INTRODUCTION

Broadly speaking, the use of fire models falls into two areas: design and simulation. To determine the design specification of the fire protection systems in a building, fire models, often in the form of tables and charts in design guides, provide data such as escape widths, distance between buildings and the capacity of smokecontrol systems. In many cases, this is all that is required as it can be shown that the advice in the Approved Documents has been followed, and this is deemed to satisfy the requirements of the Building Regulations. However, there are cases where the advice in the Approved Documents is not relevant or has not been applied, in which case fire models can be used to perform simulations to show how the proposed fire protection systems will function and to demonstrate that the requirements of the Building Regulations are met.

The role of fire models as part of a fire investigation is usually for hypothesis testing. Evidence collected from a fire incident can either be very precise (eg ‘When the firefighters arrived at 04:42 the bedroom had flashed over') or vague and possibly conflicting (eg different witnesses may report seeing different things at approximately the same time). Using fire models to conduct simulations can confirm that a sequence of events is (or is not) feasible or address ‘What if?' questions (eg ‘What happens if the door opens before the window breaks?'). This can help investigators establish a time line for the incident. The King's Cross underground fire in 1987 provides a good example of the use of fire modelling as a part of a fire investigation. Other examples are cited in BRE Digest DG 515.