Keywords
decision support systems, real time natural hazard management, environmental risk, forest fire, resource allocation problems
Start Date
1-7-2004 12:00 AM
Abstract
In the emergency management phase relevant to the occurrence of a catastrophic natural event, the efficiency of the emergency system can be deeply influenced by a correct assignment of the available resources to the “demand” centers, i.e., those elements of the territory that are directly involved in the event. In this paper, a general formulation of the real time optimal resource allocation problem is presented, and it is formalized as a mathematical programming problem. The dynamics relevant both to the resources over the territory and to the prediction of the behavior of the natural phenomenon are suitably taken into account. A graph model is introduced in order to describe the territory under consideration, and resources can be located either in the nodes or are in transit over the direct links. The phenomenon dynamics can be associated to the nodes on the territory, and specifically to the demand centers. The general approach has then been applied to the forest fire hazard. A simple model aiming to describe the spread of a fire over the territory has been developed, and it has been used for the formulation of the mathematical programming problem. A specific case study, relevant to the forest fire hazard in Liguria region (Italy), has been here defined and tested, and the main results are briefly reported and discussed.
Real Time Optimal Resource Allocation in Natural Hazard Management
In the emergency management phase relevant to the occurrence of a catastrophic natural event, the efficiency of the emergency system can be deeply influenced by a correct assignment of the available resources to the “demand” centers, i.e., those elements of the territory that are directly involved in the event. In this paper, a general formulation of the real time optimal resource allocation problem is presented, and it is formalized as a mathematical programming problem. The dynamics relevant both to the resources over the territory and to the prediction of the behavior of the natural phenomenon are suitably taken into account. A graph model is introduced in order to describe the territory under consideration, and resources can be located either in the nodes or are in transit over the direct links. The phenomenon dynamics can be associated to the nodes on the territory, and specifically to the demand centers. The general approach has then been applied to the forest fire hazard. A simple model aiming to describe the spread of a fire over the territory has been developed, and it has been used for the formulation of the mathematical programming problem. A specific case study, relevant to the forest fire hazard in Liguria region (Italy), has been here defined and tested, and the main results are briefly reported and discussed.
