FMEA Severity Rating Scale: 1-10 Definitions for Manufacturing and Assembly Failures

Every FMEA session eventually hits the same debate: “Is that a severity 7 or an 8?” The argument stalls the meeting, frustrates the cross-functional team, and rarely gets resolved with confidence. The root cause is almost always the same—the team is working from vague definitions instead of anchored, manufacturing-specific criteria for the FMEA severity rating scale.

This guide defines each severity level (1–10) with concrete manufacturing and assembly failure examples drawn from the AIAG-VDA FMEA Handbook methodology. Use it as a calibration reference during your next FMEA review session.

What the FMEA Severity Rating Scale Measures

Severity rates the worst-case consequence of a failure mode on the end user, downstream process, or regulatory compliance. It answers one question: if this failure occurs and reaches the customer, how bad is it?

Key rules that practitioners sometimes forget:

Severity is rated on the most serious effect of the failure mode. If a failure causes both a cosmetic defect (S=4) and a safety hazard (S=9), severity is 9.
Severity cannot be reduced by adding controls. Only a design change can lower the severity rating. Detection and prevention controls affect Occurrence and Detection ratings, not Severity.
Ratings of 9 or 10 automatically designate the associated characteristic as a Critical Characteristic (CC), triggering mandatory safety/regulatory actions per IATF 16949.

Tip If your team frequently debates whether a severity is 7 or 8, your rating criteria may be too generic. The table below anchors each level to specific manufacturing consequences—share it before the session starts, not during the argument.

FMEA Severity Rating Scale: 1–10 Definitions for Manufacturing

Rating	Effect Category	Manufacturing/Assembly Example	Customer Impact
1	No discernible effect	Slight variation in surface finish within spec; no functional or cosmetic impact	Customer cannot detect any difference
2	Very minor effect	Minor appearance blemish visible only under close inspection (e.g., faint tool mark on non-visible surface)	Perceptive customers may notice; no performance effect
3	Minor effect	Cosmetic imperfection on a visible surface (e.g., slight color variation, minor paint orange peel)	Customer notices and is mildly dissatisfied; no functional degradation
4	Low effect	Fit or finish issue requiring minor rework at customer end (e.g., gasket slightly misaligned but sealable)	Customer inconvenienced; performance not significantly affected
5	Moderate effect	Reduced product life or performance degradation (e.g., bearing noise develops after 50% of expected service life)	Customer dissatisfied; reduced confidence in product quality
6	Significant effect	Product or subsystem inoperable for a non-essential function (e.g., interior light fails in a vehicle; HVAC fan runs only on high speed)	Customer experiences loss of comfort or convenience feature
7	Major effect	Product operable but at significantly reduced performance (e.g., engine power derated, transmission shifts harshly)	Customer very dissatisfied; likely warranty claim or return
8	Extreme effect	Product or subsystem inoperable for a primary function (e.g., vehicle will not start, braking assist fails but manual braking still functional)	Customer unable to use product for intended purpose; high warranty/recall risk
9	Hazardous with warning	Safety-related failure with advance warning (e.g., steering becomes progressively difficult before full loss; brake warning light illuminates before failure)	Potential injury or regulatory non-compliance; customer may have time to react
10	Hazardous without warning	Safety-related failure with no warning (e.g., sudden tire blowout from defective bead seat; airbag deploys unexpectedly)	Potential serious injury or death; no opportunity for customer to react

Common Mistake Teams often cluster ratings around 5–7 to avoid triggering critical characteristic requirements. This “severity compression” defeats the purpose of FMEA. If a failure could cause injury, it is a 9 or 10—period. The AIAG-VDA handbook is explicit: severity 9–10 triggers Critical Characteristic designation regardless of occurrence or detection.

How Severity Connects to Action Priority

Under the AIAG-VDA Action Priority (AP) system, severity carries more weight than occurrence or detection. This is a deliberate correction of the older RPN calculation method, where multiplying S × O × D could mask safety-critical failures behind low-occurrence scores.

