An integrated user-system approach for shelter location and evacuation routing

Abstract

Disasters are catastrophic events that can severely affect the life conditions of entire communities. Disaster-related issues are usually dealt with according to the Disaster Operations Management framework, which is composed of four phases: mitigation, preparedness, response, and recovery. This work focuses on two crucial operations belonging to the response phase: shelter location and evacuation of endangered populations. Specifically, the ultimate scope of this paper is to present some applications of a scenario-based mixed-integer two-stage linear program which integrates shelter location with two different types of evacuation, self- (or car-based) evacuation and supported- (or bus-based) evacuation, namely the Scenario-Indexed Shelter Location and Evacuation Routing (SISLER) model. The SISLER model is solved using an off-the-shelf optimization software, whose performance is improved through the addition of some valid inequalities which are added at the root node of the solution tree to improve the lower bound. Computational results are reported for both testbed instances and a realistic case study (Sioux Falls network). The analysis of the solutions provides some useful managerial insights for relevant stakeholders working within the shelter location and evacuation planning area, such as emergency management practitioners and public service providers.