Fire Modeling and Computational Fluid Dynamics in Modern Fire Engineering

What Is Fire Modeling?

Fire modeling is the use of mathematical models — ranging from simple hand calculations to sophisticated computational fluid dynamics (CFD) simulations — to predict how fire, smoke, heat, and toxic gases will behave in a specific building under specific fire scenarios. It is the analytical foundation of performance-based fire protection design and is increasingly used to supplement — or in some cases replace — prescriptive code compliance on complex projects.

Fire modeling answers questions that prescriptive codes cannot: How long will the smoke layer remain above head height in a large atrium before conditions become untenable? What sprinkler head spacing is needed to suppress a specific commodity stored at a specific height? Will the proposed stair pressurization system maintain positive pressure under worst-case door opening scenarios? These are engineering questions, not code lookup exercises — and they require modeling to answer reliably.

Types of Fire Models

Hand Calculations and Algebraic Models

The simplest fire models are algebraic correlations derived from experimental fire research. The NFPA Handbook and the SFPE Handbook of Fire Protection Engineering contain hundreds of these correlations for estimating flame height, plume temperatures, ceiling jet velocities, smoke filling rates, and detector response times. These models are fast, transparent, and appropriate for simple geometries where the assumptions embedded in the correlations are clearly met.

Zone Models

Zone models divide a room into two layers — a hot upper zone containing fire gases and a cool lower zone containing relatively fresh air — and solve simplified equations governing the heat and mass transfer between them. The most widely used zone model is CFAST (Consolidated Fire and Smoke Transport), developed by NIST. Zone models are appropriate for single rooms and simple multi-compartment geometries where the two-zone assumption is valid.

CFD Models (Field Models)

CFD fire models solve the full Navier-Stokes equations governing fluid motion throughout the modeled space, providing a three-dimensional, time-resolved picture of fire behavior. The dominant CFD fire model in professional practice is FDS (Fire Dynamics Simulator), developed by NIST and freely available. FDS can model:

• Smoke movement through complex multi-story spaces including atria, tunnels, and transit stations
• Sprinkler activation timing under specific fire scenarios
• Detector response under specific smoke production and transport conditions
• The effectiveness of smoke control systems under specified ventilation conditions
• Temperature and visibility conditions along egress paths over time

Defining Fire Scenarios

The output of any fire model is only as reliable as the input fire scenario — and defining credible, worst-case fire scenarios is one of the most consequential judgments in performance-based fire engineering. NFPA 101 and SFPE guidelines identify key scenario parameters:

• Fire location — The most challenging location for occupant egress and suppression system response, not necessarily the most probable ignition location
• Heat release rate (HRR) — The rate at which the fire produces energy, typically expressed in kilowatts or megawatts. HRR curves for common fuel types (upholstered furniture, storage commodities, cooking oils) are available from fire research databases.
• Fire growth rate — How quickly the fire reaches its peak HRR. Standard growth rates (slow, medium, fast, ultra-fast) are characterized by t-squared fire models.
• Suppression — Whether sprinklers are assumed to activate and control the fire, and at what point in the scenario
• Ventilation conditions — Door and window positions, HVAC operation mode, and whether smoke control systems are active

Evacuation Modeling

Fire modeling addresses the fire side of the life safety equation — but occupant safety depends on both when conditions become untenable and when the last occupant reaches safety. Evacuation modeling addresses the latter, simulating the movement of building occupants from their locations at the time of alarm to points of safety outside the building.

The dominant evacuation modeling tools in professional practice are Pathfinder (Thunderhead Engineering) and buildingEXODUS (Fire Safety Engineering Group, University of Greenwich). These tools simulate:

• Pre-movement delay (the time between alarm and the start of occupant movement — often the most variable and consequential parameter)
• Walking speeds for different occupant types (ambulatory, mobility-impaired, elderly, children)
• Door and stair flow rates based on occupant density and geometry
• Behavioral responses including counter-flow, occupant merging at stair landings, and exit choice

The output of an evacuation model is compared against the output of the fire model to determine whether the “available safe egress time” (ASET) — the time until conditions become untenable in egress paths — exceeds the “required safe egress time” (RSET) — the time needed for all occupants to escape. ASET must exceed RSET with an appropriate safety factor.

Where Fire Modeling Is Applied

• Atria and large open spaces — Prescriptive codes are not well-suited to the unique smoke dynamics of multi-story atria; CFD modeling is routinely used to design and verify smoke exhaust systems
• Transit tunnels and stations — NFPA 130 performance criteria (tenable conditions in egress paths) are verified through CFD and evacuation modeling
• High-rise egress analysis — Verification that phased evacuation strategies provide adequate ASET margins
• Sprinkler design for unusual storage arrangements — Custom commodity classifications and storage geometries that fall outside standard NFPA 13 tables
• Historic buildings — Demonstrating equivalent life safety to prescriptive requirements when physical constraints prevent code-compliant modifications

Limitations and Responsible Use

Fire modeling is a powerful tool — and a tool that can produce convincingly wrong answers when misused. Key limitations that engineers and AHJs must understand:

• CFD models require validation against the specific scenario being modeled; applying a model outside its validated range produces unreliable results
• Grid sensitivity analysis is required to verify that results are independent of the computational mesh resolution chosen
• Uncertainty in input parameters (HRR, pre-movement time, material properties) should be explored through sensitivity analysis, not treated as fixed values
• The SFPE Guide to Performance-Based Fire Protection provides the professional framework for responsible fire modeling practice

Conclusion

Fire modeling has transformed fire protection engineering from a code-lookup discipline into an analytical engineering practice. The ability to simulate fire, smoke, and occupant behavior computationally — and to verify performance against quantified safety criteria — is what makes performance-based design possible. Used responsibly, with appropriate scenario selection, validated models, and honest uncertainty analysis, it is one of the most powerful tools available to the modern fire protection engineer.