Strategic emergency preparedness network design integrating supply and demand sides in a multi-objective approach

We consider integration of fast evacuation and cost-effective relief distribution objectives, the two critical aspects of emergency management, to design a strategic emergency preparedness network for foreseen disasters, such as hurricanes. To this end, we introduce the design of a three-tier system, involving evacuation source, shelters, and distribution centers, that integrates the relief (supply) and evacuation (demand) sides of an emergency preparedness network. This is motivated by the realization that the shelters are shared facilities at the interface of the supply and demand sides. Although primarily intended for strategic decision making, our model can also make tactical decisions, thus spanning two separate time frames before a disaster’s occurrence. To solve models for large-scale instances, we adopt a Benders Decomposition approach with an implementation that solves only one instance of the master problem. We also determine that, in this framework, tuning of master tree search parameters along with the strengthening of Benders cuts significantly impact convergence. We conduct an extensive computational study to examine the impact of algorithmic improvements and further consider a realistic case study based on geographic information system (GIS) data from coastal Texas and examine the effects of changing problem parameters. By comparing our approach with current practice, we illustrate that a pro-active strategic integration of evacuation and distribution can relieve the resource-constrained large urban areas, traditionally considered as shelter locations.