Glossary
FMEA (Failure Mode and Effects Analysis) is a systematic methodology for identifying potential failures in a product or process before they occur, ranking them by risk, and prioritising preventive action. The methodology was developed by the US military in the 1940s, refined by NASA in the 1960s, and is now standard across automotive, aerospace, medical devices, and increasingly apparel. This guide covers the two main FMEA types, the Risk Priority Number (RPN) calculation, and an apparel-sourcing example.
FMEA (Failure Mode and Effects Analysis) is a structured method for identifying ways in which a product, design, or process could fail; evaluating the consequences of those failures; and prioritising preventive action against the highest-risk failure modes. Unlike reactive methodologies (CAPA, 8D) that respond to problems that have occurred, FMEA is proactive: it anticipates failures during the design and planning stages and addresses them before they reach the customer.
The methodology originated in the US military in 1949 (Military Procedure MIL-P-1629) and was adopted and further refined by NASA during the Apollo programme in the 1960s. The Society of Automotive Engineers (SAE) and the Automotive Industry Action Group (AIAG) standardised the methodology for automotive use in the 1990s.
The current reference standard for automotive FMEA is the AIAG-VDA FMEA Handbook (2019), which harmonised the previously separate American and German automotive FMEA standards.
FMEA in apparel and textiles is less formalised than in automotive or aerospace but is becoming more common as brands face regulatory pressure to demonstrate proactive risk management (ESPR data, AGEC traceability) and as supply chains globalise across multiple Tier 1 and Tier 2 manufacturing partners.
FMEA has two primary variants, each suited to a different stage of the product or process lifecycle.
Applied during product design, before manufacturing begins. The analysis identifies how the product itself could fail under real-world conditions: material fatigue, structural weakness, interaction failures between components, durability under environmental stress. DFMEA is most valuable when product changes are still inexpensive (early design) and decreases in value as the design locks in.
Applied during process planning, before the first production batch runs. The analysis identifies how the manufacturing or assembly process could fail: out-of-specification operations, incorrect material handling, inadequate inspection points, missed quality checks. PFMEA is run for each manufacturing partner and revised when the process changes.
A complete FMEA programme typically runs DFMEA during product development and PFMEA before production launch, with both updated when the product or process is modified. Less commonly used variants include System FMEA (for complex multi-component systems) and Service FMEA (for service delivery processes).
The Risk Priority Number (RPN) is the FMEA's mechanism for ranking failure modes and prioritising action. It is calculated as the product of three scores:
RPN formula
RPN = Severity (S) × Occurrence (O) × Detection (D). Each factor is scored on a 1 to 10 scale.
RPN ranges from 1 to 1,000.
Each factor is scored on a 1 to 10 scale, with higher scores indicating worse outcomes. Severity rates the consequence of the failure if it reaches the customer; 1 is no impact, 10 is critical safety failure.
Occurrence rates the probability of the failure happening; 1 is virtually never, 10 is almost certain. Detection rates the probability of the failure being caught before reaching the customer; 1 is virtually certain detection, 10 is virtually impossible to detect.
Multiplying the three scores produces an RPN between 1 and 1,000. Failure modes with RPN above a threshold (typically 100 to 125, set by the organisation's risk tolerance) receive priority action.
The action may target any of the three factors: reducing the severity of the failure, reducing the occurrence rate, or improving the detection mechanism.
The AIAG-VDA 2019 FMEA Handbook replaced RPN with the Action Priority (AP) approach, which uses a structured matrix of Severity, Occurrence, and Detection rather than a multiplied score. Many organisations still use RPN because it is simpler and the existing FMEA training base is built on it.
The AP approach addresses some of RPN's known weaknesses (a high Severity score gets diluted if Occurrence is low, even though high-severity failures may warrant action regardless).
A brand is launching a new performance running jacket and conducting PFMEA on the manufacturing process with a new manufacturing partner in Vietnam.
One failure mode under analysis: incorrect placement of the brand label on the inside back yoke. The label is positioned by an operator using a marked guide; the failure mode is that the label is sewn at an angle or offset from the intended position.
Severity: 5. A misaligned label is a quality defect that triggers customer returns and damages brand perception, but does not affect product safety or function. Severity is moderate.
Occurrence: 6. The operation is manual and depends on operator attention; in similar operations at the partner's facility, the defect rate has been observed at 1.2 percent. Occurrence is moderate-to-high.
Detection: 4. The defect is visible at the partner's final inspection station and at the brand's incoming inspection. Detection is good but not certain (occasional misalignments below threshold escape both inspection points).
RPN = 5 × 6 × 4 = 120. The threshold for action is set at 100, so this failure mode is prioritised.
The action plan addresses occurrence: a fixed positioning jig is introduced at the labelling operation, eliminating the operator's marked-guide dependency. The post-action FMEA review estimates Occurrence drops from 6 to 2 with the jig in place.
New RPN: 5 × 2 × 4 = 40, below threshold.
This is one failure mode in a complete PFMEA that may evaluate 30 to 80 operations across the manufacturing process. The full FMEA produces a prioritised list of action items, scheduled into the production launch plan.
FMEA software automates the analysis workflow: structured failure-mode templates, RPN or AP calculations, action tracking, and revision management as the FMEA is updated through the product lifecycle. The category has matured significantly over the past decade.
For apparel and textiles, the FMEA-software selection runs alongside the broader quality and PLM stack. Brands typically choose one of three approaches: a dedicated FMEA platform, an FMEA module within a broader QMS, or an FMEA capability built into the PLM.
Each approach has trade-offs around depth of FMEA functionality, integration with adjacent systems, and total cost of ownership.
The selection criteria that matter for apparel: how the tool handles supplier-side FMEA participation (many process FMEAs are co-developed with the manufacturing partner), how the tool integrates with the operational systems that produce real-world failure data (PIM, QMS, returns analytics, rework coordination), and how the tool supports the AIAG-VDA Action Priority approach if the brand wants to move beyond classical RPN.
Growing search demand for FMEA software reflects the broader regulatory and supply-chain pressure on brands to formalise risk management. The trend is most visible in industries already running formal FMEA programmes; apparel is following several years behind automotive and medical devices.
Flexireo is not an FMEA platform. The platform's purpose is the operational coordination of rework, refurbishment, and disposition decisions for apparel and footwear.
FMEA is the design-and-process risk management discipline running upstream of the operational work.
What Flexireo contributes is the operational failure data that informs FMEA revision: when a particular defect type recurs across batches, when a particular manufacturing partner shows a higher defect rate for a specific operation, when a specific failure mode appears in real-world rework volumes that was not predicted in the original FMEA. The Flexireo record is the empirical evidence that closes the loop from PFMEA prediction to operational reality.
For brands running mature FMEA programmes, the Flexireo operational data feeds back into FMEA updates: occurrence scores get refined based on observed defect rates, detection scores get refined based on observed escape rates, and new failure modes get added when the operational data surfaces failures the original FMEA did not anticipate.
Book a 30-minute demo and we will show you how the operational data from your rework coordination feeds back into FMEA updates: observed defect rates by failure mode, observed escape rates by inspection point, new failure modes surfaced in real production. The data your FMEA needs to stay current is the data your rework operation produces.