The AP decision is a rule table over Severity, Occurrence and Detection. The grid below shows the simplified view used in training; the production engine implements the full AIAG-VDA 2019 ruleset.
Unlike RPN, Action Priority ensures Severity always dominates the decision. A safety-critical failure is never masked by low occurrence.
The legacy RPN approach (Severity × Occurrence × Detection) produced misleading scores. An S=10 / O=2 / D=2 failure scored 40, while an S=3 / O=5 / D=4 failure scored 60 — even though the first failure can kill a customer and the second is a cosmetic nuisance. Teams chased the higher number and ignored the safety issue.
AP fixes this by anchoring priority to Severity first, then layering Occurrence and Detection. See the full comparison in RPN vs Action Priority.
| Severity | Occurrence | Detection | Action Priority |
|---|---|---|---|
| 9–10 (safety / regulatory) | Any | Any | High |
| 7–8 (loss of primary function) | 4–10 | Any | High |
| 7–8 | 2–3 | 5–10 | High |
| 7–8 | 2–3 | 1–4 | Medium |
| 4–6 | 6–10 | 5–10 | High |
| 4–6 | 4–5 | Any | Medium |
| 4–6 | 2–3 | 1–4 | Low |
| 2–3 (annoyance) | Any | Any | Low |
| 1 (no effect) | Any | Any | Low |
Failure mode: drilled Ø8.0 H7 hole exceeds upper limit. Severity = 7 (press fit lost, customer return). Occurrence = 4 (drill wear historically frequent). Detection = 6 (sample bore-gauge check). AP = High.
Adding a tool-life counter drops Occurrence to 2. With Severity 7, Occurrence 2, Detection 6, AP becomes High still — Severity 7 with weak Detection stays High. Adding 100% in-line bore gauging takes Detection to 2 and finally drops AP to Medium. Full example: PFMEA CNC machining.
| AP | Required action | Documentation |
|---|---|---|
| High | Identify and implement actions; justify if none taken | Mandatory written justification |
| Medium | Should take action; consider design / process changes | Recommended written rationale |
| Low | Action optional; monitor | Record the decision |
Qhubio calculates AP deterministically from the rated S/O/D values using the full AIAG-VDA 2019 matrix. AI never adjusts the result — the logic is fixed in the backend, so two reviewers always see the same AP for the same ratings. Try it in the FMEA generator or read about the AI FMEA software.
This page targets the primary keyword FMEA Action Priority and the cluster of queries actually used by quality engineers and auditors searching for AP guidance.
| Query type | Examples | Where answered on this page |
|---|---|---|
| Definition | what is action priority in fmea, ap fmea meaning, aiag vda action priority definition | Intro, decision grid, plain-English section |
| Comparison | rpn vs action priority, ap vs rpn fmea, action priority replaced rpn | Why AIAG-VDA replaced RPN section + dedicated RPN page |
| Calculation | how to calculate action priority, ap matrix aiag vda, action priority table excel | Action Priority decision table |
| Decision logic | high medium low fmea action, when is ap high, action priority high meaning | How High/Medium/Low translate into actions |
| Examples | action priority example, fmea ap example, pfmea action priority case study | Worked CNC machining example + linked example pages |
| Audit | action priority audit findings, justify low ap, ap documentation iatf | Common AP mistakes + auditor guidance below |
Auditors do not check the Action Priority value in isolation. They check whether the decision behind it is defensible. A High AP without a documented action plan, or a Low AP that hides a real risk by inflating Detection, both fail the same test: the team cannot explain the logic in front of a reviewer.
Pair this page with the FMEA audit checklist before any IATF 16949 or customer audit, and review the common FMEA mistakes list for the engineering anti-patterns behind these findings.
Industry context changes the practical interpretation of AP. The same S=7 / O=3 / D=5 row carries very different downstream actions in automotive vs medical vs electronics.
| Industry | Failure mode | S/O/D | AP | Typical action |
|---|---|---|---|---|
| Automotive (Tier-1) | Stamping burr causing assembly interference | 7 / 3 / 5 | High | Install 100% in-line vision; update Control Plan reaction |
| Medical device | Sterilization cycle parameter drift | 9 / 2 / 4 | High | Add redundant sensor + parameter lock; update DHR per ISO 13485 |
| Electronics / PCBA | Cold solder joint on safety-critical net | 8 / 3 / 6 | High | Add AOI with auto-reject and X-ray sampling on high-risk nets |
| CNC machining | Bore oversize due to tool wear | 7 / 4 / 6 | High | Tool-life counter + in-line air gauge — see CNC example |
| Packaging | Label mis-orientation | 3 / 5 / 4 | Low | Document and monitor; reaction plan only |
| Assembly | Under-torque on non-safety fastener | 5 / 4 / 5 | Medium | Add data-logging torque tool + audit frequency |
For more detail see the PFMEA CNC machining example, the PCB assembly example, and the broader FMEA example page.
If your team still maintains legacy FMEAs alongside new AIAG-VDA 2019 work, the table below covers the five questions that come up in nearly every migration workshop.
| Question | RPN behavior | Action Priority behavior |
|---|---|---|
| Does it weigh Severity first? | No — equal weight | Yes — Severity dominates |
| Can a safety failure score 'low'? | Yes — common | No — never |
| Is it auditable as a decision rule? | No — threshold per site | Yes — single AIAG-VDA matrix |
| Can teams game it? | Yes — inflate Detection | Limited — rule table resists it |
| Recommended for new programs? | No | Yes |
Full migration walkthrough: FMEA RPN vs Action Priority guide. For automated AP calculation that cannot drift between projects, see AIAG-VDA FMEA software and the FMEA generator.
How Action Priority connects to other FMEA concepts, standards, examples and software.
Three tiers, automatically derived from the Qhubio knowledge graph. The tier containing this guide is highlighted.
Curated next steps — methodology guides, worked examples, and the relevant tool.
Move from theory to a working FMEA. Same methodology, three entry points depending on how you want to start.
Qhubio applies the AIAG-VDA methodology automatically — no Excel formulas, no inconsistent rating scales, no scattered spreadsheets.
Generate your first FMEAHelp us improve Qhubio. Analytics help us understand which FMEA features are useful, which pages need improvement, and where users encounter problems. No advertising. No selling personal data. See our Cookie Policy.