CoChat

Application of Monte Carlo Simulation to Study the Probability of Confidence Level under the PFMEA’s Action Priority

Open access
Sun, Jia-Jeng, Yeh, Tsu-Ming, Pai, Fan-Yun · 2022 · Mathematics

The DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index; 1 of the 42 sources its claims cite were checked for support.

Abstract

Failure mode and effects analysis (FMEA) is the most commonly used risk evaluation tool in industry and academia. After four revisions, the US Automotive Industry Action Groups (AIAG) and German Association of the Automotive Industry (VDA) issued the latest FMEA manual, called AIAG and VDA FMEA Handbook Edition 1, in June 2019. Risk priority number (RPN) in the old-edition FMEA is replaced with action priority (AP), where the numerical evaluation of severity (S), occurrence (O), and detection (D) are referred to in the AP form for judging high (H), medium (M), and low (L) priority in order to ensure appropriate actions for improving prevention or detection control. When evaluating design (D) or process (P) in FMEA, the FMEA team has to refer to the evaluation criteria for S, O, and D, so as to reduce the difference in the evaluation reference and fairness. Since the criteria evaluation form is the qualitative rating standard with semantic judgment, evaluation errors are likely to occur when the team judges S, O, and D. The FMEA cases in this study are preceded by the confidence level (CL) of the S, O, and D evaluation standards and the setting of a confidence interval (CI) for the actual evaluation events. With discrete nonuniform distribution as the simulation setting, Monte Carlo simulation is applied several times to evaluate the probability before and after the evaluation, which is compared with the AP form to confirm the probability values of high, medium, and low priority. It provides reference for the FMEA cross-functional team, improving the originally non-AP events. Finally, the AP calculated in the simulation is compared and analyzed with the RPN sequence to verify the judgment of better actions with AP.

Overview

CoChat Summary

TL;DR

This paper applies Monte Carlo simulation with a 95% confidence level/interval to PFMEA severity, occurrence, and detection ratings, then maps simulated S/O/D combinations to action priority (AP) H/M/L probabilities. Using two semiconductor-plant PFMEA cases, it shows AP judgments can contradict old-edition RPN-based improvement decisions.

Key findings

  • Case 1: simulated AP probabilities were H = 95.243%, M = 4.757%, L = 0.000%; the old-edition RPN was 96, which would normally suggest no improvement, but AP H >95% and L = 0% indicated improvement was reasonable.
  • Case 2: simulated AP probabilities were H = 2.335%, M = 5.291%, L = 92.374%; the old-edition RPN was 108, which would normally suggest high risk and improvement, but AP L was 92.374%, the opposite conclusion.
  • Under the RPN criterion of 100, Case 1 had RPN ≥100 probability 44.444% and <100 probability 55.556%; Case 2 had RPN ≥100 probability 51.851% and <100 probability 48.149%.
  • The paper argues RPN weights S, O, and D equally, while AP prioritizes severity first, then frequency, then detection; examples include S=9, O=6, D=1 giving AP H with RPN 54, and S=9, O=2, D=2 giving AP L.

Methods

  • A cross-functional team evaluates PFMEA process steps according to the AIAG and VDA FMEA Handbook, rating S, O, and D and judging H/M/L from the AP form.
  • CL 95% is set for each FMEA rating: professional scoring is assumed to fall within the 95% CL, using ±1.96σ and a ±1 cutoff; S, O, and D are simulated with discrete nonuniform distributions in Crystal Ball (CB) software over 1000-time individual evaluations, with adjacent ratings at 2.5%, the main rating at 95%, and the other adjacent rating at 2.5%.
  • The simulation generates 27 S/O/D combinations (3 × 3 × 3), multiplies their probabilities, maps each to AP H/M/L, verifies the probability sum is 100%, and compares results with old-edition RPN. Two PFMEA cases from a semiconductor plant equipment FMEA are used, with original S/O/D of 8, 6, 2 (Case 1) and 9, 3, 4 (Case 2).

