Stochastic Programming Model for Fuel Treatment Management

Abstract

Due to the increased number and intensity of wild fires, the need for solutions that minimize the impact of fire are needed. Fuel treatment is one of the methods used to mitigate the effects of fire at a certain area. In this thesis, a two-stage stochastic programming model for fuel treatment management is constructed. The model optimizes the selection of areas for fuel treatment under budget and man-hour constraints. The process makes use of simulation tools like PHYGROW, which mimics the growth of vegetation after treatment, and FARSITE, which simulates the behavior of fire. The model minimizes the costs of fuel treatment as well as the potential losses when fire occurs. Texas Wild re Risk Assessment Model (TWRA) used by Texas Forest Service (TFS) is used to quantify risk at each area. The model is applied at TX 12, which is a re planning unit under the administration of TFS. Results show that the total of the expenditures on fuel treatment and the expected impact justify the efforts of fuel treatment.