Integrated scheduling of truck and drone fleets for cargo transportation in post-disaster relief: A two-stage stochastic optimization approach

Abstract

• From the perspective of the transportation method, we propose an enhanced version of the traditional TSP-D method. • From the modeling perspective, our model incorporates a drone allocation scheme to address this issue. • We developed an effective two-stage heuristic algorithm based on parallel computing and enhanced this algorithm. • We conducted numerous experiments using Fangshan District, Beijing, China, as the case study subject. This paper develops a two-stage stochastic optimization approach to handle the integrated scheduling problem of truck and drone fleets for cargo transportation in post-disaster relief. Initially, a two-stage stochastic optimization model is introduced to account for the uncertainty of the traveling time of trucks. The first stage involves an integer nonlinear optimization model to determine the drone allocation scheme and truck scheduling scheme, and the second stage employs a mixed-integer linear optimization model to establish the drone scheduling scheme. Subsequently, an efficient heuristic algorithm based on parallel computation is developed to solve the problem. Finally, some experimental tests were conducted using real disaster data from extreme rainstorms in Fangshan District, Beijing in July 2023. The extensive experiments demonstrate that the proposed algorithm consistently identifies high-quality solutions efficiently compared to the exact algorithm. The numerical results suggest that considering the drone allocation scheme can reduce relief cargo transportation time and enhance transportation efficiency.