A shortage risk mitigation model for multi-agency coordination in logistics planning

We introduce a two-stage stochastic program to handle typical disaster preparedness activities under uncertainty from a multi-agency perspective. The model explicitly takes into account the number of people without healthcare attention, relief aid, and shelter support. We build a function that represents the total number of people at risk of not receiving proper humanitarian assistance using a bi-objective approach in which expected logistics costs are also minimized. The benefit of our approach is assessed through real flood cases in Mexico in which GIS analysis was used to enhance data gathering and to provide risk maps that could be potentially used by policy-makers in practical settings. The overall results suggest that sheltering decisions have to be closely coordinated with the management of material and human resources to avoid an increased number of people deprived of attention and relief aid. The Pareto Frontier also reveals that some solutions exhibit a quite interesting trade-off, e.g., it is possible to improve the overall relief assistance by almost 17% at the expense of less than 14% in the logistics costs.