Limitations

  • The paper states no dedicated limitations section; it acknowledges that AP research is scant, the study is initial, qualitative S/O/D ratings may produce evaluation errors, and AP H/M/L is judged as a single result.
  • Evident from the text: the method assumes expert scores fall within the 95% CL and treats S/O/D evaluation as based on a normal distribution; the simulation uses 1000 trials and restricts S/O/D to 1–10.
  • Evident from the text: only two PFMEA cases from one semiconductor plant are analyzed, with no external validation or comparison to actual failure outcomes stated.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/math10152596

Cites

35 of 35 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Evolution of the approach to FMEA. Data source: Dariusz et al . [21] Figure 1. Evolution of the approach to FMEA. Data source: Dariusz et al. [21].
Figure 1 Evolution of the approach to FMEA. Data source: Dariusz et al . [21] Figure 1. Evolution of the approach to FMEA. Data source: Dariusz et al. [21].CC-BY · From the PDF
Figure 2 Outputting reliability, tolerance, and confidence interval with simulation. Data source: Wittwer [31].
Figure 2 Outputting reliability, tolerance, and confidence interval with simulation. Data source: Wittwer [31].CC-BY · From the PDF
Figure 3 Construct of a CI percentile. Data source: Glen [44]. Figure 3. Construct of a CI percentile. Data source: Glen [44]. Values of 90%, 95%, or 99% are generally used as the CL indicator in normal distri-
Figure 3 Construct of a CI percentile. Data source: Glen [44]. Figure 3. Construct of a CI percentile. Data source: Glen [44]. Values of 90%, 95%, or 99% are generally used as the CL indicator in normal distri-CC-BY · From the PDF
Figure 4 N (0,1) 90%, 95%, 99% cutoff values. Data source: Heumann and Schomaker [45]. Figure 4. N(0,1) 90%, 95%, 99% cutoff values. Data source: Heumann and Schomaker [45].
Figure 4 N (0,1) 90%, 95%, 99% cutoff values. Data source: Heumann and Schomaker [45]. Figure 4. N(0,1) 90%, 95%, 99% cutoff values. Data source: Heumann and Schomaker [45].CC-BY · From the PDF
Figure 5 The histogram of PMF discrete nonuniform distribution. Data source: Cross Validated Figure 5. The histogram of PMF discrete nonuniform distribution. Data source: Cross Validated [48].
Figure 5 The histogram of PMF discrete nonuniform distribution. Data source: Cross Validated Figure 5. The histogram of PMF discrete nonuniform distribution. Data source: Cross Validated [48].CC-BY · From the PDF
Figure 6 Graph showing േ 1.96 σ and 95% CI for the mean . Data source: Altman et al. [49]. Figure 6. Graph showing ± 1.96 σ and 95% CI for the mean. Data source: Altman et al. [49].
Figure 6 Graph showing േ 1.96 σ and 95% CI for the mean . Data source: Altman et al. [49]. Figure 6. Graph showing ± 1.96 σ and 95% CI for the mean. Data source: Altman et al. [49].CC-BY · From the PDF
Figure 7 Custom assumptions of CB software for the Monte Carlo simulation Figure 7. Custom assumptions of CB software for the Monte Carlo simulation. Figure 7. Custom assumptions of CB software for the Monte Carlo simulation
Figure 7 Custom assumptions of CB software for the Monte Carlo simulation Figure 7. Custom assumptions of CB software for the Monte Carlo simulation. Figure 7. Custom assumptions of CB software for the Monte Carlo simulationCC-BY · From the PDF
Figure 8 Forecast values of the Monte Carlo simulation with 1000 trials according to research definition in CB software . Figure 8 . Forecast values of the Monte Carlo simulation with 1000 trials according to research definition in CB software . Figure 8. Forecast values of the Monte Carlo simulation with 1000 trials according to research definition in CB software.
Figure 8 Forecast values of the Monte Carlo simulation with 1000 trials according to research definition in CB software . Figure 8 . Forecast values of the Monte Carlo simulation with 1000 trials according to research definition in CB software . Figure 8. Forecast values of the Monte Carlo simulation with 1000 trials according to research definition in CB software.CC-BY · From the PDF
Image 9
CC-BY · From the PDF
Figure 9 Case 1: Discrete nonuniform distribution of S in CB software. Figure 9. Case 1: Discrete nonuniform distribution of S in CB software. , x FOR PEER REVIEW
Figure 9 Case 1: Discrete nonuniform distribution of S in CB software. Figure 9. Case 1: Discrete nonuniform distribution of S in CB software. , x FOR PEER REVIEWCC-BY · From the PDF
Figure 10 Case 1: Discrete nonuniform distribution of O in CB software. Figure 10. Case 1: Discrete nonuniform distribution of O in CB software. Figure 10. Case 1: Discrete nonuniform distribution of O in CB software.
Figure 10 Case 1: Discrete nonuniform distribution of O in CB software. Figure 10. Case 1: Discrete nonuniform distribution of O in CB software. Figure 10. Case 1: Discrete nonuniform distribution of O in CB software.CC-BY · From the PDF
Image 12
CC-BY · From the PDF
Figure 13 Case 1: Monte Carlo simulation of O in CB software. Figure 13. Case 1: Monte Carlo simulation of O in CB software. Figure 13. Case 1: Monte Carlo simulation of O in CB software.
Figure 13 Case 1: Monte Carlo simulation of O in CB software. Figure 13. Case 1: Monte Carlo simulation of O in CB software. Figure 13. Case 1: Monte Carlo simulation of O in CB software.CC-BY · From the PDF
Figure 14 Case 1: Monte Carlo simulation of D in CB software. Figure 14. Case 1: Monte Carlo simulation of D in CB software.
Figure 14 Case 1: Monte Carlo simulation of D in CB software. Figure 14. Case 1: Monte Carlo simulation of D in CB software.CC-BY · From the PDF
Image 15
CC-BY · From the PDF
Image 16
CC-BY · From the PDF
Figure 17 Case 2: Monte Carlo simulation of D in CB software. Figure 17. Case 2: Monte Carlo simulation of D in CB software.
Figure 17 Case 2: Monte Carlo simulation of D in CB software. Figure 17. Case 2: Monte Carlo simulation of D in CB software.CC-BY · From the PDF

