Simulation-Assisted Failure Mode and Effects Analysis
Also known as: Simulation-FMEA, Monte Carlo FMEA, Simulation-based FMEA, SA-FMEA
Simulation-assisted FMEA enhances the classical Failure Mode and Effects Analysis by replacing point-estimate occurrence ratings with probabilistic simulation — typically Monte Carlo — to quantify failure probability distributions across a system's components. This yields statistically grounded Risk Priority Numbers (RPNs) rather than expert guesses, enabling more rigorous identification and prioritization of critical failure modes in complex engineering systems.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use simulation-assisted FMEA when failure rate data or physics-of-failure models are available and the system is complex enough that point-estimate RPNs are likely to mislead prioritization. It is particularly valuable for safety-critical systems (aerospace, medical devices, nuclear) where uncertainty in occurrence rates is high and the cost of misclassifying a critical failure mode is large. It is also appropriate when management requires probabilistic risk statements rather than ordinal ratings. Do not use it as a substitute for standard FMEA when failure data are absent — assigning arbitrary distributions without empirical basis inflates apparent precision without improving accuracy. For simple, well-understood systems with stable historical failure rates, standard FMEA is sufficient.
Strengths & limitations
- Replaces subjective integer Occurrence ratings with statistically quantified failure probabilities, improving RPN credibility.
- Produces confidence intervals around RPNs, making uncertainty in rankings explicit and guiding data collection priorities.
- Enables sensitivity analysis to identify which input uncertainties drive the most RPN variability.
- Supports quantitative verification that corrective actions achieve target risk reductions by re-running simulations post-change.
- Integrates naturally with reliability block diagrams and physics-of-failure models already used in engineering programs.
- Requires failure rate data or well-reasoned distributional assumptions — if data are poor, the simulation output is spuriously precise.
- Substantially more time and computational effort than standard FMEA; may not be justified for low-criticality systems.
- Simulation results are only as valid as the system model; incorrect functional decomposition or missing failure modes are not corrected by adding simulation.
- Requires statistical expertise to select appropriate distributions, validate model inputs, and interpret simulation outputs.
Frequently asked
What simulation method is most commonly used with FMEA?
Monte Carlo simulation is by far the most common approach. It samples randomly from the probability distributions assigned to Severity, Occurrence, and Detectability (or a subset) and computes RPN distributions across thousands of iterations. Discrete event simulation is used when the focus is on system availability and failure sequences over an operational timeline.
How many simulation iterations are needed?
As a practical rule, 10,000 iterations are sufficient for stable mean and variance estimates of RPNs for most engineering applications. If you need accurate estimates of tail probabilities (e.g., probability that RPN exceeds a critical threshold), 100,000 or more iterations may be needed. Always check for convergence by comparing outputs from successive run sizes.
Can simulation-assisted FMEA be performed in standard software?
Yes. Monte Carlo sampling can be implemented in spreadsheet add-ins such as @RISK or Crystal Ball, in MATLAB, Python (SciPy/NumPy), or dedicated reliability software such as ReliaSoft BlockSim. The FMEA worksheet structure remains the same; simulation replaces the fixed Occurrence integer with sampled draws.
Is this approach required by any industry standard?
No standard mandates simulation-assisted FMEA by name, but probabilistic safety assessment standards in aerospace (SAE ARP4761), nuclear (IEEE Std 352), and medical devices (ISO 14971) effectively require quantitative failure probability estimates that align naturally with this approach.
When should I use standard FMEA instead?
Use standard FMEA when failure rate data are unavailable or the system is in early concept stages, when the team lacks statistical expertise, or when a rapid, qualitative screening is the goal. Simulation adds value only when credible distributional inputs exist; without them, it adds computational complexity without improving decision quality.
Sources
- Stamatis, D. H. (2003). Failure Mode and Effect Analysis: FMEA from Theory to Execution (2nd ed.). ASQ Quality Press. ISBN: 978-0873895989
- Failure mode and effects analysis. Wikipedia. link ↗
How to cite this page
ScholarGate. (2026, June 3). Simulation-Assisted Failure Mode and Effects Analysis. ScholarGate. https://scholargate.app/en/experimental-design/simulation-assisted-failure-mode-and-effects-analysis
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Event Tree AnalysisReliability↔ compare
- Failure Mode and Effects AnalysisExperimental design↔ compare
- Fault Tree AnalysisReliability↔ compare
- MONTE-CARLO-SIMULATIONDecision-making↔ compare
- Reliability AnalysisReliability↔ compare
- Simulation-assisted reliability analysisExperimental design↔ compare