In high-hazard chemical environments, a single undetected risk can cascade into a catastrophic event. Safety professionals must therefore employ rigorous, structured methods to identify, evaluate, and control risks before they materialize. Two of the most widely adopted techniques are Failure Mode and Effects Analysis (FMEA) and Bowtie Analysis. While both serve the same ultimate purpose—preventing harm to people, assets, and the environment—they approach risk from distinct perspectives. Understanding their differences, strengths, and appropriate applications is essential for building a robust process safety program. This article provides a deep comparative study of FMEA and Bowtie Analysis specifically for chemical hazards, offering practical guidance on when and how to use each method effectively.

Failure Mode and Effects Analysis (FMEA)

Origins and Core Concept

FMEA was first developed in the 1940s by the U.S. military to improve the reliability of munitions systems and was later adopted by NASA and the automotive industry. Its core premise is straightforward: systematically examine each component or step in a process, identify all possible ways it could fail (failure modes), determine the consequences of those failures (effects), and assign a risk priority number (RPN) based on severity, occurrence, and detection. For chemical hazards, FMEA is often applied to unit operations such as reactors, distillation columns, storage tanks, and transfer lines.

Methodology and Key Steps

A typical FMEA follows these steps:

  1. Define the system scope and boundaries. What equipment, process steps, or operating modes will be analyzed?
  2. List all components or process steps. For chemical processes, this includes valves, pumps, instruments, control logic, manual operations, and interlock systems.
  3. Identify each potential failure mode. Examples: a pressure relief valve fails to open, a temperature sensor drifts high, a gasket leaks, a batch reactor overfills.
  4. Determine the effects of each failure. Could the failure lead to a loss of containment, an exothermic runaway, a flammable release, or a toxic exposure?
  5. Assign three numerical ratings: Severity (S) – how severe is the effect on safety, operations, or the environment (1-10 scale); Occurrence (O) – how likely is the failure to happen (1-10); Detection (D) – how easily could the failure be detected before causing harm (1-10).
  6. Calculate the Risk Priority Number (RPN = S × O × D).
  7. Prioritize failures with high RPNs and recommend actions to reduce severity, occurrence, or improve detection.
  8. Reassign ratings after corrective actions to verify improvement.

Strengths of FMEA for Chemical Hazards

  • Systematic and thorough. Every component is examined, leaving little room for oversight, especially in complex processes.
  • Quantitative prioritization. The RPN provides a clear, numbers-based ranking that helps teams allocate resources.
  • Well-documented. FMEA produces a detailed spreadsheet that serves as a living document for audits, training, and regulatory compliance.
  • Effective for early design stages. It can be applied during process design reviews (PHA) to catch weaknesses before equipment is built.
  • Familiar and standardized. Methods like AIAG FMEA, SAE J1739, and AIChE’s CCPS guidelines make training consistent.

Limitations of FMEA

  • Isolation of failures. FMEA typically evaluates single-point failures; it does not naturally handle combinations of failures or human errors that occur in sequence.
  • Limited causal visualization. The tabular format does not show how multiple threats can lead to the same consequence or how barriers are positioned.
  • Detection rating subjectivity. The D rating can be inconsistent between teams, leading to RPN values that may not reflect true risk.
  • Not ideal for analysing barrier effectiveness. FMEA can identify controls but does not model how barriers degrade or interact under stress.
  • Time-consuming for large systems. A full FMEA on a chemical plant with hundreds of equipment items can be overwhelming and may dilute focus on the most critical risks.

Bowtie Analysis

Origins and Core Concept

Bowtie Analysis, also known as a “bowtie diagram”, originated in the 1970s from the University of Queensland and was later popularized in the oil and gas industry, particularly through the work of the Centre for Chemical Process Safety (CCPS). It merges the left side of a fault tree (threats leading to a top event) with the right side of an event tree (consequences of the top event). The result is a single, intuitive diagram shaped like a bowtie, with the knot representing the loss of control or hazardous event.

Structure and Elements

A complete bowtie diagram consists of:

  • Hazard – the source of danger (e.g., flammable liquid stored under pressure).
  • Top Event – the moment when control over the hazard is lost (e.g., a pipeline rupture).
  • Threats – causes or initiating events that could lead to the top event (e.g., corrosion, overpressure, third-party damage).
  • Consequences – outcomes resulting from the top event if not mitigated (e.g., jet fire, toxic gas release, explosion).
  • Preventive Barriers – controls placed between threats and the top event to prevent it from happening (e.g., corrosion monitoring, pressure relief valves, operating procedures).
  • Mitigation Barriers – controls placed between the top event and consequences to reduce severity (e.g., fire suppression systems, blast walls, emergency shutdown valves).
  • Escalation Factors – conditions that can cause a barrier to fail (e.g., a corroded relief valve, a poorly trained operator).
  • Barrier Degradation Controls – actions to manage escalation factors (e.g., inspection schedules, competency assessments).

