Skip to content
Rektron

Search

    Explained · FMEA · Failure Mode and Effect Analysis

    FMEA: finding failure modes before they reach the customer

    Failure Mode and Effect Analysis (FMEA) is a structured method for finding, early in development, how a product or process can fail, what the effects would be and why it might happen. It rates the risks so that preventive actions go where they do most good.

    A failure that could have been foreseen

    A new bracket goes into production. After a few weeks, the customer finds brackets where a press-fit bush has worked loose. The investigation finds that the bore diameter depends on the temperature of the die, which nobody measures. Looking back, the team agrees: everything needed to foresee this was known before the start of production. Nobody had asked the question systematically.

    That systematic questioning is what an FMEA does.

    What an FMEA is

    Failure Mode and Effect Analysis is a team method for working through a design or a process and asking, element by element: how could this fail, what would happen if it did, and why might it happen? It is done during product and process development, when changes are still cheap.

    The main types are:

    • Design FMEA (DFMEA): failures of the product design, for example a part that is too weak or a tolerance that does not ensure function.
    • Process FMEA (PFMEA): failures in manufacturing and assembly, for example a wrong torque or a missing part.
    • FMEA for monitoring and system response (FMEA-MSR): introduced in the AIAG & VDA handbook for systems that detect faults in the field and react, such as vehicle diagnostics.

    The chain: failure mode, effect, cause

    At the heart of an FMEA is a chain of three elements:

    • Failure mode: how the product or process can fail. The bush is loose in the bore.
    • Effect: what the customer or the next step notices. The bracket rattles and wears; in the worst case the bush falls out.
    • Cause: why the failure occurs. Bore oversize due to die temperature.

    How does an FMEA work from failure to action?

    Illustration

    An FMEA describes what can go wrong, what effect it has and why it happens, and rates the risk so that actions go where they do most good.

    Step by step

    • Failure mode: how the product or process can fail.
    • Effect: what the customer or the next step notices. Rated as severity.
    • Cause: why the failure occurs. Rated as occurrence.
    • Action: prevent the cause or improve detection. Rated as detection.
    • Actions are followed up and the analysis is updated.

    Each element is rated on a scale from 1 to 10:

    • Severity (S) of the effect
    • Occurrence (O) of the cause, given the current prevention controls
    • Detection (D): how likely the current detection controls are to find the cause or the failure before the part leaves

    From ratings to priorities

    Traditionally, the three ratings were multiplied into a Risk Priority Number, RPN = S × O × D, and the highest RPNs were treated first. The problem is that very different risks can share the same number: a severity of 10 with low occurrence can have the same RPN as a minor nuisance that happens often.

    The AIAG & VDA FMEA Handbook (1st edition, 2019) replaces this with Action Priority (AP): a table that assigns each combination of S, O and D a priority of high, medium or low, giving most weight to severity, then occurrence, then detection.

    Example. For the loose bush: S = 8 (loss of function), O = 6 (the cause occurs from time to time), D = 7 (no measurement of bore diameter). RPN = 8 × 6 × 7 = 336. With high severity and moderate occurrence, Action Priority would also be high: action is required. The ratings are illustrative.

    The seven steps

    The AIAG & VDA handbook organises the work in seven steps:

    1. Planning and preparation
    2. Structure analysis
    3. Function analysis
    4. Failure analysis
    5. Risk analysis
    6. Optimisation
    7. Results documentation

    Steps 2 and 3 break the system into elements and describe what each element must do. Step 4 builds the failure chains, step 5 rates them, and step 6 defines and follows up the actions.

    From FMEA to production

    The analysis is only useful if it changes something. The preferred actions reduce occurrence, by changing the design or the process. Where that is not possible, detection is improved. In a PFMEA, the controls end up in the control plan: which characteristic is checked, how, with which gauge, how often, and what to do if it fails.

    How does a risk in the PFMEA become a check in production?

    Illustration · Example data

    The control plan turns the risks into concrete checks: which characteristic, with which method and gauge, how often, and what happens when the requirement is not met.

    Step by step

    • A risk in the PFMEA concerns the length.
    • The control plan specifies that the length is measured with a caliper on every part.
    • The reaction plan says what is done if the length is outside: quarantine and sort.

    In the example, the action could be to control die temperature and to measure the bore on a sample of parts each shift, plotted on a control chart. The gauge used for that check should itself be verified with a measurement system analysis. In this way the FMEA becomes part of quality assurance rather than a document filed for the customer.

    Common pitfalls

    • Done too late. An FMEA written after the design is frozen documents risks instead of reducing them.
    • Done alone. The method depends on a cross-functional team with design, production and quality knowledge.
    • Never updated. Complaints, scrap and process changes should flow back into the analysis. An FMEA is a living document.
    • Ratings negotiated down. Lowering a detection rating without a real new control only hides the risk.

    In short

    An FMEA asks what can go wrong before it does. It links failures to effects and causes, rates the risks, and turns the highest priorities into actions and controls that are actually used on the shop floor.

    Key takeaways

    • An FMEA links failure modes, effects and causes, and rates them for severity, occurrence and detection.
    • The AIAG & VDA handbook (2019) uses seven steps and Action Priority instead of ranking by the RPN alone.
    • The value is in the actions: an FMEA should lead to design changes, process controls and a control plan that is used in production.

    Where Rektron can help

    Rektron FMEA is a separate tool for DFMEA, PFMEA and control plans. It supports the AIAG & VDA methodology.

    Sources

    • AIAG (Automotive Industry Action Group) and VDA (Verband der Automobilindustrie), Failure Mode and Effects Analysis - FMEA Handbook, 1st edition, 2019
    • International Organization for Standardization (ISO), Quality management systems - Requirements (ISO 9001), 2015

    Written by: Rektron. Reviewed by Lana Donoga (provisionally approved 2026-10-21) 2026-10-21. Published 2026-10-21.