The challenge that all fire protection engineers face is that nobody knows exactly when or where a fire will occur, under what conditions, and who will be at risk. In part, this is because one cannot predict the future. However, it is also because building and fire regulations are used to manage the risk. The building and fire regulatory system is complex and comprehensive, which for most buildings results in a generally tolerable level of fire performance. It also means that unknown or unacceptable life safety or financial loss concerns might exist in any given building, particularly if there are attributes of the building, its occupants, processes or mission, which are not specifically addressed by applicable codes and standards. One way to determine whether such a potential exists is by undertaking a fire risk assessment of the building or facility.


The general aim of fire risk assessment (FRA) is to identify and characterize the fire risks of concern and provide information for fire risk management decisions. The intent is to answer three basic questions: what can happen (what can go wrong), how likely is it that it will happen, and if it does happen, what are the consequences?3 FRA is distinguished from fire hazard analysis (FHA) and consequence analysis by the inclusion of an estimate of likelihood of occurrence, in addition to assessment of those factors that could lead to a fire and the impact should a fire occur.

FRA involves several steps, including identifying the objectives of the assessment, the metrics for assessment, the hazards of concern and the potential fire scenarios, conducting frequency and consequence analyses on the scenarios of concern, and estimating the risk associated with the scenarios. In some cases, FRA may be extended to assessment of options to mitigate the risk (either through reducing the likelihood of occurrence or magnitude of consequences), although this is also part of the risk management process. One framework for the FRA process is shown in Figure 1.4

Figure 1. Fire Risk Assessment Process4

Risk Assessment Objectives, Metrics and Thresholds

Some of the most important steps in the FRA process are identifying the objectives of the risk assessment, the measure(s) that will be used to express risk, and how the risk measures will be presented or communicated for decision making purposes.

For example, a high-level goal might be to "provide an environment for occupants that is reasonably safe from fire by protection of occupants not intimate with initial fire development and improvement of the survivability of occupants intimate with the initial fire development.”5

A first question might be: which occupants – all of them, only the most sensitive population, another subgroup? One then needs to ask under what conditions – smoke inhalation, radiant energy exposure, high temperatures, all, others, at any time of day, or under any circumstances? Characterizing the population and their risk thresholds is important as it will help drive scenarios of consideration and risk estimation and evaluation later in the process.

The same holds true for financial loss objectives, as might be associated with property loss (direct) or operational continuity (indirect). Is the focus structure only, contents, all contents or only some, contents and the structure? Is the impact related only to local operations, or is there an exposure somewhere in the supply chain or market delivery? What one chooses to address can influence the assessment, and whether or not all scenarios of concern are selected will depend on the focus.

How one chooses to measure and present the risk is equally important and can make a significant difference in terms of how the risk is perceived. For example, the life safety risk discussed above could be expressed in terms of the ratio of fire deaths to population, which could be expressed as 9.6×10-6 for the general population, or 2.99×10-7 for nonresidential buildings. The audience may or may not have a feel for what this means, so a comparative number might be given, such as risk of death in an automobile accident, which is much higher at 1.03×10-4. They might still not appreciate the numbers, so one might use reciprocals, i.e., 1 in 104,161, 1 in 3,344,481 and 1 in 9,708 respectively. One might also choose fire-per-building type, risk of untenable conditions, or some other metric.

Hazard, Event and Scenario Identification
As used here, a hazard is a condition or physical situation with a potential to result in harm to the focus of the risk assessment (e.g., people, property, operational continuity). The threat posed by the hazard is a basis for identifying scenarios. If the potential for undesirable consequences of the hazard manifests in an occurrence, that constitutes an event. A fire scenario is a fire incident characterized by a sequence of events. Fire hazards include heat sources (for ignition) and fuels (type, arrangement, products of combustion). An initiation hazard might be heating equipment. A contributing hazard might be an earthquake, which could cause the heating equipment to come in contact with an unintended fuel source. If the fuel ignites because the heating equipment was knocked over in an earthquake, that is an event. If the fire spreads to adjacent fuels and continues to grow, that is a scenario.

