AIAG-VDA FMEA 2019: Why Action Priority Replaced RPN and How to Apply It
If your organization is transitioning from the traditional FMEA approach to the AIAG & VDA FMEA Handbook (first edition, 2019), the biggest practical change is how you prioritize risk. The old method multiplied severity, occurrence, and detection into a Risk Priority Number. The new method uses a logic-based lookup called Action Priority. This guide explains what changed, why it changed, and how to apply Action Priority alongside or instead of RPN in your FMEAs.
FMEA Action Priority vs. RPN: What Changed in the AIAG-VDA 2019 Handbook
The AIAG & VDA FMEA Handbook merged the North American (AIAG 4th Edition) and German (VDA Volume 4) approaches into a single standard. Among the changes—the 7-step process, structure analysis, foundation FMEAs—the replacement of RPN with Action Priority is the one that affects every FMEA session immediately.
How RPN Works (Traditional Method)
RPN = Severity × Occurrence × Detection. Each factor is rated 1–10. The product ranges from 1 to 1,000. Teams set a threshold (commonly 100 or 125) and require corrective action for any failure chain above it.
How Action Priority Works (AIAG-VDA Method)
Action Priority uses the same three 1–10 input ratings but does not multiply them. Instead, it feeds them through a logic table that outputs one of three levels:
- High (H) — Action required. The failure mode presents an unacceptable risk and must be reduced through new or improved controls.
- Medium (M) — Action should be taken. The failure mode presents a risk that the team should evaluate for reduction opportunities.
- Low (L) — Action could be taken. The current risk level is acceptable, though improvement is optional.
The critical difference: AP always evaluates severity first. A severity 9–10 failure (safety or regulatory impact) is rated High priority regardless of how rare or detectable it is. RPN cannot do this because multiplication treats all three factors equally.
Side-by-Side Comparison: RPN vs. Action Priority
| Dimension | RPN (Traditional) | Action Priority (AIAG-VDA) |
|---|---|---|
| Formula | S × O × D (multiplication) | Logic table lookup (severity-first) |
| Output | Numeric score: 1–1,000 | Categorical level: H, M, or L |
| Severity weighting | Equal to O and D | Evaluated first; severity 9–10 always triggers H |
| Threshold | Organization-defined (100, 125, 200 are common) | Built into the standard: H = must act, M = should act, L = optional |
| Known flaw | Different risk profiles produce identical RPNs (e.g., S=10,O=1,D=1 = 10 vs. S=2,O=5,D=1 = 10) | Logic table resolves these: the S=10 case is always H |
| Adoption | Still used in SAE J1739, IEC 60812, ISO 14971, medical device FMEAs | Required for automotive IATF 16949 supply chain (AIAG-VDA alignment) |
| Audit expectation | Accepted where customer or standard specifies RPN | Expected by OEM customers in automotive; increasingly adopted in aerospace |
The RPN Flaw That Drove the Change
The fundamental problem with RPN is that multiplying ordinal scales produces mathematically misleading results. Consider three failure chains from the same PFMEA:
| Failure Chain | S | O | D | RPN | AP |
|---|---|---|---|---|---|
| Missing crimp (safety hazard, no warning) | 10 | 2 | 3 | 60 | H |
| Cosmetic scratch (customer annoyance) | 3 | 5 | 4 | 60 | L |
| Dimensional drift (assembly fit issue) | 6 | 4 | 5 | 120 | M |
The missing crimp and cosmetic scratch have identical RPNs of 60. If your threshold is 100, both fall below it—neither triggers action. But one is a severity-10 safety hazard. Under Action Priority, the missing crimp is High (mandatory action) while the cosmetic scratch is Low. This is the outcome a quality engineer would reach intuitively; AP encodes that intuition into a repeatable method.
For a full walkthrough of how to assign the severity, occurrence, and detection ratings that feed into both methods, see our guide on calculating RPN in FMEA.
How the AP Lookup Table Works
The AIAG & VDA FMEA Handbook provides separate AP tables for DFMEA and PFMEA. The logic follows this hierarchy:
- Severity 9–10 (safety/regulatory) — Almost always High, regardless of occurrence and detection. The only exception is extremely low occurrence (1) combined with near-certain detection (1), which may yield Medium in some table entries.
- Severity 5–8 with moderate-to-high occurrence — Typically Medium or High depending on detection effectiveness.
- Severity 1–4 (minor effects) — Usually Low unless both occurrence and detection are poor (high numbers).
The table is not a simple formula you can memorize. It contains over 100 S-O-D combinations with specific H/M/L assignments. This is by design: the ASQ and AIAG intended the table to replace subjective threshold-setting with a standardized, severity-weighted decision framework.
You can look up the AP result for any combination of ratings using our RPN and Action Priority calculator, which implements both the PFMEA and DFMEA tables.
When to Use RPN, AP, or Both
Switching methods is not always straightforward. Here is a decision framework based on what we see in practice:
- Automotive supply chain (IATF 16949) — Use AP. Your OEM customers expect AIAG-VDA format, and auditors will look for the AP column. If you still carry RPN for legacy comparison, that is fine, but AP is the governing metric.
- Aerospace (AS9100) — RPN is still the primary method per SAE J1739 and ARP5580. Some programs are adopting AP voluntarily because the severity-first logic aligns with safety-critical culture.
- Medical devices (ISO 14971) — Risk management under ISO 14971 uses a risk matrix (probability vs. severity) rather than RPN or AP directly. However, many medical device FMEAs still calculate RPN internally. AP adoption is rare in this sector.
- General manufacturing (ISO 9001) — No standard mandates either method. Choose based on your risk philosophy: if severity-first prioritization matters, use AP. If your teams are comfortable with RPN thresholds and they produce reasonable outcomes, there is no compliance reason to switch.
Transition Approach: Running Both Methods in Parallel
Most teams that migrate from RPN to AP do it gradually. The practical approach:
- Add the AP column to your existing FMEA template. Keep the RPN column for now. Calculate both from the same S, O, D ratings—no additional rating work required.
- Compare prioritization for your top 20 failure modes. Where RPN and AP agree, the transition is transparent. Where they disagree (RPN says low, AP says High), examine the failure mode. These are almost always high-severity, low-occurrence items—exactly the cases RPN handles poorly.
- Shift action decisions to AP. Once your team sees that AP catches safety-critical items RPN misses, the transition argument makes itself. Drop the RPN column from new FMEAs once confidence is established.
- Update your control plan linkage. If your control plan references RPN thresholds for enhanced controls, update them to reference AP = H or AP = M instead.
Key Takeaways
Action Priority is not a cosmetic rename of RPN. It fixes a structural flaw—the equal weighting of severity, occurrence, and detection—that caused real misalignment between RPN rankings and engineering judgment. If your current FMEA process occasionally lets a severity-10 failure slip below the action threshold because its occurrence and detection scores are low, AP will fix that immediately.
To test the difference on your own data, enter a few failure chains into the RPN and Action Priority calculator and compare the outputs side by side.