TABLE OF CONTENTS In this chapter and Chapter 12 two models will be considered. One model describes fire growth and the other describes smoke spread. The stochastic version of the models is obtained by introducing randomness in the model.
The introduction of randomness in a deterministic model is discussed in Ref. [28]. Two methods are described. The first method, the more general one, consists in replacing the deterministic inputs to the model by random variables. The method implies that the major source of randomness is in the inputs, an assumption that is often realistic for fire models. The second method is to introduce random terms in the model equations, thus implying that the source of the randomness lies in the imperfection of the model. It is also possible to introduce randomness with both methods, as illustrated in Ref. [28].
The first method will be illustrated by the fire growth model. The second method will be illustrated by the smoke spread model in Chapter 12.
The fire growth model that will be analysed is called CESARE-Risk. It is based on the original article by Takeda and Yung [62], a recent report by Cooper and Yung [18], and the modification to the model by researchers at the Centre for Environmental Safety and Risk Engineering, at Victoria University of Technology (see He [34]). It has been used to generate temperature and smoke data for an integrated system model that...