A fire scenario is a qualitative, time-sequence-based description of a fire that identifies key events that characterize the fire. It should describe the fire from initiation until burnout or extinguishment, including performance of passive or active fire protection systems that may be present. It should also identify outcomes in terms of whom or what is exposed to the fire and what the magnitude or severity of the harm is. Scenario clusters are groups of scenarios that have some, but not necessarily all characteristics in common, and are expressed at a level of detail appropriate for engineering analysis. Scenario clustering is needed because any individual scenario (sequence of events) will have negligible frequency data. For example, a fire scenario may start as: "A lit candle tips over onto an upholstered chair in a living room. The chair ignites and the fire grows.” Frequency data for this exact scenario may not exist. However, a scenario cluster might be: "An open flame ignites combustible fuel package within a living area.” Data to support frequency assessment of this scenario cluster could be found in resources such as NFPA fire statistics reports.

Frequency Analysis, Consequence Analysis and Data
A key factor that distinguishes FRA from fire hazard analysis (FHA) is the inclusion of an estimate of the likelihood that an event or scenario will occur. For the frequency analysis, data are needed from reliable sources. This may include entities such as NFPA, which report fire statistics; insurance companies, which collect fire data; and manufacturers or others, such as the Center for Chemical Process Safety, which have component and system reliability data.6 Databases such as NFIRS7 capture data on extent of fire spread (e.g., contained to item of origin, room of origin, etc.), which can be helpful in looking at reliability of containment.

Consequence analyses are often undertaken using analytical or computational tools, assessing such factors as performance of a building’s fire protection systems for the defined fire scenario. However, they can also make use of historical data, from similar operations or occupancies, at least to benchmark the process. Expert judgment can also be applied for screening purposes.

Risk Estimation
To develop a risk estimate, one combines information generated during the frequency and consequence analyses of the scenarios of concern. This can be accomplished in a variety of ways, including qualitatively, semiquantitatively and quantitatively. Qualitative approaches treat both frequencies and consequences qualitatively, and include methods such as risk matrices and risk indices. The NFPA Fire Safety Evaluation System,8 the risk matrix approach in MIL-STD-882D,9 and the risk binning approach outlined in DOE-STD-300910 are examples of this. An example of the risk binning and risk matrix approach is illustrated in Tables 1 and 2 and Figure 2.11

Consequence Level Impact on Populace Impact on Property/Operations
High (H) Immediate fatalities, acute injuries—immediately life threatening or permanently disabling Damage > $XX million – building destroyed and surrounding property damaged
Moderate (M) Serious injuries, permanent disabilities, hospitalization required $YY < damage < $XX million – major equipment destroyed, minor impact on surroundings
Low (L) Minor injuries, no permanent disabilities, no hospitalization Damage < $YY – reparable damage to building, signifi cant operational downtime, no impact on surroundings
Negligible (N) Negligible injuries Minor repairs to building required, minimal operational downtime

Table 1. Possible Consequence Ranking Criteria11

Acronym Description Frequency Level
(median time to event)
A Anticipated, expected >10–2/yr
(<100 years)
Common incidents that may occur several times during the lifetime of the building
U Unlikely 10–4 < f <10–2/yr
(100 to10,000 years)
Events that are not anticipated to occur during the lifetime of the facility. Natural phenomena of this probability class include UBC-level earthquake, 100-year flood, maximum wind gust, etc.
EU Extremely unlikely 10–6 < f <10–4/yr
(10,000 to 1 million years)
Events that will probably not occur during the life cycle of the building
BEU Beyond extremely unlikely <10–6/yr
(>1 million years)
All other accidents

Table 2. Example Frequency Criteria Used for Probability Ranking11

Figure 2. Consequence Ranking, Frequency Ranking and Risk Matrix11

Semi-quantitative approches combine quantitative and qualitative aspects. Semi-quantitative frequency approaches use sources such as actuarial data, which provides data for quantitative frequency analysis, but qualitative consequence analysis. Semi-quantitative consequence approaches use fire effects modeling for quantitative consequence analysis and treat frequencies qualitatively. These approaches can be used with event trees or other analysis frameworks. Event tree analysis (ETA) is often used to analyze complex situations with several possible scenarios, where several fire or life safety systems are in place or are being considered. Event trees are developed for a scenario, with frequencies and consequences described, and the risk then estimated. One method for quantifying fire risk from multiple fire scenarios is given as:11,12

