Optimal deployment of emergency resources in sudden onset disasters

In this paper, we study the problem of allocating emergency units to affected sites and casualty groups in the initial hours after sudden-onset disasters to maximize the number of survivors. Survival rate of extricated casualties is high in the first hours following disasters, but the number of emergency (rescue and medical) units available to service casualties is often inadequate. Optimal allocation of units to the affected sites and casualty groups is necessary to maximize the number of survivors. In this paper, we determine: I) the best strategy to allocate rescue and medical units to affected sites; and II) the best strategy to allocate medical units to casualty groups at each site. The optimization models developed to answer the questions are tested on a case problem in the New Madrid Seismic Zone of Illinois to find simple and practical strategies for deploying emergency units. Results show that emergency units should be distributed among the sites in proportion to their casualty populations, regardless of the casualty mix ratios. We show that at each site, medical units should be streamed fairly among the casualty groups. Fair streaming leads to even workloads among the medical units assigned to each group of casualties. We show that casualty overflow does not make for any significant increase in the expected number of survivors, if the medical units are divided fairly among the casualty groups. However, the casualty overflow does make the expected number of survivors more robust against any likely biases in the medical units' streaming.