A Path-Based Model for Efficient Logistics Distribution in Damaged Transportation Networks

This study presents a model for post-disaster goods distribution from warehouses to depots over partially damaged transportation networks. Unlike classical transportation models that assume static and fully accessible infrastructure, the proposed model accounts for intermediate intersections, vehicle capacity constraints, limited road accessibility, and time-dependent restoration schedules. To improve computational tractability, the model employs a preprocessing step that enumerates all feasible delivery paths before optimization, thereby significantly reducing the solution space. The core of the model integrates flow allocation with shortest path decisions to minimize the latest arrival time across all demand points. It dynamically adapts to network conditions by selecting alternative paths when restoration delays are prohibitive or leveraging restored links when they offer faster delivery. Numerical examples demonstrate the model's effectiveness in generating time-efficient delivery plans under varying levels of infrastructure disruption. The results highlight the framework's practical potential for supporting emergency logistics planning and resource deployment in disaster-affected regions, particularly in contexts of government preparedness planning and decision-support systems. Furthermore, the model provides a basic for future extensions involving uncertainty, dynamic demand profiles, multi-period planning, and multimodal transportation strategies.