What is a Fire and Explosion Risk Assessment (FERA)
- Carsten Stegelmann | Principal Consultant
- 1 day ago
- 5 min read
In this article:
In the design or modification of process facilities that handle flammable materials posing a fire or explosion risk, it is Critical to conduct a Fire and Explosion Risk Analysis (FERA)
The purpose of FERA is to systematically identify potential fire and explosion hazards in a process plant, evaluate their likelihood and consequences, and establish Design Accidental Loads (DeALs) for fires and explosions to be applied to support plant design and risk assessment.
FERA also evaluates the effects of Risk Reducing Measures (RRMs) proposed to either reduce the consequences of the DeALs (e.g. fire and gas detection, emergency shutdown, emergency depressurization) or to strengthen critical equipment and structures to withstand the DeALs (e.g. passive fire protection (PFP), structural survivability and redundancy. Ultimately, the FERA ensures that the plant design is robust against credible fire and explosion scenarios so that an initial event does not result in unacceptable escalation.
Examples of escalation include failure of structural integrity, loss of separation between hazardous and non-hazardous plant areas, or large releases of flammable or toxic inventories from process or storage facilities.
The purpose of this insight is to present the typical FERA methodology applied in process plant design. A schematic of a typical FERA process is shown in Figure 1.
Figure 1 Typical steps in FERA
Information Gathering
All FERAs begin with information gathering to understand the facility and its operation. This requires close collaboration with the design or operations teams. Accident and operating experience from similar facilities should also be reviewed where available.
Typical input documentation required includes Piping and Instrumentation Diagrams (P&IDs), heat and mass balances, process inventories, layout drawings, and the overall safety philosophy of the plant.
HAZID
Before starting the FERA, it is essential to identify all potential fire and explosion hazards. This is best achieved through a structured, multidisciplinary Hazard Identification (HAZID) workshop. Any hazards not identified at this stage will be missing from subsequent FERA modelling, regardless of how advanced the analysis software may be.
Based on the HAZID outcomes, relevant fire and explosion scenarios to be included in the FERA are identified. These typically cover Loss of Containment (LoC) scenarios of various sizes and locations, potentially leading to events such as jet fires, pool fires, or vapour cloud explosions (VCEs).
Frequency Analysis
In the frequency analysis, the LoC scenarios are defined (e.g. specific leak sizes or instantaneous inventory losses). Several LoC scenarios must often be analysed to achieve sufficient resolution of the fire and explosion risk.
Frequencies for each scenario are usually derived from historical data, such as the TNO "Purple Book", IOGP Risk Assessment Data Directory, or similar references. Alternatively, fault tree analysis (FTA) may be used to calculate frequencies based on identified failure mechanisms.
Since a release is only the initiating event, event trees are used to model subsequent outcomes depending on factors such as ignition probability, ignition timing (early or delayed), wind direction and speed, and safety barrier performance. The event tree quantifies the frequency of each outcome (fire or explosion) based on the initiating event and conditional probabilities.
Sometimes, a deterministic rather than a probabilistic approach is adopted in FERA. In that case, credible worst-case scenarios are selected, and their fire or explosion loads are used directly for design purposes. In such situations, frequency analysis becomes less relevant.
Consequence Analysis
The different fire and explosion accident event outcomes of a LoC scenario will typically represent different sorts of consequences such as jet fire, pool fire, flash fire, explosion etc.
The effect distances of these consequences can be calculated by empirical formulas, phenomenological models or ultimately by Computational Fluid Dynamics (CFD).
The consequence calculations are based on the release conditions of the LoC and fluid type released. Furthermore, ambient weather conditions and the general surroundings impact the consequence outcomes.
Effect distances of heat radiation levels and blast loads of certain levels are typically calculated.
Especially for fire consequences it is also important to calculate potential duration of the fire as part of the consequences. It is not the heat load itself but rather the heat dose that determines if escalation will occur for structural impairment.
Fire and Explosion DeALs
Based on frequency and consequence analyses, the fire and explosion Design Accidental Loads (DeALs) are determined. For probabilistic approaches, exceedance curves of heat dose or overpressure versus frequency are developed for various fire types and explosions in different plant areas.
DeALs are then defined based on target return frequencies (e.g. 10-5 to 10-4 per year). Critical structures, process equipment, piping, and safety systems - meaning components whose failure could cause significant escalation - must be able to withstand these loads for the risk to be judged As Low As Reasonably Practicable (ALARP).
Establishing Escalation Scenarios
Critical escalation targets (e.g. major equipment, structures, safety systems) are identified for each plant area. They are then evaluated for their ability to survive the area-specific DeALs. If a target cannot withstand them, design modifications or RRMs are required.
RRMs may aim to lower the DeALs (by reducing the likelihood or magnitude of events) or to increase the resistance of targets (e.g. by higher design strength or additional PFP). FERA is therefore an iterative process: RRMs are implemented until residual escalation risks are acceptable.
Commonalities with QRA
FERA and Quantitative Risk Assessment (QRA) share many methodological steps — Planning, System Definition, HAZID, Frequency Analysis, and Consequence Analysis. However, their objectives differ.
FERA ensures that plant design can withstand fires and explosions without leading to major escalation, while QRA quantifies the individual and societal risks primarily to personnel. Moreover, QRA addresses all Major Accident Hazards (MAHs), not only fire and explosion hazards.
FERA results feed directly into QRA, providing data on residual escalation risks. Both analyses apply risk acceptance criteria and ALARP principles, although their criteria differ.
Linkage to Other Risk and Safety Documentation
FERA may be reported as a single document or as several sub-studies, such as:
Fire Risk Assessments (possibly split by plant area).
Explosion Analyses.
Passive Fire Protection (PFP) Assessments.
Structural or Mechanical Integrity Studies.
FERA results inform key safety deliverables such as Fire and Explosion Strategies (FES), Safety Strategies, QRA, Fire and Gas (F&G) Philosophies, Emergency System Survivability Assessment (ESSA) etc.
Assumptions and Uncertainties
FERA modelling is not an exact science and relies on assumptions. As with any model, “garbage in equals garbage out” regardless of mathematical sophistication. Therefore, assumptions must be carefully documented, justified, and validated wherever possible.
Uncertainties in critical parameters should be examined through sensitivity analyses. FERA should therefore always be recognized as a decision-support tool, not a source of absolute truth.
