Analysis of Energy Efficiency in Truck-Drone “Last Mile” Delivery Systems
Truck-drone delivery systems have the potential to improve how the logistics industry approaches the “last mile problem”. For the purposes of this study, the “last mile” refers to the portion of the journey between the last transportation hub and the individual customer that will consume the product. Drones can deliver packages directly, without the need for an underlying transportation network but are limited by their range and payload capacity. Studies have developed multiple truck-drone configurations, each with different approaches to leverage the benefits and mitigate the limitations of drones. Existing research has also established the drone’s reduction to package delivery time over the traditional truck only model. Two key model factors that have not been considered in previous research are the distribution of package demand, and the distribution of package weight. This study analyzes the drone’s impact to the energy efficiency of a package delivery system, which has taken a backseat to minimizing delivery time. Demand distribution dictates the travel distances required for package delivery, as well as the proportion of delivery locations that are in range for drone delivery. Package weight determines the energy consumption of a delivery and further restricts the proportion of drone eligible packages. The major contributions of this study are the development of a truck-drone tandem mathematical model which minimizes energy consumption, the construction of a population-based package demand distribution, a realistic package weight distribution, and a genetic algorithm used to solve the mathematical model developed for problems that are too computationally expensive to be solved optimally using an exact method. Results show that drones can only have a significant impact to energy efficiency in package delivery systems if implemented under the right conditions. Using truck-drone tandem systems in areas with lower package demand density affords the drone the potential for larger energy savings as larger portions of the truck distance can be replaced. Further, the lower density translates to greater differences between the road-restricted driving distance and the flying distance between delivery points. Finally, energy savings are highly dependent on the underlying package weight distribution of the system. A heavier average package weight increases the energy consumption of the system, but more importantly the portion of packages above the drone’s payload capacity severely limit the savings afforded by the incorporation of drones.