考量無人機補貨及調貨之卡車配送最佳化

Abstract

隨著科技與網路通訊蓬勃發展，無人機市場漸趨擴增，應用層面亦愈加廣泛。於物流配送之應用方面，為因應全球物流需求逐時成長，物流業者期望能提升配送效率並增加配送服務量，於是利用無人機進行配送之物流方式興起。無人機運作受到其飛行距離及電池續航力以及載貨能力等限制，因此現階段無人機物流之發展模式，以卡車搭配無人機進行混合配送之研究為主。此模式經文獻證實得以提升整體配送效益，然而現實環境中欲將無人機配送落實於應用仍是困難挑戰。有鑑於種種無人機配送服務之現實層面疑慮，本研究欲提出一考量無人機補貨與調貨之卡車配送模式，以提升實際應用之可行性。本模式以無人機協助卡車補貨與調貨得以縮短貨物補給時間，並且減少卡車來回倉庫補貨及前往他處調貨之成本消耗，亦降低增派卡車之額外成本，進而提升整體配送效益。本模式具體運作方式為卡車搭載無人機從倉庫出發配送，途中無人機可於卡車停等位置起飛，前往倉庫進行補貨或他車進行調貨工作後返回卡車位置降落，以達增加卡車載貨量之目的。其中，無人機調貨之作用可使模式更具彈性，未來可將模式延伸為動態問題，亦可調整為不同情境設計，例如行動衛星倉儲、大小卡車之間調貨等。本研究透過建立混合整數線性規劃模型，以最小化配送完成時間為目標求解問題，並發展以數學模型為基礎之啟發式演算法提升求解效率。本研究藉由不同規模之案例進行測試，針對不同顧客數與卡車數之案例，驗證演算法之求解品質並展現其求解效率。研究結果顯示，利用無人機補貨與調貨運作得以增加卡車服務配送量，且使卡車之閒置貨物獲得流通調度之管道。最後透過敏感度分析，針對卡車初始載貨量分配及無人機最大飛行時間限制，給予策略與建議。

With the rapid development of technology and internet, the unmanned aerial vehicle (UAV) market has been expanding, and its’ applications have been widely expended to various domain. In the field of logistics and delivery, logistics companies aim to enhance delivery efficiency and increase service capacity to meet the growing global logistics demands. As a result, the utilization of delivery with drones has emerged as a logistics method. However, the operational range, battery life, and payload capacity of drones impose certain limitations on their operations. The current workaround for this issue is using hybrid truck-drone delivery. Previous studies have confirmed the effectiveness of this model in improving overall delivery efficiency. However, implementing this approach in practical environments poses significant challenges. Considering the practical concerns, this study proposes a truck delivery mode considering goods resupplied and transferred by drones. This research aims to enhance the feasibility of practical applications. The proposed model introduces drones to assist trucks in resupplying and transferring goods, thereby reducing the time required for replenishment and minimizing the costs associated with truck trips to warehouses or other trucks. Additionally, this model helps mitigate the extra costs of deploying additional trucks, ultimately enhancing overall delivery efficiency. The operational process involves trucks carrying drones for delivery, and the drones can take off from designated locations while the truck is stationary. The drones can travel to warehouses for resupply or to other vehicles for transfer before returning and landing at the truck's location, thus increasing the truck's capacity. The inclusion of drone transfer provides flexibility to the model, enabling exploration of dynamic and complex scenarios in future research, such as mobile satellite depot or transfer between trucks of varying sizes. To address the problem, a mixed integer linear programming model is constructed in this study, with the objective of minimizing the total delivery time. Furthermore, a mathematical model-based heuristic algorithm is developed to improve the computational efficiency. By testing with diverse scenarios involving varying numbers of customers and trucks, the algorithm’s solution quality is validated, and its’ computational effectiveness is demonstrated. The results indicate that using drones for resupply and transfer increases the delivery capacity of trucks and provides a mechanism for redistributing idle goods. Finally, sensitivity analysis is conducted to provide strategic recommendations regarding the initial loading distribution of trucks and the maximum flight time constraint for drones.