Abstract
Effective support of emergency materials is a prerequisite for post-disaster rescue. This paper addresses the optimization of transportation from supply points to distribution centers, accounting for road resistance, weather-driven speed attenuation, and disaster intensity. By minimizing the sum of transportation, construction, and time penalty costs, the proposed mixed-integer linear programming (MILP) model provides a scientific basis for resilient logistics planning.
The Challenge of Post-Disaster Logistics
In the immediate aftermath of a natural disaster, the logistics of delivering food, water, and medical supplies is a race against time. This process typically occurs in two stages: moving materials from national supply points (SPs) to regional distribution centers (DCs), and then from those DCs to specific rescue points (RPs) within the disaster zone.
Most existing models focus on minimizing distance or standard costs. However, disasters create chaotic environments where weather conditions and road damage (disaster intensity) significantly degrade vehicle speeds. Ignoring these variables leads to "optimal" plans that fail in reality due to underestimated travel times and the high social cost of delivery delays.
Methodology: Integrating Environmental Factors
The study constructs a multi-objective optimization model formulated as a mixed-integer linear programming (MILP) problem. The objective function minimizes the Total Cost (TC), which comprises:
- Transportation costs: Based on actual travel time and distance.
- Fixed investment costs: The expense of establishing distribution centers.
- Variable processing costs: Handling expenses at the DCs.
- Time penalty costs: A penalty for every hour a delivery exceeds the expected arrival window.
The model's innovation lies in its calculation of transportation speed. It uses an attenuation factor ($\theta$) and disaster intensity ($r_{ki}$) to adjust maximum speeds. By linking these environmental factors directly to unit costs and time penalties, the algorithm avoids high-risk or weather-impacted routes, even if they are geographically shorter.
Key Results and Sensitivity Analysis
The model was validated using a numerical simulation involving 4 supply points, 4 potential distribution centers, and 8 rescue points. The results highlight that weather is a critical mathematical weight rather than just an inconvenience.
As shown in the experimental data, there is a near-linear relationship between the weather attenuation coefficient and the total distribution cost. As weather conditions worsen (higher $\theta$), costs rise predictably. Sensitivity analysis revealed that while traffic flow ratios have minor impacts on DC selection, weather and road intensity are decisive. For instance, increasing the attenuation factor from 40 to 80 caused the model to entirely reconfigure the network, dropping certain distribution centers in favor of more "weather-resilient" hubs.
Conclusion and Future Directions
The proposed model successfully prioritizes routes with lower disaster intensity, aligning mathematical optimization with the intuitive logic of avoiding damaged infrastructure. However, the current model assumes static meteorological conditions. Future research should explore dynamic, hour-by-hour weather updates and multi-modal transport options (e.g., air or rail) for scenarios where road intensity leads to total blockage. Testing against historical data from real-world events like major hurricanes would further validate its predictive power for emergency response.
Summary updated