Σ Riski = Σ Lossi x Fi

Riski = Risk associated with scenario i
Lossi = Loss associated with scenario i
Fi = Frequency of scenario i occurring

A final type of risk estimation technique is the benefit-cost approach, which either determines costs required to achieve various levels of risk reduction, or determines optimum levels of fire protection based against expected losses. These approaches are often employed within the insurance industry and by facility management to balance acceptance, mitigation, transfer, or avoidance decisions. Output is often expressed in terms of expected losses or costs, which include capital expenditures, maintenance costs, and expected losses.


Given the growing interest in the use of risk assessment techniques for building fire safety evaluation, a number of organizations have prepared guidance documents that are useful to designers and approval authorities (i.e., AHJs) in relation to buildings.* These guides are not risk assessment methodologies or risk analysis techniques. Rather, they are directed at assisting practitioners in selecting the appropriate methodology for any given building and ensuring that the process of risk assessment and approval is undertaken in a proper engineering manner.

SFPE Engineering Guide: Fire Risk Assessment
The SFPE Engineering Guide: Fire Risk Assessment4 is aimed at qualified practitioners who are undertaking design and evaluation of buildings and/ or process fire safety. The document provides guidance on the selection and use of risk assessment techniques and provides a recommended process to follow. The SFPE Fire Risk Assessment Guide does not specify particular risk assessment methods or techniques. However, it highlights

  • A recommended process for fire risk assessment (Figure 1)
  • Tools that may be used for hazard identification
  • Sources of data for risk assessment
  • Approaches to consequence modeling
  • Methods for calculating fire risk
  • Documentation of fire risk assessment

The SFPE Guide is structured to follow the flowchart represented in Figure 1, providing guidance and information association with each step in the process. This information is supported with many references and a comprehensive list of information sources for further reading for each step of the risk assessment process.

NFPA 551, Guide for the Evaluation of Fire Risk Assessments
NFPA 551, Guide for the Evaluation of Fire Risk Assessments,13 was developed in the United States in recognition of the fact that fire risk assessment methods are increasingly being used in developing fire and life safety solutions for buildings and other facilities. This guidance document is directed at those responsible for approving or evaluating fire and life safety solutions based on a fire risk assessment. It provides a framework that describes the properties of a fire risk assessment, particularly where it is being used in a performance-based regulatory framework. As a result, this guide is suited to a building or fire official or other authority having jurisdiction required to evaluate or approve a building design where the design is being supported by a fire risk assessment. Like the SFPE Engineering Guide: Fire Risk Assessment, NFPA 551 neither specifies particular fire risk assessment methods nor attempts to set acceptance criteria. Rather, it sets out the technical review process and documentation that should be used by those evaluating or approving. The review process is illustrated in Figure 3.

Figure 3. NFPA 551 Review Process
(Reprinted with permission from NFPA 551-2013, Guide for the Evaluation of Fire Risk Assessments, Copyright © 2013, National Fire Protection Association, Quincy, MA. This reprinted material is not the complete and official position of the NFPA on the referenced subject, which is represented only by the guide in its entirety.)

NFPA 551 defines five categories of fire risk assessment methods in order of increasing complexity, namely

  • Qualitative methods
  • Semi-qualitative criteria-based methods
  • Semi-qualitative consequence methods
  • Quantitative methods
  • Cost-benefit risk methods

It highlights the importance of identifying the objectives of any fire risk assessment and other factors that should be considered by those undertaking fire risk assessments. For each of the five categories of methods, the characteristics of each approach are identified, and issues of inputs and outputs, assumptions and limitations, selection of fire scenarios, and uncertainty are discussed.

BS 7974-7, Probabilistic Risk Assessment
The British Standards Institute (BSI) provides a number of fire-related design standards. A framework for the application of fire safety engineering principles for the design of buildings is provided within BS 7974. This document is supported by the Published Document series PD 7974 Parts 0 to 7. The final document, Part 7, provides guidance for the probabilistic risk assessment of buildings.14 The document provides a framework for risk assessment commensurate with a number of approaches. Specifically, the document provides guidance with regard to acceptance criteria for life safety and financial assessments, which may use either comparative or absolute methodologies.

