no need to additionally construct other facilities specifically for drone battery swapping. Through the use of existing rural resources and without any additional facility construction, the proposed model enables drone charging to become more flexible during the delivery process. At the same time, this model avoids the additional costs brought by the construction of charging stations that would otherwise be required. Under the proposed collaborative delivery model, drones replace batteries in villages that are equipped with battery-swapping capability. After battery replacement is completed in villages with battery-swapping capability, drones carry out delivery tasks along pre-planned routes. Meanwhile, trucks transport parcels and batteries from the distribution center to villages with battery-swapping capability. At the same time, trucks collect depleted batteries during the delivery process while completing the transportation of parcels and batteries as part of their assigned tasks. Considering constraints related to both drones and trucks, including payload capacity constraints and endurance constraints, a mixed-integer programming model is established. The mixed-integer programming model is formulated with the objective of minimizing the total cost of the delivery system under the given constraints. To solve the mixed-integer programming model efficiently, a two-stage hybrid algorithm is adopted.In the first stage of the two-stage hybrid algorithm, an improved K-means algorithm is used. During the clustering process, the improved K-means algorithm simultaneously considers the distance from each demand point to the corresponding cluster center and the distance from the distribution center to each cluster center. By considering both the distance between demand points and cluster centers and the distance between the distribution center and cluster centers in the clustering process, the improved K-means algorithm overcomes the defect of the traditional K-means algorithm, which ignores the degree of closeness between the distribution center and the cluster centers. Through this improvement, global optimality is ensured. In the second stage of the two-stage hybrid algorithm, a genetic algorithm with adaptive control of crossover segment length is introduced. By retaining longer crossover segments during the crossover operation, the genetic algorithm maintains the superior structure of parent solutions. Through retaining longer crossover segments, the transmission of high-quality genes is accelerated during the evolutionary process of the genetic algorithm.The results of case study analysis show that, compared with existing truck and drone collaborative delivery modes and truck-only delivery modes, the truck and drone collaborative delivery model based on integrated battery-swapping functionality achieves improved performance. In terms of delivery cost, delivery costs are reduced by 22.55% and 39.07%, respectively. In terms of delivery time, delivery time is reduced by 32.7% and 68.43%, respectively. In addition, the two-stage hybrid algorithm exhibits stronger optimization capability and faster solution speed.