What should the RPN threshold be?

There is no universal threshold. Values such as 100 are mentioned as examples, but the right threshold depends on your organisation's procedure, the scale and the risk. The threshold is a policy choice that decides which rows need action; here it is used only for flagging.

If two failure modes have the same RPN, which comes first?

RPN cannot decide it. The general approach is to put the one with higher severity first; at equal RPN the tool sorts the higher-severity row on top. The team should still assess how often the cause occurs and how likely detection is.

Why are severity 9 and 10 highlighted separately?

Effects of severity 9–10 usually concern safety or legal compliance. Even when occurrence or detection difficulty looks low, the RPN may stay small; these rows should be handled without looking at RPN. The option keeps them visible in every case.

How should I assign the scores?

Take the scale definitions (what 1 and 10 mean) from your own FMEA guideline and have the whole team use the same scale. This tool provides no scale definitions and does not suggest scores.

How do I bring my table in from Excel?

Select the failure mode, S, O, D columns in Excel, copy and paste them into the "Paste / import CSV" field (tab-separated text is read). You can take the result out with "Download CSV".

FMEA RPN calculator

Enter failure modes with severity (S), occurrence (O) and detection (D) scores; the tool calculates RPN = S × O × D, ranks the rows, flags those above a threshold and draws a Pareto chart. Rows with high severity can be highlighted regardless of RPN.

Free tool · Predictive maintenance

Failure modes

Severity (S): how serious the effect is (Row 1)Occurrence (O): how often the cause occurs (Row 1)Detection (D): how hard it is to notice the failure before it reaches the customer (10 = cannot be detected) (Row 1)RPN168
Severity (S): how serious the effect is (Row 2)Occurrence (O): how often the cause occurs (Row 2)Detection (D): how hard it is to notice the failure before it reaches the customer (10 = cannot be detected) (Row 2)RPN160
Severity (S): how serious the effect is (Row 3)Occurrence (O): how often the cause occurs (Row 3)Detection (D): how hard it is to notice the failure before it reaches the customer (10 = cannot be detected) (Row 3)RPN105
Severity (S): how serious the effect is (Row 4)Occurrence (O): how often the cause occurs (Row 4)Detection (D): how hard it is to notice the failure before it reaches the customer (10 = cannot be detected) (Row 4)RPN140
Severity (S): how serious the effect is (Row 5)Occurrence (O): how often the cause occurs (Row 5)Detection (D): how hard it is to notice the failure before it reaches the customer (10 = cannot be detected) (Row 5)RPN120
Severity (S): how serious the effect is (Row 6)Occurrence (O): how often the cause occurs (Row 6)Detection (D): how hard it is to notice the failure before it reaches the customer (10 = cannot be detected) (Row 6)RPN162

Severity (S): how serious the effect is
Occurrence (O): how often the cause occurs
Detection (D): how hard it is to notice the failure before it reaches the customer (10 = cannot be detected)
Scores are whole numbers from 1 to 10. Take the scale definitions (what 1 and 10 mean) from your own FMEA guideline.

RPN threshold

Rows at or above the threshold are flagged. 100 is only an example; your own FMEA procedure sets the threshold.

Highlight rows with severity 9 and 10 regardless of RPN

A low RPN can hide a high-severity failure; this option always flags S ≥ 9 rows and lists them separately.

Paste / import CSV

One row per line: failure mode ; S ; O ; D (separator ; tab or comma). A header row is skipped; the rows replace the current ones. At most 200 rows.

The calculation runs in your browser; the failure modes and scores you enter are not sent anywhere.

RPN ranking

6 / 6 rows at or above the threshold

RPN ranking
Rank	Failure mode	S·O·D	RPN	Cumulative share
1	Bearing wear	7·6·4	168 (At or above threshold)	20%
2	Motor winding insulation breakdown▲ S ≥ 9	9·3·6	162 (At or above threshold)	39%
3	Lubrication loss	8·4·5	160 (At or above threshold)	57%
4	Temperature sensor drift	4·5·7	140 (At or above threshold)	74%
5	Shaft fracture▲ S ≥ 9	10·2·6	120 (At or above threshold)	88%
6	Loose connection	5·7·3	105 (At or above threshold)	100%

