Graph fragmentation problem for natural disaster management

Abstract

Natural disasters represent a threaten for the existence of human beings. Given its remarkable importance, operational researchers should contribute to provide rationale decisions. In this paper we study a purely combinatorial problem that models management disasters, called Graph Fragmentation Problem, or GFP for short. The problem belongs to the NP -Hard class. As corollary, finding the optimal protection scheme is prohibitive for large populations. First, we review the problem and its properties. Then, we introduce a mathematical programming formulation and exact resolution for small instances. Finally, we discuss feasible model extensions and trends for future work.