Abstract
While evacuation along with opening public shelters is vital to save the lives of disaster survivors, many of them run the risk of refusing evacuation thereby jeopardising their safety. The primary reason for taking such a risk is the inferior conditions commonly experienced in public shelters, including the noise of crowds, constant lighting, privacy concerns, and safety apprehensions, combined with the absence of other viable sheltering options. Moreover, the potential transmission of infectious diseases, such as tuberculosis and COVID-19, deters disaster victims further from seeking refuge in these public shelters. In this study, we propose a four-echelon network design to allow for small capacity shelters that can enhance welfare and public health. A two-stage stochastic mixed-integer linear programming model is formulated to optimise the selection of opening nodes in each of the four echelons under different disaster scenarios, in addition to determining the logistical operations. This model considers the tradeoff between maximising total welfare-enhanced sheltering occupancy and minimising associated costs. A novel problem-specific scenario-based cutting-plane algorithm is developed to efficiently solve the model while providing practical advantages. The effectiveness of our developed approach is demonstrated through a case study in Texas – the most disaster-prone state in the United States.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call