The absolute criteria for individual risks and societal risk are provided. The logic tree is illustrated using both event trees and fault trees. An assessment methodology using complex analysis techniques is also provided. The annex to this document provides guidance about the probability of fire starting, depending on the type and use of the building. Further, the average area damaged and the distribution of damage are provided.

There are also valuable statistics on the frequency distribution of the numbers of deaths attributed to fire, the probability of flashover, and reliability data concerning active and passive fire safety systems. These data are principally based on U.K. fire statistics recorded over a representative sample period and as such are considered a valuable source of information, although generally applicable to U.K. projects.

ISO 16732-1 Fire Safety Engineering – Fire Risk Assessment
ISO 16732-115 provides the conceptual basis for fire risk assessment by stating the principles underlying the quantification and interpretation of fire-related risk. The principles and concepts outlined in the standard can be applied to any fire safety objectives, including life safety, conservation of property, business continuity, preservation of heritage, and protection of the environment.

The fire risk principles discussed in the standard apply to all fire-related phenomena and user applications, which means that the principles can be applied to all types of fire scenarios. In ISO 16732-1, principles underlying the quantification of risk are presented in terms of the steps to be taken in conducting a fire risk assessment. These quantification steps are initially placed in the context of the overall management of fire risk and then explained within the context of fire safety engineering.

The use of scenarios and the characterization of probability (or the closely related measure of frequency) and consequence are then described as steps in fire risk estimation, leading to the quantification of combined fire risk. Guidance is also provided on the use of the information generated, i.e., on the interpretation of fire risk.

Finally, there is guidance on methods of uncertainty analysis, in which the uncertainty associated with the fire risk estimates is determined and the implications of that uncertainty are interpreted and assessed. As described by ISO 16732-1, risk management includes risk assessment, but also typically includes risk treatment, risk acceptance, and risk communication (see Figure 4).

Figure 4. ISO Fire Risk Management Concept15


There have been new chapters added to the SFPE Handbook of Fire Protection Engineering on various aspects of fire risk assessment, e.g., by industry, occupancy type, and sector (built environment, transportation), with each new edition. As another indicator of the growing interest in fire risk assessment, and the desire for information relative to tools and techniques for fire risk assessment, a number of textbooks have been published in the last decade.

Evaluation of Fire Safety,16 while not strictly a text on fire risk assessment, includes many aspects of fire risk assessment throughout. Written by a collection of five leading authorities in fire safety engineering, the text includes chapters on sources of statistical fire loss data, measurements of fire risk, various fire risk evaluation methods (e.g., point systems, logic trees, stochastic fire risk modeling, and the fire safety concepts tree and derivative approaches). It provides a comprehensive suite of information for anyone embarking on fire safety evaluation of the built environment.

Following the tragic events of Sept. 11, 2001, the text Extreme Event Mitigation in Buildings: Analysis and Design17 was published to provide a resource for understanding and assessing building performance under extreme events. While not focused solely on fire, the text provides information on assessing likelihood of occurrence, potential impacts, and strategies for mitigation for a wide range of extreme events – natural, technological, and deliberate, while aiming to achieve a balance of acceptable levels of risk, performance, and cost. The text outlines how risk-informed performance-based analyses can be used to help make important risk mitigation decisions.

In 2007, a trio of risk experts from Australia published the book, Risk Analysis in Building Fire Safety Engineering.18 As the title implies, this text is focused on tools and techniques that are fundamental to applying risk concepts in fire safety engineering. It starts with elements of probability theory required for the understanding of risk analysis, then transitions into various tools for risk analysis, including the beta reliability index, Monte Carlo analysis, event tree and fault tree analysis, and cost benefit analysis. Several chapters are then provided relative to modeling the probabilistic and stochastic aspects of fire safety systems. Case studies are provided to illustrate the application of these concepts in performance-based fire safety design.

