RPN in Simple Terms
Each failure mode in an FMEA gets three ratings, usually from 1 to 10. RPN multiplies them: RPN = Severity x Occurrence x Detection. Scores range from 1 to 1000, and higher numbers suggest a higher priority for action.
Limits of RPN
RPN treats the three ratings as equally important, which they are not. Very different risks can share the same score:
| Severity | Occurrence | Detection | RPN |
|---|---|---|---|
| 10 | 2 | 5 | 100 |
| 2 | 10 | 5 | 100 |
The first row could be a safety issue, the second a minor cosmetic defect. Teams also used fixed thresholds, such as "act on anything over 100", which led to rating games rather than real risk reduction.
For these reasons the AIAG-VDA FMEA Handbook (2019) replaced RPN with Action Priority, which uses tables that weigh severity first, then occurrence, then detection.
Example
In a PFMEA for a brake line crimp, "crimp too loose" has severity 9, occurrence 4 and detection 6, giving an RPN of 216. The team adds a crimp force monitor that rejects bad parts, which lowers detection to 3 and the RPN to 108. Under Action Priority, the high severity alone would still flag this item for review.
Is RPN Still Used?
Yes. Many FMEAs written to the older AIAG 4th edition still use RPN, and some industries and customers ask for it. New automotive FMEAs usually follow the AIAG-VDA method. OperaHive FMEA software works out Action Priority from your own rating tables. Read our FMEA guide for both approaches.