Verification

Claims the paper makes

1 Valid · 2 Not valid · 32 Not checked · 32 Own results and interpretations

10 more claims

Create a free account to see all 67 claims and to verify the 22 references still open.

See all claims — free account
  • Probably not a substantial claim. 43
  • It may state two separate claims as one. 25
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

4 of 4

Citations

Cites

Show all 35 references

Research works printed but not linked: 10

  1. Yang, 2009, Dose FMEA try hard to get rid of RPN constraint?, Qual. Mag., 45, 37
  2. Hsu, 2013, How to determine a sample size in survey research, J. Taiwan Sport Pedagog., 8, 89
  3. Lin, 2008, Establishing E-service quality model based on FMEA: An example of 3g mobile-communication industry, Logist. Manag. Rev., 3, 47
  4. Frunza, 2020, Study regarding the application of the FMEA (failure modes and effects analysis) method to improve food safety in food services, Sci. Pap. Ser. D Anim. Sci.—Int. Sess. Sci. Commun. Fac. Anim. Sci., 63, 360
  5. Anackovski, 2021, Action priority in new FMEA as factor for resources management in risk reduction, Int. J. Sci. Eng. Res., 12, 59
  6. Malvin, H.K., and Paula, A.W. (2008). Monte Carlo Methods, Wiley. [2nd ed.].
  7. Law, A.M., and Kelton, W.D. (2000). Simulation Modeling and Analysis, McGraw-Hill.
  8. Eduardo, 2017, Stochastic risk analysis: Monte Carlo simulation and FMEA (Failure Mode and Effect Analysis), Rev. Espac., 38, 26
  9. Weiss, N.A. (2016). Introductory Statistics, Pearson & Addison Wesley. [10th ed.].
  10. Altman, D., Machin, D., Bryant, T., and Gardner, M. (2013). Statistics with Confidence: Confidence Intervals and Statistical Guidelines, John Wiley & Sons.

Cited by

Metadata

Authors
Sun, J.-J., Yeh, T.-M., & Pai, F.-Y.
Affiliations
  • IBM (United States) (US)
  • National Quemoy University (TW)
  • National Changhua University of Education (TW)
Corresponding author
  • Yeh, Tsu-Ming
  • Pai, Fan-Yun
Venue
Mathematics
Publisher
MDPI AG
Published
2022-07-25
Volume
10
Issue
15
Pages
2596
ISSN
2227-7390 (ISSN-L)
OpenAlex
W4287838586
Cited by
10
References
34
Citations by year
2026: 4 · 2025: 1 · 2024: 3 · 2023: 1 · 2022: 1
FWCI
0.91
Type
journal article
Language
en
Open access
Locations
  • Mathematics (publishedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Ministry of Science and Technology of Taiwan (MOST 108-2221-E-507-004-MY3)
Topics
  • Physical Sciences › Engineering › Safety, Risk, Reliability and Quality › Risk and Safety Analysis
  • Physical Sciences › Engineering › Safety, Risk, Reliability and Quality › Reliability and Maintenance Optimization
  • Social Sciences › Business, Management and Accounting › Strategy and Management › Management Systems and Quality Improvement
Keywords
failure mode and effects analysis, action priorities, risk priority number, confidence level, confidence intervals, evaluation criteria, risk assessment

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.