Principles of Fire Risk Assessment in Buildings19 is presented in two parts: Part I overviews simple approaches to fire risk assessment, and Part II outlines a fundamental approach to fire risk assessment—considering fire growth, smoke spread, occupant response, and other factors using fire risk assessment concepts. This book was authored by an expert in the field who has developed models for fire risk assessment.

Most recently, two renowned fire risk experts from the UK collaborated on the 2011 text, Quantitative Risk Assessment in Fire Safety.20 This text presents a broad ranging discussion of qualitative, semi-quantitative and quantitative risk assessment techniques—discussing sources of data, structuring of the assessment technique, assessment, and evaluation. Probabilistic and stochastic analysis of fire development and spread and response of fire safety systems is also provided. Reliability of fire safety systems, performance of people, and effectiveness of the fire services are also presented.

These texts, as well as others written for specific industries, hazards, and risks, provide fire protection engineers with additional resources for tackling the challenges of building fire risk analysis.

Brian J. Meacham is with Worcester Polytechnic Institute.

*The discussion is excerpted from the chapter, Building Fire Risk Analysis, to be published in the 5th Edition of the SFPE Handbook of Fire Protection Engineering.


  1. An Overview of the US Fire Problem, NFPA, Quincy, MA, September, 2012
  2. Badger, S. "Large-Loss Fires in the United States in 2011,” NFPA Journal, November/December 2012, pp. 60-67.
  3. Kaplan, S. and Garrick, B. "On the Quantitative Definition of Risk,” Risk Analysis, Vol. 1, No. 1, 1981.
  4. SFPE Engineering Guide – Fire Risk Assessment, Society of Fire Protection Engineers, Bethesda, MD, November 2006.
  5. NFPA 101, Life Safety Code, National Fire Protection Association, Quincy, MA, 2012.
  6. Guidelines for Process Equipment Reliability Data, Center for Chemical Process Safety, American Institute of Chemical Engineers, New York, NY, 1989.
  7. National Fire Incident Reporting System Version 5.0, Federal Emergency Management Agency, Washington, DC, 2011.
  8. NFPA 101A. Guide to Alternative Approach to the Life Safety Code, NFPA, Quincy, MA, 2013.
  9. MIL-STD-882D. System Safety, US Department of Defense Standard Practice, US Department of Defense, Washington, DC, 2010.
  10. Preparation Guide for US Department of Energy Nonreactor Nuclear Facility Documents Safety Analyses, US Department of Energy, Washington, DC, 2006.
  11. Meacham, B., Johnson, P., Charters, D., and Salisbury, M., "Building Fire Risk Analysis,” SFPE Handbook of Fire Protection Engineering, National Fire Protection Association, Quincy, MA, 2008.
  12. SFPE Engineering Guide to Performance-Based Fire Protection, Society of Fire Protection Engineers, Bethesda, MD, 2007.
  13. NFPA 551, Guide for the Evaluation of Fire Risk Assessments, National Fire Protection Association, Quincy, MA, 2013.
  14. BS 7974-7, "Part 7—Probabilistic Risk Assessment,” Code on the Application of Fire Safety Engineering Principles to the Design of Buildings, British Standards Institute, London, UK, 2003.
  15. ISO 16732-1:2012, Fire Safety Engineering—Guidance on Fire Risk Assessment, International Organization for Standardization, Geneva, Switzerland, 2012.
  16. Rasbash, D., Ramachandran, G., Kandola, B., Watts, J., and Law, M. Evaluation of Fire Safety, John Wiley & Sons Ltd, Chichester, England, 2004.
  17. Meacham, B. and Johann, M., Eds., Extreme Event Mitigation in Buildings: Analysis and Design, National Fire Protection Association, Quincy, MA, 2006.
  18. Hasofer, A., Beck, V., and Bennetts, I. Risk Analysis in Building Fire Safety Engineering, Butterworth-Heinemann, Oxford, England, 2007.
  19. Yung, D. Principles of Fire Risk Assessment in Buildings, John Wiley & Sons Ltd, Chichester, England, 2008.
  20. Ramachandran, G. and Charters, D. Quantitative Risk Assessment in Fire Safety, Spon Press, London, England, 2011.