Dynamic resource allocation for forest fire risk management

The occurrence of one or more simultaneous forest fires requires that decision makers are able to solve a complex problem of resources allocation to fight the fires. The efficiency of the emergency management system can be deeply affected by these decisions, which, in turn, are influenced by the predicted behavior of forest fires. In this paper, a methodology is introduced, based on system modeling and optimization, to manage resource allocation, for real time control of forest fires, over a regional area. While several studies have proposed different models of forest fire dynamics, a proper formalization related to the decision support aspect has not been so deeply investigated. In this work, a graph model and a mathematical formalization are introduced in order to describe the territory under consideration, and the dynamics of the detected fires. A specific case study in Liguria region (Italy) is presented, in order to highlight the feasibility of the proposed approach.