743. Comparison of metaheuristics to solve a simulation-based optimization problem for firebreak allocation
Invited abstract in session HB-7: Analytics for Prescribed Burning and Firebreaks Location in Forest Fires Prevention, cluster Use of Analytics in Forest Fires Management.
Thursday, 10:30-12:00Room: CE-210
Authors (first author is the speaker)
| 1. | David Palacios
|
| Departamento de Ingeniería Civil Industrial, Universidad de Chile | |
| 2. | Jaime Carrasco
|
| University of Chile | |
| 3. | Sebastián Dávila
|
| Industrial Engineering, Universidad de Santiago de Chile | |
| 4. | Andrés Weintraub
|
| Industrial engineering, University of Chile |
Abstract
Global warming and the overpopulation of previously uninhabited areas has led to an increase in forest fire episodes; it becomes urgent to control these events. One of the most used techniques are firebreaks - interventions in the vegetation of the forest to prevent the advance of fire. In this work, metaheuristics are studied, implemented and compared to locate firebreaks optimally in the landscape, prior to a forest fire episode and, thus, to reduce the impact of this event. For this purpose, research is carried out in the literature to find similar works using metaheuristics to find any algorithm able to adapt to the problem. Subsequently, Genetic Algorithm, GRASP and Tabu Search are tested in pilot implementations, and a greedy methodology is proposed to perform the local search of the algorithms in a more effective way. Results in the final implementation show good performance of Genetic Algorithm in scenarios of medium- and high-operational capability, as well as in low- and medium-stochasticity scenarios. Finally, these metaheuristics could be useful in the prevention against forest fires in certain geographical areas, such as Chile.
Keywords
- Metaheuristics
- Simulation
- Applications, Agriculture and Forestry
Status: accepted
Back to the list of papers