Stochastic modelling for risk assessment

Abstract

A probabilistic risk assessment method is described in this paper for the analysis of expected risk to life. The method involves stochastic, as well as deterministic, modeling of building occupant evacuation process and the attainment of untenable conditions due to fires in buildings. Both the number of occupants leaving a building at a given time and the time to untenable conditions are treated as random variables. A general expression for calculating the expected number of death is given. With the postulated Poisson distribution for the number of occupants leaving a building and uniform distribution for the time to untenability, a simple expression for the calculation of the expected number of deaths is obtained. An example is given to illustrate the application of the method. Copyright © International Association for Fire Safety Science.