The AP lookup logic works like this:

Severity 9–10: Action Priority is always High (H) regardless of O and D values. This is the core improvement over RPN.
Severity 5–8: AP depends on the combination of O and D. High occurrence with low detection = High AP; low occurrence with high detection = Medium or Low AP.
Severity 1–4: AP is typically Low (L) unless both occurrence and detection are also elevated.

You can test this yourself with the RPN & Action Priority calculator—enter a severity of 10 with occurrence of 1 and detection of 1. Under RPN, that combination gives a score of just 10 out of 1,000. Under Action Priority, it is classified as High, requiring mandatory action. That single comparison illustrates why the AIAG-VDA handbook replaced RPN with Action Priority.

Worked Example: Rating Severity for an Injection Molding PFMEA

Consider a plastic connector housing produced by injection molding. The cross-functional team is conducting a process FMEA and needs to assign severity ratings to three failure modes:

Worked Example

Failure Mode 1: Short shot (incomplete cavity fill)

Effect: Connector housing is structurally weak; retention clips may break during assembly at the customer plant.

Severity assessment: The customer cannot assemble the part, making the product inoperable for its primary function. Severity = 8 (product/subsystem inoperable—loss of primary function).

Failure Mode 2: Flash on parting line

Effect: Excess material on the sealing surface prevents proper mating with the housing cover. Assembly is possible but seal integrity is compromised, potentially allowing moisture ingress over time.

Severity assessment: Product is operable at reduced performance—the seal works initially but degrades. Severity = 7 (major effect—significantly reduced performance).

Failure Mode 3: Contamination in material (metal particles in resin)

Effect: In the final application (automotive electrical connector), metal contamination could cause a short circuit in the wiring harness, leading to an electrical fire with no warning.

Severity assessment: Safety hazard without warning. Severity = 10 (hazardous without warning—potential serious injury). This automatically designates the material purity characteristic as a Critical Characteristic (CC).

Notice the logic: the team rates based on the end effect at the customer level, not the defect itself. A short shot is a process defect, but the severity is about what happens when that defect reaches the customer assembly line or the end user.

Calibrating Your Team: Three Anchoring Techniques

Severity ratings are only useful when applied consistently. Here are three calibration techniques used by experienced FMEA facilitators:

1. Anchor-and-adjust

Start with clear anchors: “We all agree a severity 10 means potential injury without warning. We agree a severity 1 means no one can detect any difference.” Then work inward from both ends for ambiguous cases.

2. Effect-first rating

Describe the effect before showing the severity table. Ask each team member to independently write down a rating, then compare. Discuss divergences—this surfaces hidden assumptions about what “reduced performance” means for a specific product.

3. Historical benchmarking

Reference severity ratings from past FMEAs on similar products within your organization. If a cracked housing was rated severity 8 on the previous program, the same failure mode on a similar housing should receive the same rating unless the application context changed.

Tip Before your next FMEA session, distribute the severity scale definitions (the table above) to all team members at least 24 hours in advance. Teams that pre-read the criteria spend less time debating ratings and more time identifying failure modes—which is where FMEA actually adds value.

When to Challenge a Severity Rating

Severity should be stable once assigned—but these situations justify re-evaluation:

Design change: A new material or geometry eliminates a failure effect entirely (e.g., switching from a brittle polymer to a ductile one may reduce fracture severity from 8 to 5).
Application change: The same part is used in a different end product with different safety implications (e.g., a connector housing used in a seat heater vs. a braking system).
New regulatory requirement: A regulation classifies a previously non-critical attribute as safety-relevant (e.g., updated flammability standards for interior materials per FMVSS 302).
Field data: Warranty returns or customer complaints reveal a more severe consequence than originally anticipated.

Severity ratings should not change because the team added a detection control. A 100% inspection plan does not make a potential safety hazard less severe—it makes it more detectable.

Key Takeaways

The FMEA severity rating scale runs 1–10, from no discernible effect to hazardous without warning.
Rate based on the worst-case end effect at the customer level, not the process defect itself.
Severity 9–10 always triggers Critical Characteristic designation and High Action Priority under the AIAG-VDA system.
Severity cannot be reduced by adding controls—only by design changes.
Distribute rating criteria before the session, not during the debate.