Optimization of the Resources Management in Fighting Wildfires |
| |
Authors: | SUSANA MARTIN-FERNÁNDEZ EUGENIO MARTÍNEZ-FALERO J MANUEL PÉREZ-GONZÁLEZ |
| |
Institution: | (1) Departamento de Economia y Gestión de las Explotaciones e Industrias Forestales, E.T.S.I. de Montes., Ciudad Universitaria s/n, 28040 Madrid, Spain, ES;(2) Departamento de Economia y Gestión de las Explotaciones e Industrias Forestales, E.T.S.I. de Montes., Ciudad Universitaria s/n, 28040 Madrid, Spain, ES;(3) Departamento de Matemática Aplicada a los Recursos Naturales, E.T.S.I. de Montes., Ciudad Universitaria s/n, 28040 Madrid, Spain, ES |
| |
Abstract: | Wildfires lead to important economic, social, and environmental losses, especially in areas of Mediterranean climate where
they are of a high intensity and frequency. Over the past 30 years there has been a dramatic surge in the development and
use of fire spread models. However, given the chaotic nature of environmental systems, it is very difficult to develop real-time
fire-extinguishing models. This article proposes a method of optimizing the performance of wildfire fighting resources such
that losses are kept to a minimum. The optimization procedure includes discrete simulation algorithms and Bayesian optimization
methods for discrete and continuous problems (simulated annealing and Bayesian global optimization). Fast calculus algorithms
are applied to provide optimization outcomes in short periods of time such that the predictions of the model and the real
behavior of the fire, combat resources, and meteorological conditions are similar. In addition, adaptive algorithms take into
account the chaotic behavior of wildfire so that the system can be updated with data corresponding to the real situation to
obtain a new optimum solution. The application of this method to the Northwest Forest of Madrid (Spain) is also described.
This application allowed us to check that it is a helpful tool in the decision-making process. |
| |
Keywords: | : Bayesian methods Mockus adaptive model Simulated annealing Expansion models Wildfire management |
本文献已被 PubMed SpringerLink 等数据库收录! |
|