FMEA Ranking Tables (AIAG-VDA 2019)

Severity captures customer impact. Occurrence captures how often the cause is expected. Detection captures the ability of current controls to catch the failure. AP is determined by a fixed rule table; the three ratings are not multiplied.

• Pick anchor examples from your own historical data for each rating.
• Do not redefine the AIAG-VDA wording — customers will reject non-standard scales in audit.
• Keep Severity essentially fixed across reviews; it reflects customer impact, not your improvement actions.
• Recalibrate Occurrence whenever ppm data or process capability changes meaningfully.

For any FMEA performed under IATF 16949 or to an automotive customer's Customer-Specific Requirements (CSRs), the AIAG-VDA 1–10 scales are the expected baseline. Most OEMs (Ford, GM, Stellantis, VW Group, Renault-Nissan, Toyota) reference them directly or through their own derivative tables. Deviating from the structure — collapsing to 1–5, renumbering, or rewording rating anchors — is a routine source of audit findings and PPAP rejections.

Outside automotive (medical, aerospace, consumer electronics) the AIAG-VDA scales are not mandated, but most organizations adopt them because the structure is well understood and software, training and audit checklists are built around it. FMEA Standard covers which standard applies in which industry.

AIAG-VDA explicitly allows organizations to add company- or product-specific examples to the rating anchors, provided the underlying definition is unchanged. Customization is allowed; redefinition is not.

1. Pick 2–3 anchor examples per rating from your own historical data — scrap reports, warranty data, PPAP.
2. Publish the anchored tables as a controlled document; reference the document ID in every FMEA header.
3. Re-anchor Occurrence whenever ppm data or process capability changes meaningfully.
4. Train new team members on the anchored table before they participate in their first workshop.
5. Audit cross-team consistency twice a year — pick 5 random rows from different FMEAs and re-rate as a group.

The AIAG-VDA scale definitions themselves change only when the handbook is revised (last major change: 2019). Your company-specific anchors should be reviewed:

• Annually as part of the management review.
• Whenever a new product family or process technology is added.
• Whenever a customer escape forces a rating recalibration.
• When ppm performance or Cpk for a process family changes by more than one rating band.

• The anchored table is a controlled document — version, owner and revision date are checked.
• Random FMEA rows are compared to the anchored examples; mismatches become findings.
• Detection ratings below 5 are cross-checked against Control Plan reaction plans.
• Occurrence ratings are compared against the latest ppm or Cpk data.
• Severity values are compared against the customer impact actually documented in warranty records.

• Maintaining a separate "customer table" and "internal table" that drift apart.
• Letting individual teams write their own anchors with no central control.
• Using ppm anchors that have not been updated for years.
• Treating Severity as adjustable — Severity reflects customer impact and rarely changes (see Severity Rating Guide).
• Forgetting that the three scales feed Action Priority, not RPN — see RPN vs AP for the migration logic.

A modern FMEA software tool should enforce the 1–10 structure, store your anchored examples, surface them in the rating cell during workshops, and apply the AP lookup automatically. Spreadsheet templates leave all of this to the user, which is why anchors drift across teams. The Qhubio FMEA generator uses the AIAG-VDA scales as a fixed reference and binds every generated rating to the appropriate anchor.

• Severity Rating Guide
• Occurrence Rating Guide
• Detection Rating Guide
• Action Priority Guide
• RPN vs Action Priority
• AIAG-VDA 7-Step Process
• FMEA Audit Checklist