Analysis of Drone-based Last-mile Delivery Systems under Aerial Congestion: A Continuum Approximation Approach

<p>This dataset presents a systematic analysis and design framework for a spectrum of last-mile delivery systems that leverage unmanned aerial vehicles (UAVs). Four distinct modes are considered: (1) direct drone deliveries from a fixed depot; (2) drone deliveries from parked trucks that carry bulk parcels to customer neighborhoods; (3) drone deliveries from nonstopping trucks that tour customer neighborhoods; and (4) as a benchmark, traditional truck-based home deliveries. We present a new continuum-approximation approach that is used for analysis of both truck routing and aerial-UAV traffic. We compared the operational cost and efficiency of different delivery schemes to reveal how a certain scheme is the most efficient under various scenarios. We demonstrate the applicability of our model on expansive real-world roadway networks and further conduct analysis on grid networks, yielding key analytical insights. The drone-based delivery is demonstrated to be superior to conventional truck-only delivery, suggesting a significant potential for socioeconomic benefit. It is observed that when servicing a relatively low demand over a small area, dispatching drones directly from the distribution facility is the most efficient method. As the demand grows or spans a wider area, collaborative strategies are preferred, as they better cope with certain aspects, such as the long line-haul cost or the ramping aerial congestion, by striking a balance between efficiency and flexibility.</p>