Methodology and Key Steps

  1. Define the hazard and top event. Choose a specific hazard scenario (e.g., handling of anhydrous ammonia).
  2. Identify all credible threats. Through brainstorming, historical data, or process hazard analysis (PHA) checklists.
  3. Identify all credible consequences. What are the worst-case and likely outcomes?
  4. Determine preventive barriers for each threat. Each barrier must be independent and auditable.
  5. Determine mitigation barriers for each consequence.
  6. Identify escalation factors that could weaken each barrier (e.g., “corrosion under insulation” for a pipeline integrity barrier).
  7. Add barrier degradation controls such as inspections, testing, or training.
  8. Review and validate the diagram with a cross-functional team.

Strengths of Bowtie Analysis for Chemical Hazards

  • Excellent visual communication. The diagram clearly shows how threats propagate to consequences and where barriers are placed. This makes it easier to explain complex risk scenarios to operators, management, and regulators.
  • Focus on barrier health. Bowtie forces the team to explicitly identify and manage barriers and their degradation factors, aligning with the principles of barrier-based safety management.
  • Handles multiple threats and consequences. A single top event can be linked to many initiating causes and outcomes, providing a holistic view.
  • Captures causal relationships. Unlike FMEA, Bowtie shows the chain of events, making it ideal for incident analysis and designing layered protection.
  • Auditable and dynamic. Bowtie diagrams can be updated as barriers are modified, and they serve as a powerful tool for demonstrating “adequate” risk control to regulators.

Limitations of Bowtie Analysis

  • Complexity and time investment. Building a thorough bowtie for a single hazard can take several hours, and linking multiple hazards into a unified bowtie management system can be daunting.
  • Subjectivity in barrier identification. Teams may disagree on what constitutes a valid barrier or how to judge its effectiveness.
  • Not naturally quantitative. Although some software tools allow adding probabilities, Bowtie is primarily qualitative. It does not easily produce a risk ranking like RPN.
  • Requires skilled facilitators. Effective bowtie workshops demand expert knowledge of the process, hazard analysis, and barrier management.
  • Risk of diagram clutter. For processes with many threats and consequences, the diagram can become overcrowded, reducing clarity.

Comparative Analysis: FMEA vs. Bowtie for Chemical Hazards

To help safety professionals decide which method to use—or whether to combine them—the table below compares the two techniques across key dimensions.

Focus and Scope

  • FMEA: Component- or step-focused. Evaluates every failure mode of each element in a process. Best for detailed analysis of single equipment or sequential operations.
  • Bowtie: Hazard- or top event-focused. Centers on a single loss-of-control event and maps all credible threats and consequences. Best for high-consequence scenarios where barrier integrity is critical.

Methodology and Output

  • FMEA: Tabular checklist with S-O-D scores and RPN. Output is a ranked list of failure modes with recommended actions.
  • Bowtie: Graphical diagram showing the complete risk pathway. Output is a visual map of barriers, escalation factors, and control measures.

Risk Prioritization

  • FMEA: RPN provides a numerical priority that can be sorted. However, the multiplication of three ordinal scales can produce misleading rankings if not carefully calibrated.
  • Bowtie: Priority is qualitative, based on the number and effectiveness of barriers. Risk level is often assessed through expert judgment or supplementary layer-of-protection analysis (LOPA).

Handling of Multiple Causes

  • FMEA: Treats each failure mode independently; does not naturally show how two or more failures could combine to cause an incident.
  • Bowtie: Explicitly illustrates multiple threats converging on the same top event, and multiple consequences diverging. It can also represent common-cause failures via escalation factors.

Data Requirements

  • FMEA: Requires detailed component failure rates, historical data on failure modes, and detection capabilities. Suitable when failure statistics are available.
  • Bowtie: Requires a thorough understanding of the hazard scenario, barrier performance, and degradation mechanisms. Often relies on expert elicitation rather than hard data.

Team Involvement

  • FMEA: Typically facilitated by a reliability or safety engineer with input from design and operations. Can be done with a small team.
  • Bowtie: Requires a cross-functional workshop including operators, process engineers, safety specialists, and management to ensure barrier credibility.

Best Use Cases

  • FMEA: New process design, early hazard identification, reliability improvement, and regulatory compliance (e.g., ISO 9001, ICH Q9 for pharmaceutical chemistry).
  • Bowtie: High-risk process safety scenarios, major accident hazard management (e.g., COMAH, Seveso), incident investigation, safety critical element assurance, and communicating with non-specialists.

