Looking back to the early 1980s, the use of fire models in the profession of fire safety engineering was limited. Since this time, the world has seen an unprecedented growth in computer technology and the design community experienced an increase in the demand for performance-based fire safety design. Because of these rapid changes in technology and the need to construct innovative buildings, the use of fire models has become considerably more common in our profession. In fact, fire models and are now considered an essential component of the fire protection engineer’s toolbox.
With this new technology, when used properly, fire protection engineers can more efficiently estimate fire related phenomena such as gas layer temperature and depth, flame height, heat flux, gas species concentration heat release rate, visibility and detector response to name a few. Unfortunately, as with any technological advancement that outpaces society’s ability to adapt quickly, harm can result. That is the case when an engineer does not select the appropriate fire model for the given application.
When a fire model is needed for a fire safety engineering analysis, selecting the best model can be a challenging task. For example, in many instances one or multiple models may be available. In other situations, an appropriate model may not be available. At the same time, there are numerous available models that vary in complexity. These models range from simple algebraic correlations to zone models to models that are based on the principles of computational fluid dynamics. In some cases, the most complex model may not be the best choice as is the case when a simple calculation performed on a spreadsheet can provide the same result.
It is the fire safety engineer’s responsibility to determine if a fire model is suitable. Unfortunately, with few exceptions, there is no formal process by which fire models are approved for a design project. To assist the engineer in selecting an appropriate fire model, in 2011, SFPE published its Engineering Guide for Substantiating a Fire Model for a Given Application.
The intent of the SFPE guide is to provide a framework for determining and documenting the suitability of a fire model for use in a specific fire protection application. Specifically, according to the SFPE guide, determining the suitability of a fire model is a five-step process: 1) defining the problem of interest, 2) selecting a candidate model, 3) verification and validation (V&V), 4) user effects and 5) documentation. The first three steps apply before the engineer performs an analysis using a model, and the last two steps occur after the analysis has been completed.
When looking at the first three steps in the SFPE guide, it seems V&V is a step that is often overlooked by the user. Basically, verification is a check that the equations in the model are solved correctly, and validation is a check that the right equations are being solved. Responsible model developers do a respectable job in showing their models are applicable over a wide range of uses as they have the expertise and resources to perform this type of work. Despite this fact, it is impossible for them to address every possible application. As such, before selecting a model for a particular problem, the model user cannot assume the model is capable of generating a useful result without any additional analysis. To assist the user with the V&V effort, the SFPE guide provides a list of exercises that the user can conduct to supplement the verification efforts of the developer and provides a procedure that the fire safety engineer can apply to validate a model for a specific application.
Documentation is another essential step in the fire model selection process that should not be overlooked. A report that documents the process should include how the model addresses the key physics involved in the problem being solved. The level of detail in this report may vary based on the complexity of the specific project, but should show how the steps in the SFPE guide were addressed.