An Agent Based Model for Suppression of Wildfire Contagion
Abstract
As wildfires threaten Northwestern America, it becomes increasingly important to contain a fire soon after its ignition. Given a fire’s origin, the environment it propagates over, and a constrained amount of resources, we propose a framework for analyzing different containment strategies that aim to minimize wildfire impact. Inspired by probabilistic cellular automata models including Highly Optimized Tolerance, we simulate fire propagation according to biased stochastic probability functions on a 2 dimensional, hexagonal lattice. Rational artificial agents are tasked with building fire breaks to contain the spread in real time. Given only spatial context, agents are allowed to act on sites in their immediate neighborhood. By introducing different tuneable parameters such as topography, fire strength, and resource scarcity, we compare different containment strategies. Our goal is for simulations in this framework to inform fire fighting efforts in near real time.
An Agent Based Model for Suppression of Wildfire Contagion
As wildfires threaten Northwestern America, it becomes increasingly important to contain a fire soon after its ignition. Given a fire’s origin, the environment it propagates over, and a constrained amount of resources, we propose a framework for analyzing different containment strategies that aim to minimize wildfire impact. Inspired by probabilistic cellular automata models including Highly Optimized Tolerance, we simulate fire propagation according to biased stochastic probability functions on a 2 dimensional, hexagonal lattice. Rational artificial agents are tasked with building fire breaks to contain the spread in real time. Given only spatial context, agents are allowed to act on sites in their immediate neighborhood. By introducing different tuneable parameters such as topography, fire strength, and resource scarcity, we compare different containment strategies. Our goal is for simulations in this framework to inform fire fighting efforts in near real time.