Multiple emergency resource allocation for concurrent incidents in natural disasters

Abstract

Emergency resource allocation, which mainly deals with how to efficiently allocate emergency resources of rescue agencies to process emerging incidents and reduce casualties and economic losses, has been one of the most fundamental and essential problems in emergency management for natural disasters. However, to our best knowledge, the existing deterministic methods concentrate on emergency resource allocation only for a single incident with constraints on response time or emergency resource cost separately, which is too simple to be applied in complex emergency scenarios. To this end, in this paper we commit ourselves to a challenging problem of allocating multiple emergency resources to multiple concurrent incidents in a parallel manner, taking response time and emergency resource cost altogether. For this purpose, we first define disaster response coalition as a form of team play. Besides, we propose a multiply constrained integer linear programming model and develop a differential evolution and heuristic based search algorithm to explore overlapping disaster response coalitions for concurrent incidents. Finally, the comparative experimental results show that technically, the parallel allocation of emergency resources for concurrent incidents is significantly more economical and efficient than the traditional serial allocation.