Evolutionary-Statistical System: A parallel method for improving forest fire spread prediction

• The ESS method addresses the problem of uncertainty in input parameters.
• ESS combines statistical concepts, HPC and Parallel Evolutionary Algorithms.
• ESS has been validated using a set of real burns carried out in Portugal.
• ESS has been compared against S2F2M, an already validated method based on statistics.

Fighting fires is a very risky job, where loss of life is a real possibility. Proper training is essential. Several firemen academies offer courses and programs whose goal is to enhance the ability of fire and emergency services to deal more effectively with fire. Among the tools that can be found in the training process are fire simulators, which are used both for training and for the prediction of forest fires. In many cases, the used simulators are based on models that present a series of limitations related to the need for a large number of input parameters. Moreover, such parameters often have some degree of uncertainty due to the impossibility of measuring all of them in real time. Therefore, they have to be estimated from indirect measurements, which negatively impacts on the output of the model. In this paper we present a method which combines Statistical Analysis with Parallel Evolutionary Algorithms to improve the quality of the model output.