Integrating FMEA and Bowtie Analysis for Chemical Hazard Management

Rather than viewing these methods as competitors, leading process safety programs use them as complementary tools. A common workflow is to start with a broad Hazard Identification (HAZID) or Preliminary Hazard Analysis (PHA) to list major hazards. For each major hazard, a bowtie diagram is built to map the full risk scenario. Then, for the most critical barriers or for specific equipment (e.g., a safety-instrumented system), an FMEA is performed to ensure that the barrier itself is reliable and that all its internal failure modes are understood.

For example, consider a facility handling hydrogen fluoride (HF). The bowtie might show the top event “Loss of containment from a storage sphere”, with threats including “corrosion under insulation”, “overfilling”, and “impact from a forklift”. One preventive barrier is the high-pressure alarm and shutdown system. A separate FMEA on this barrier could evaluate its components: pressure transmitter, logic solver, and shutdown valve—assigning RPNs to each failure mode such as “transmitter drift”, “valve stuck open”, or “logic solver failure”. This integrated approach ensures both the big picture and the detailed component reliability are covered.

Example: Applying Both Methods to a Chemical Plant

Step 1 – Bowtie for the main hazard

Hazard: Flammable liquid (toluene) in storage at ambient temperature and pressure.
Top Event: Catastrophic rupture of storage tank. Threats: Overpressure due to blocked vent, external fire, corrosion of tank bottom, pipe failure during transfer, operator error.
Consequences: Pool fire, vapor cloud explosion (VCE), toxic exposure to benzene (if toluene contains benzene), environmental contamination.
Preventive barriers: Vent system with flame arrestor, nitrogen blanketing, corrosion monitoring, permit-to-work for hot work near tank.
Mitigation barriers: Secondary containment dike, firewater system, gas detection and emergency shutdown, evacuation plan.

Step 2 – FMEA on one critical barrier

Suppose the “gas detection and emergency shutdown” mitigation barrier is identified as critical. An FMEA on that system would list failure modes such as: - Failure of the gas detector to detect toluene vapor (due to poisoning, calibration drift, or obstruction).
- Failure of the logic solver to process the signal.
- Failure of the isolation valve to close on command.
Each mode is rated for severity, occurrence, and detection. Actions could include dual detection, auto-diagnostic testing, and quarterly valve stroke testing.

Practical Considerations for Chemical Hazards

Regulatory and Standards Alignment

Many regulatory frameworks explicitly require hazard analysis methods. For example, the OSHA Process Safety Management (PSM) standard (29 CFR 1910.119) mandates a Process Hazard Analysis (PHA) using methods such as HAZOP, What-If, or Checklist—but does not specify FMEA or Bowtie. However, Bowtie is increasingly recognized by regulators in the UK (HSE), the Netherlands, and Australia for demonstrating “reasonably practicable” risk control under safety-case regimes. FMEA is more common in reliability-centered maintenance (RCM) programs and in industries governed by functional safety standards like IEC 61511 / ISA-84.

Software Tools

Both methods are supported by commercial software. For Bowtie, tools like BowTieXP, BowTie Pro, and the CCPS BowTie Library offer templates and built-in barrier libraries. For FMEA, packages such as RAM Commander, ReliaSoft, and custom spreadsheets are widespread. Some platforms now integrate both approaches, allowing users to link an FMEA table to a bowtie diagram barrier element.

Common Pitfalls

  • FMEA: Using generic RPN thresholds (e.g., RPN > 100) without calibrating to the specific risk tolerance of the facility. Also, focusing too much on low-severity failures and missing high-consequence, low-probability events.
  • Bowtie: Creating diagrams that are too large—trying to cover every hazard on one sheet, leading to confusion. Another pitfall is treating barriers as binary (present/absent) without assessing their actual performance under real-world conditions.

Conclusion

Failure Mode and Effects Analysis and Bowtie Analysis offer distinct yet complementary lenses for understanding and managing chemical hazards. FMEA excels at detailed, systematic identification of component-level failures and provides a prioritized, quantitative output that is ideal for early design reviews and reliability engineering. Bowtie Analysis shines in conveying the full risk pathway from hazard to consequence, emphasizing barrier health and communication across diverse stakeholders.

For a comprehensive process safety program, safety professionals should not default to one method alone. Instead, consider using Bowtie as the framework for major accident scenarios and then deploying FMEA to assure the reliability of critical barriers and equipment. This layered approach not only satisfies regulatory requirements but also builds a deeper, shared understanding of risk among engineers, operators, and management.

By selecting the right tool for the right context—and knowing how to combine them—organizations can significantly reduce the likelihood and severity of chemical incidents, protecting workers, communities, and the environment.

Further Reading and Resources

For deeper exploration of these topics, consult the following: