Evacuation management is an essential service for large-scale (e.g., hurricane) evacuations because of the difficulties inherent in moving a large number of people long distances using infrastructures not designed for the purpose. One common problem is massive congestion, which causes longer, costlier, and more stressful evacuations, lower compliance rates, and increased risk to the population. This research will study demand-based strategies to manage large-scale evacuations, that is, strategies that organize the evacuees so as to improve the efficiency of the evacuation. Specifically, aggregate-level staging and routing and network flow controls will be studied. Current evacuation management is mainly supply-based (e.g., increasing the evacuation capacity using contra-flow). However, a purely supply-based management approach often does not sufficiently reduce congestion (and thus risk). Because of this, several jurisdictions have started augmenting their evacuation plans using demand-based strategies. Nevertheless, how best to use these strategies is unknown; unfortunately, research suggests that poorly developed strategies can decrease evacuation efficiency. The research methodology of this award will be to study analytical models to understand the structural properties of optimal evacuation strategies and to gain insights into this difficult problem. These structural properties and insights will then be used to develop effective and efficient demand-based evacuation strategies. These strategies will be tested under more complex (and realistic) settings using simulation.
