A two-stage stochastic optimization for disaster rescue resource distribution considering multiple disasters

In disaster rescue, multiple resource allocation strategies before and after disasters are a key research issue. Response planning for sudden-onset disasters must take into account the inherent uncertainties associated with the catastrophe and potential ensuing disasters. This article investigates the problem of resource allocation optimization in the context of natural disasters accompanied by the occurrence of multiple secondary hazards. A two-stage stochastic optimization model is proposed to simulate the scenario of random occurrence and severity of multiple natural hazards to optimize the resource allocation of rescue teams, warehousing items and medical resources when multiple natural hazards occur in combination. A particle swarm optimization (PSO) algorithm is designed and compared with a commercial solver (CPLEX) and lower bound at various scales. The results show that PSO performs optimally for large-scale cases. A case study, undertaken in Beijing, demonstrates that this method can improve the estimated casualty rescue success rate to 90.6%.