■ At or above threshold▲ S ≥ 9

Pareto chart: RPN, largest to smallest

Rows with severity 9 or above (review regardless of RPN)

Motor winding insulation breakdown S9 · O3 · D6 · RPN 162
Shaft fracture S10 · O2 · D6 · RPN 120

Limitations of RPN

The same RPN hides very different risks: S=9, O=2, D=4 (fatal but rare and visible) and S=2, O=6, D=6 (mild but frequent and invisible) both give 72. RPN cannot tell you which to address first.
The product takes only 120 distinct values, not 1000, and the scale is not evenly spaced; an RPN of 200 does not mean twice the risk of 100. Scores are rankings, not measurements.
S, O and D carry equal weight although severity is often decisive. That is why there is a separate highlight option for high-severity rows.
Scores vary with the person and the team; the same team should use the same scale. This tool does not apply the AIAG-VDA "Action Priority" (AP) approach and does not replace it.

Number of RPN values shared by rows with different severity: 0

RPN = S × O × D (1 – 1000) · cumulative share = running sum of the sorted RPNs / total RPN

RPN is an aid for ranking priorities; it is not a measure of risk and does not replace the team's judgement. For critical and safety-related failure modes, decide according to your organisation's FMEA procedure and the relevant standards.

Let's look at your high-priority failure modes together with monitoring data and decide which ones can give early warning.

Request a conversation

How to use it

A
Enter the failure modes and their severity, occurrence and detection scores (1–10), or paste your table as CSV.
B
Read the RPN ranking and the Pareto chart; set the threshold according to your own procedure.
C
Switch on the high-severity option to review serious failures hidden by a low RPN, and download the result as CSV.

How is RPN calculated?

The risk priority number is the product of three scores: severity (S), how serious the effect of the failure is; occurrence (O), how often the cause appears; detection (D), how hard it is for current controls to notice the failure before it has an effect. Each score is given from 1 to 10, so RPN lies between 1 and 1000.

Rows are sorted by RPN from largest to smallest. The cumulative share shows what percentage of the total RPN the first n rows make up, which helps you see where to concentrate the work.

Why not look at RPN alone?

RPN has known weaknesses: the same number can describe very different situations, the scale is not proportional, and severity is multiplied with equal weight to the other two scores. For this reason many organisations do not use RPN alone; high-severity failures are addressed regardless of RPN.

The tool flags rows with severity 9 and 10 independently of RPN and lists them separately. This is not a weighting method but a safety net. The AIAG-VDA "Action Priority" table is not part of this tool; if you use it, refer to the table in your own guideline.

How does monitoring data change the scores?

The detection score (D) expresses how well current controls notice the failure. If continuous measurement such as vibration, temperature or current monitoring can show the failure weeks before its effect, the D score drops and the RPN of that failure mode falls.

An FMEA study and a monitoring plan should therefore be thought of together: failure modes with a high RPN and a measurable physical symptom are the first candidates for monitoring. Do not change D without a real change in controls just to lower the score.

FAQ

What should the RPN threshold be?: There is no universal threshold. Values such as 100 are mentioned as examples, but the right threshold depends on your organisation's procedure, the scale and the risk. The threshold is a policy choice that decides which rows need action; here it is used only for flagging.
If two failure modes have the same RPN, which comes first?: RPN cannot decide it. The general approach is to put the one with higher severity first; at equal RPN the tool sorts the higher-severity row on top. The team should still assess how often the cause occurs and how likely detection is.
Why are severity 9 and 10 highlighted separately?: Effects of severity 9–10 usually concern safety or legal compliance. Even when occurrence or detection difficulty looks low, the RPN may stay small; these rows should be handled without looking at RPN. The option keeps them visible in every case.
How should I assign the scores?: Take the scale definitions (what 1 and 10 mean) from your own FMEA guideline and have the whole team use the same scale. This tool provides no scale definitions and does not suggest scores.
How do I bring my table in from Excel?: Select the failure mode, S, O, D columns in Excel, copy and paste them into the "Paste / import CSV" field (tab-separated text is read). You can take the result out with "Download CSV".

Turn your FMEA into a monitoring plan

Let's look at which of your high-priority failure modes give a measurable symptom and where monitoring could start.

Request a conversation Explore the predictive maintenance service →