Performance methodology and fire scenario selection

Introduction

The performance-based methodology in fire protection engineering provides the opportunity to design solutions adapted to the particular characteristics of each building or installation. However, this design freedom must be accompanied by rigorous scenario selection and evaluation criteria that ensure a level of safety at least equivalent to that guaranteed by prescriptive regulatory solutions.

This article reviews the key aspects of these methodologies, emphasizing the need to contemplate different fire levels (“low fire” and “high fire”), the possibility of failures in fire protection systems and the use of safety margins that compensate for the uncertainties of simulation models.

2. The Role of ISO 16733-1 and ISO 23932-1 in Scenario Selection

The international standards ISO 16733-1 and ISO 23932-1 contemplate a series of analysis steps to properly select and define fire scenarios.

1. ISO 16733-1:2017 (section 6.3.6 and 8 steps for choosing scenarios)
    

• Emphasizes the importance of special conditions that can lead to atypical or more severe fires (external actions such as earthquakes, vandalism or extreme weather, as well as interaction with occupants).
• Highlights the reliability and availability of systems (sprinklers, SHEVS, detectors…) and consideration of their possible failure or poor maintenance.
• Provides risk classification matrices that help prioritize and justify the inclusion or exclusion of certain scenarios in the analysis.

2. ISO 23932-1:2017 (sections 11.1.3 and 11.1.4)
    

• Emphasizes the need to evaluate the probability of occurrence of each scenario, not only its severity.
• Proposes the introduction of safety factors or additional margins to compensate for the uncertainties inherent in any computational model or analytical simplification.

In practice, it is not enough to simulate a single “worst case”; it is essential to justify why certain scenarios are discarded and why others, seemingly less frequent but with very harmful consequences, must be analyzed.

3. Performance-Based Analysis: Key Aspects

3.1. Representative Set of Scenarios

Performance-based analysis is not complete if it does not include a range of situations that may occur in reality:

• Different fire sources or intensities: from incipient fires controlled with few sprinklers to hypothetical high-power fires.
• System failures: sprinkler systems that do not operate, fire doors that remain open, electrical failures in smoke extraction, etc.
• Variations in fire load or in occupant response (for example, reactions that worsen propagation or delay evacuation).

3.2. Safety Margins

Simulation models, both CFD and zone models, contain uncertainties (modeling of convective and radiant heat, water droplet distributions, evaporated water fraction, sprinkler activation assumptions, etc.). Hence the obligation to:

• Include safety factors in the most critical assumptions.
• Analyze the sensitivity of results to changes in main parameters. If a small adjustment can mean the difference between meeting or not meeting the desired temperature or visibility, an additional margin will be required.

4. “Low Fire” and “High Fire”: How to Integrate Them in SHEVS?

The UNE 23585 standard insists on differentiating between “low fire” and “high fire” in the design of the smoke and heat exhaust ventilation system (SHEVS). Often, performance-based studies focus on fire controlled quickly by few sprinklers (equivalent to “low fire”), and stop the heat release rate curve as soon as the first or third sprinkler activates. However:

• What happens if the fire grows to a level that requires massive sprinkler activation (close to the design operation area)?
• What happens if sprinklers take longer than expected to activate or if fire evolves vertically very quickly?

Saying that the probability is low does not exempt from studying how smoke would worsen (darker, denser) and its effect on visibility and evacuation. Fire safety engineering, by definition, does not design only for the most frequent case, but also for less probable events with critical consequences.

Something similar occurs when there are no sprinklers. High fire must also be considered, especially if there may be structural impact.

5. Smoke Flooding and System Reliability

5.1. Partial or Total Smoke Flooding

UNE 23585 admits flooding of stored loads only in specific cases (e.g., high-height silos with sprinklers). But in many performance-based designs, total smoke mixing is assumed due to system activation delays, and evaluation is focused on the 2-meter height layer for evacuation.

• Is this considered equivalent to regulatory compliance?
• Is the loss of smoke control in the upper part adequately justified, if standards require maintaining a smoke-free cushion above merchandise or at a certain distance from the ceiling?

If SHEVS is unable to recover smoke stratification because it is undersized for both “low fire” and “high fire”, the supposed equivalence with prescriptive regulations may be called into question.

5.2. Reliability and Failure Condition

No system enjoys absolute reliability. Statistics show that automatic sprinklers usually operate in 85%-90% of cases, but this implies that in 10%-15% of fires they may fail totally or partially (water supply failure, closed valves, obstructed heads, etc.). For SHEVS, average reliability may be even lower as it depends on dampers, fans, detectors, electrical panels, etc.

• In critical scenarios, SHEVS failure could seriously endanger evacuation and damage control, so it must be evaluated whether redundancy, independent supplies or greater robustness in electrical or pneumatic supply is required.

6. Conclusions and Recommendations

1. Comply with performance-based methodology
    

• ISO 16733-1 and ISO 23932-1 standards require selecting and justifying fire scenarios transparently, also contemplating the failure of protection systems.

2. Analyze scenarios with and without sprinklers
    

• “Low fire” does not exhaust design possibilities. Studying a “high fire” (or at least justifying its dismissal) is essential to ensure that SHEVS is well dimensioned.

3. Use safety margins
    

• Fire model assumptions and interaction with sprinkled water are subject to uncertainty, so safety factors or sensitivity studies must be included to strengthen conclusions.

4. Ensure equivalence
    

• If seeking “equivalence” with prescriptive solutions, it is advisable to prove that the actual capacity of SHEVS or structural behavior under fire conditions with or without sprinklers is not inferior to what standards require.

5. Do not forget reliability
    

• Since systems are not infallible, performance-based engineering must carefully assess the probability of failure and its impact on occupant evacuation and building integrity.

In conclusion, the performance-based methodology is a very powerful path to design efficient fire protection systems adapted to each case, but it is essential to rigorously comply with the steps and requirements established by reference standards. Only this way can we ensure that the required safety level is not reduced and that the resulting protection is truly equivalent —or even superior— to what prescriptive solutions intend to guarantee.

Summary References

• ISO 16733-1:2017. Fire safety engineering — Selection of design fire scenarios and design fires.
• ISO 23932-1:2017. Fire safety engineering — General principles.
• UNE 23585:2017. Requirements and calculation and design methods for designing SHEVS.
• Community of Madrid Guide for Review of Performance-Based Fire Protection Projects (2024).
• Sprinkler reliability statistics (various international sources: NFPA, FM Global, etc.).