The use of subjective probability assessments to predict forest fire occurrence

Publisher Website
Google Scholar
Abstract
This paper addresses the problem of predicting forest fire occurrence. A simple methodology is developed to elicit information, from experienced fire managers, for deriving subjective probability assessments concerning the number of fires that will be reported in their districts each day. The approach is based upon the assumption that such individuals can identify and classify similar fire environments with satisfactory consistency. Using the concept of subjective probability, a methodology is developed for combining classified experience with a decision maker's initial assessment concerning assessor behavior. A scoring rule is used to measure the accuracy of the assessments. The results of an experiment which was conducted in northern Ontario during the 1973 fire season are presented.

This publication has 0 references indexed in Scilit: