FMEA / RPN Calculator

What this calculator computes

This tool is a working FMEA worksheet. You enter each potential failure mode with its three ratings, Severity, Occurrence, and Detection on the standard one-to-ten scales, and it computes two things for every row: the traditional Risk Priority Number, the product of the three ratings, and the modern Action Priority, a High, Medium, or Low level based on the severity-weighted logic introduced by the AIAG-VDA FMEA handbook. It then ranks all the failure modes, puts the most urgent at the top, counts how many fall into each priority level, and plots the risks so the worst offenders are obvious at a glance.

The point of FMEA is to spend limited engineering effort where it matters, and that requires a defensible way to compare failure modes. The Risk Priority Number was the long-standing answer, a single number from one to a thousand that rolls severity, frequency, and detectability into one score. But a single product hides important differences, so the newer Action Priority keeps the three dimensions distinct and lets severity dominate, ensuring a dangerous failure is never buried beneath a moderate RPN. This calculator shows both, because seeing them together is the clearest way to prioritise well and to understand why the two can disagree.

What sets this calculator apart is that it is a multi-row worksheet, not a single-score gadget. Most free RPN tools compute one number for one failure mode; this one takes your whole list, ranks it by Action Priority and then RPN, highlights the single top item to tackle first, and shows the spread of High, Medium, and Low across your analysis. It colours each row by its action level, plots a Pareto-style chart of the biggest risks, and exports the ranked worksheet, all in your browser with nothing stored.

How to use this calculator, step by step

Start by listing your failure modes, one per row. For each, give it a short name and rate the three dimensions. Severity is how bad the effect is, from one for no real effect to ten for a hazardous, safety- or regulation-affecting failure. Occurrence is how often the cause is expected, from one for extremely unlikely to ten for almost certain. Detection is how well your current controls catch the problem before it escapes, and it runs the opposite way to intuition: one means detection is almost certain and ten means it will almost surely be missed, so better controls earn a lower number.

Add as many rows as you need with the add-row button, and remove any with the small delete control. The calculator recomputes live as you type, filling in each row’s RPN and colour-coded Action Priority, and updating the ranking and the chart. It opens with a worked example of three failure modes, a bearing seizure, a seal leak, and a sensor drift, so you can see a complete, ranked worksheet immediately and replace the rows with your own. Keep every rating an integer from one to ten; values outside that range are ignored.

Read the result from the top. The summary line names the single highest-priority failure mode, its RPN, and its Action Priority, and gives the count of High, Medium, and Low items across your worksheet. The table shows each row’s RPN and action level, colour-coded so High items stand out in red, Medium in amber, and Low in green. The chart ranks the biggest risks so you can see the Pareto shape of where the risk concentrates, and you can download the ranked worksheet as a PDF or CSV. Everything runs locally.

Severity, Occurrence, and Detection in depth

The three ratings are the heart of FMEA, and using them well starts with understanding what each measures and how the scales run. Severity rates the consequence of the failure effect, judged from the point of view of the customer or the end result, not the cost to fix.

A one means the effect is not even noticeable; the middle of the scale covers effects that degrade performance or annoy the customer; and the top, nine and ten, is reserved for failures that affect safe operation or violate a regulation, with ten being a failure that occurs without warning.

Because severity reflects the nature of the effect, it usually cannot be reduced without a design change that removes or mitigates the effect itself, which is why it carries the most weight in prioritisation.

Occurrence rates how frequently the cause of the failure mode is expected to arise, over the life of the product or the run of the process. A one means the cause is eliminated through prevention or is so unlikely as to be practically impossible; the scale climbs through occasional and moderate frequencies to a ten, where failures from that cause are almost inevitable. Occurrence is the dimension most directly attacked by reliability improvement and process control, because reducing how often the cause happens lowers the frequency of the failure at its root, which is generally preferable to catching it after the fact.

Detection is the dimension most often misunderstood, because its scale runs opposite to the others. It rates the ability of the current controls, tests, inspections, alarms, monitoring, to detect the failure mode or its cause before the product or process moves on. Here a low number is good and a high number is bad: a one means the control will almost certainly catch the problem, while a ten means there is no control or it is effectively useless and the failure will escape. It is worth stating this explicitly because the natural instinct is to score a strong control high, which inverts the risk; in FMEA, better detection means a lower detection rating and therefore a lower RPN.

RPN versus Action Priority: why the change

For decades the Risk Priority Number was the single output of an FMEA, and it did useful work: multiplying severity, occurrence, and detection gave one comparable number, and teams set a threshold above which they acted. But the multiplication has a structural flaw that the reliability and quality community came to see as serious enough to change the standard.

Because the RPN collapses three very different dimensions into one product, many different combinations yield the same number, and those combinations are not equally dangerous.

A failure rated severity two, occurrence ten, detection ten and a failure rated severity ten, occurrence ten, detection two both produce an RPN of two hundred, yet the first is a frequent, hard-to-catch nuisance and the second is a frequent potential safety hazard that is at least usually caught. Treating them as equal risks is exactly the wrong conclusion.

The 2019 AIAG-VDA FMEA handbook, which harmonised the American and German automotive FMEA standards, responded by replacing the RPN as the primary result with Action Priority. Rather than a product, Action Priority is a logic that considers the three ratings in order of importance, severity first, then occurrence, then detection, and assigns each failure mode to High, Medium, or Low.

The design guarantees that a high severity cannot be diluted by favourable occurrence or detection scores: a severity of nine or ten pushes the priority up regardless, so a genuine safety concern is never allowed to slip below an action threshold just because its RPN happens to be moderate.

High means risk reduction is needed, Medium means it is recommended, and Low means the current controls are likely adequate but should be justified.

This calculator deliberately shows both numbers rather than choosing for you. The RPN remains useful for ranking within a priority level and for tracking whether your improvement actions have actually lowered risk over time, and many organisations still record it. The Action Priority is the better guide to what must be acted on. Seeing them side by side also teaches the lesson directly: when a row has a modest RPN but a High action priority, you are looking at a high-severity failure that the RPN would have under-ranked, and that contrast is the whole reason the standard changed.

Five worked examples you can follow

Example 1: the default worksheet

The calculator opens with three failure modes. The bearing seizure, rated severity eight, occurrence four, detection six, has an RPN of 192 and a High action priority, so it tops the list. The seal leak, severity five, occurrence six, detection five, gives an RPN of 150 and a Medium priority. The sensor drift, severity three, occurrence five, detection eight, gives an RPN of 120 but only a Low priority, because its severity is low. The ranking puts the bearing first even though all three RPNs are close, which is the point.

Example 2: two failure modes with the same RPN

Enter one row at severity two, occurrence ten, detection ten, and another at severity ten, occurrence ten, detection two. Both show an RPN of two hundred, but the first comes out Low or Medium while the second is High, because severity dominates the Action Priority. This is the textbook illustration of why RPN alone misleads and why the second failure, a frequent potential hazard, must be worked first despite the identical number.

Example 3: a high-severity, low-RPN item

Enter a failure at severity ten, occurrence two, detection two. Its RPN is only forty, well below any typical action threshold, yet its Action Priority is High because the effect is hazardous. Under a pure RPN rule this dangerous failure would be ignored; the calculator flags it, which is exactly the safety trap the Action Priority was designed to prevent.

Example 4: improving detection versus occurrence

Take a failure at severity seven, occurrence six, detection eight, an RPN of 336 and High. Lower the detection to three, simulating a better inspection, and the RPN drops to 126; now lower occurrence to two instead, simulating a fix to the cause, and the RPN drops to 112. Both help, but attacking the cause is usually preferable to relying on detection, and the calculator lets you compare the two moves instantly.

Example 5: building a ranked work list

Add half a dozen realistic failure modes with varied ratings and read the ranking. The High items cluster at the top, sorted by RPN within the level, and the summary counts show how much of your analysis is High versus Low. This ranked list, not any single score, is the practical output of an FMEA: a prioritised sequence of what to fix, in order.

Three expert tips for reliable results

The logic behind the results

The Risk Priority Number is a straightforward product: RPN equals Severity times Occurrence times Detection, each an integer from one to ten, so the RPN is an integer from one to one thousand. It has no units; it is an ordinal score whose only meaning is comparative, letting you rank failure modes and track whether a number falls after you act. The criticality this calculator also shows, Severity times Occurrence, is a related measure that ignores detectability and focuses on how often a serious effect actually occurs, which some analyses prefer because detection does not change the underlying failure rate.

The Action Priority is not a formula but a decision logic, and this calculator implements it as a transparent, severity-dominant rule that follows the AIAG-VDA principle.

Severity sets the base level: a severity of one is always Low, because an effect no one notices is not worth acting on; a severity of nine or ten is High for essentially all occurrence and detection combinations, because a hazardous effect must be addressed; and the moderate severities in between escalate toward High as occurrence and detection worsen.

Within each severity band, higher occurrence and higher detection ratings raise the level, and the rule is monotonic, meaning a worse rating on any dimension never lowers the priority. The result is a level that respects the ordering severity, then occurrence, then detection, which is exactly the intent of the handbook.

One honest caveat belongs here. The official AIAG-VDA Action Priority is a specific, proprietary lookup table covering all one thousand combinations, published in the handbook.

The Action Priority this calculator produces follows the same severity-dominant principle and reproduces its key behaviours, that severity leads, that a severity of one is always Low, that a high severity forces a High priority, and that the logic is monotonic, but it is presented as a transparent guide rather than a verbatim copy of the proprietary table.

For everyday prioritisation and teaching it behaves as expected; for a formal automotive submission that must match the handbook cell for cell, confirm the level against the official AIAG-VDA FMEA Handbook.

Where FMEA is used

FMEA is one of the most widely applied tools in quality and reliability engineering, and it appears wherever the cost of failure justifies systematic prevention.

In the automotive industry it is effectively mandatory, embedded in the quality-planning frameworks that suppliers must follow, which is why the AIAG-VDA handbook that defines the Action Priority carries such weight.

Aerospace, defence, and medical-device makers use it to satisfy safety and regulatory requirements, documenting that every credible failure mode has been considered and prioritised. In general manufacturing it guides process design and control planning, and in product development it shapes design decisions before anything is built.

Within reliability engineering, FMEA connects to the rest of this silo as the qualitative front end that identifies what can fail before the quantitative tools measure how often.

The failure modes it catalogues are the events whose rates the failure rate calculator and Weibull analysis quantify; the components it examines are the blocks whose arrangement the system reliability calculator combines; and a High-priority wear-out failure it flags is exactly the case where the preventive maintenance calculator finds a cost-optimal replacement interval.

FMEA tells you what to worry about; the other tools tell you how much.

The method also feeds continuous improvement and risk management beyond its origins. The ranked list it produces drives corrective-action programmes, directing engineering and maintenance resources to the highest-priority failures first, and the reassessment step, re-rating after action, provides a measurable record that risk has been reduced.

In maintenance specifically, FMEA underlies reliability-centred maintenance, where the analysis of failure modes and their consequences decides whether each component warrants preventive replacement, condition monitoring, or a run-to-failure strategy.

Return to the Maintenance and Reliability hub for the companion tools that quantify the failures this analysis prioritises.

Design FMEA, Process FMEA, and the FMEA process

FMEA comes in two main forms that share the same rating scheme but examine different things. A Design FMEA analyses a product design, asking how each component or function could fail and what effect that would have on the system and the customer; its occurrence reflects the likelihood of a design-related cause, and its detection reflects design verification and testing.

A Process FMEA analyses a manufacturing or assembly process, asking how each process step could produce a defect; its occurrence reflects how often the process fault happens, and its detection reflects in-process inspection and controls.

The same severity of a given effect usually carries across both, since the effect on the customer is the same however the failure arises, which is why teams often align severity ratings between the design and process analyses.

Conducting an FMEA is a team activity that follows a sequence: define the scope and break the system or process into elements, identify the potential failure modes of each element, then their effects and their causes, and rate severity, occurrence, and detection for each.

The ratings produce the RPN and Action Priority that this calculator computes, which drive the crucial next step: deciding and assigning actions for the high-priority items, then re-rating once those actions are complete to confirm the risk has fallen.

The modern handbook formalises this as a seven-step process, adding structured planning and results documentation around the familiar rating core, but the analytical heart remains the identification and prioritisation of failure modes that this worksheet supports.

Two habits separate an FMEA that improves a product from one that merely fills a form. The first is to treat the analysis as living: an FMEA created once and filed away loses value quickly, whereas one revisited when the design changes, when a new failure appears in the field, or when a process is modified keeps steering effort to the current highest risks. The second is to close the loop with real actions and honest re-rating.

A high-priority failure mode is not resolved when it is logged; it is resolved when a change reduces its severity, occurrence, or detection and the ratings are updated to reflect that change, ideally with evidence that the reduction is real rather than optimistic.

This worksheet is built for that rhythm: enter your modes, act on the High items, adjust the ratings, and watch the ranking and the High-Medium-Low counts shift as the risk profile genuinely improves, which is the record that turns an FMEA from a compliance document into a measured reduction of risk.

Reducing risk: the hierarchy of actions

Once the worksheet has ranked the failure modes, the real work is deciding what to do about the high-priority ones, and there is a clear hierarchy of how to reduce risk that mirrors the three ratings. The strongest move is to reduce severity, because severity reflects the consequence of the failure and is the dimension the Action Priority weights most.

Reducing severity almost always means a design change that removes the failure effect or makes it benign, for example designing a component so that its failure is safe rather than hazardous, or adding a fail-safe that changes a catastrophic outcome into a mild one.

Severity reductions are the hardest to achieve but the most valuable, because they lower the priority no matter how often the failure occurs or how well it is detected.

The next move is to reduce occurrence, attacking the cause so the failure happens less often. This is the province of reliability improvement and process control: stronger materials, better tolerances, error-proofing a process step, tighter supplier quality.

Reducing occurrence is generally preferable to improving detection, because a failure that never happens needs no catching, and it lowers the true failure rate rather than merely intercepting more of the failures that still occur.

The weakest of the three, though still worthwhile, is improving detection: adding or strengthening inspections, tests, alarms, or monitoring so that more failures are caught before they escape.

Detection improvements are appropriate when severity cannot be designed out and the cause cannot be eliminated, but relying on them alone leaves the failure happening at the same rate and simply hopes to catch it, which is why the standard treats detection as the least influential dimension.

Working the hierarchy in order, severity then occurrence then detection, extracts the most risk reduction for the effort, and re-rating after each action, which this worksheet makes easy, confirms that the priority has genuinely fallen rather than merely being reshuffled.

A short history of FMEA

FMEA began in the United States military in the mid-twentieth century, formalised in a procedure that asked engineers to anticipate how systems could fail and what the consequences would be, at a time when the growing complexity of aerospace and defence systems made after-the-fact fixes unacceptably risky.

It was taken up by the aerospace programmes of the 1960s, where the stakes of failure were extreme, and then spread into the automotive industry in the following decades as manufacturers sought systematic ways to build quality and safety into products before production.

The Risk Priority Number emerged in this period as a simple way to turn the three ratings into a single prioritising score, and it became deeply embedded in automotive quality-planning practice.

The method was codified over time by standards bodies and industry groups, with the American Automotive Industry Action Group and the German Verband der Automobilindustrie each publishing FMEA guidance that suppliers were expected to follow.

The two diverged enough to create friction for companies working across both markets, and in 2019 they published a harmonised AIAG-VDA FMEA handbook that unified the approach and, most consequentially, replaced the Risk Priority Number with the Action Priority as the primary output.

That change was the culmination of a long-running critique of the RPN’s flaws, and it reflected a maturing understanding that prioritising risk well requires respecting the distinct role of severity rather than blending everything into one product. Today FMEA remains one of the most widely practised risk tools in engineering, and this calculator reflects both its history and its present by showing the traditional RPN alongside the modern severity-dominant priority.

Common mistakes to avoid

A few errors recur when teams run an FMEA. Watch for them.

• Inverting the detection scale. A strong control earns a low detection rating, not a high one. Scoring good detection high inflates the RPN backwards and corrupts the whole ranking.
• Ignoring severity because the RPN is low. A high-severity failure is High priority even at a modest RPN; never let the product hide a hazardous effect.
• Chasing a single RPN threshold. Equal RPNs can mean very different risks; use Action Priority to decide what to act on and RPN only to rank within a level.
• Improving detection instead of the cause. Catching a failure more often is weaker than preventing it; attack occurrence first where the cause can be reduced.
• Rating alone. FMEA is a team method; ratings set by one person miss failure modes and drift in scale. Calibrate the scales across the team before rating.
• Stopping at the numbers. The RPN and Action Priority are the start, not the end; the value is in the actions taken on the high-priority items and the re-rating that confirms improvement.

Input format and quick reference

Enter one failure mode per row with a name and three integer ratings from 1 to 10. The calculator computes the RPN and Action Priority for each and ranks the worksheet. The reference below explains each output.

Frequently asked questions

Sources, disclaimer and editorial transparency

This calculator computes the Risk Priority Number as RPN = Severity × Occurrence × Detection (each 1–10, RPN 1–1000) and a criticality of Severity × Occurrence, and it assigns an Action Priority (High/Medium/Low) using a transparent severity-dominant, monotonic rule that follows the AIAG-VDA 2019 FMEA principle.

The Action Priority here is an educational guide, not a verbatim reproduction of the proprietary AIAG-VDA lookup table; for formal automotive submissions, confirm each level against the official AIAG-VDA FMEA Handbook.

This calculator and guide are created and reviewed by the OpsCalculators team; see our Editorial Policy for how each tool is researched, built, and tested.

Results are accurate for planning, prioritisation, and education, not certified safety or quality-engineering advice, and FMEA ratings are judgements that should be set by a qualified team. See our full Disclaimer. OpsCalculators.com is operated by MAFHH INTERNATIONAL LTD. Your data is processed in your browser and never stored; see our Privacy Policy.