Annual State of FMEA
Anonymised aggregate of how teams build, score and revise FMEAs across industries — published yearly.
FMEA research, the way industry actually needs it
Qhubio is building an independent research programme around Failure Mode and Effects Analysis. The goal is simple: publish anonymised, methodologically transparent benchmarks that engineers, auditors and OEMs can actually use — not vendor marketing dressed up as statistics.
Qhubio Research
The research programme today
No statistics are published yet. The infrastructure, methodology and privacy guarantees are in place; data collection from opted-in organisations is the next phase.
8
Planned reports
0
Active reports
0
Published reports
8
Industries covered
Privacy principles
The Qhubio research programme is bound by four principles. They are non-negotiable — if a dataset cannot satisfy all four, it does not get published.
Opt-in only
No customer data enters any published dataset without an explicit opt-in by the organisation that owns it.
Aggregates, never rows
Reports publish distributions, medians and percentiles only. Free-text fields, part numbers and organisation identifiers are never published.
Suppression thresholds
Any cell with fewer than ten organisations or fewer than thirty FMEA rows is suppressed and replaced with a notice.
Independent methodology
Methodology is published before data. Customers can replicate the calculation on their own dataset.
Methodology
Every published dataset is derived from anonymised aggregates of opted-in FMEAs and audits. Severity, Occurrence, Detection and Action Priority are recomputed server-side using the deterministic AIAG-VDA logic that powers the rest of Qhubio, so reported distributions reflect a single consistent scoring method — not the rating drift that makes vendor-to-vendor benchmarks meaningless.
- Organisations opt in per project, with a clear data-use scope.
- Row text, part numbers, customer names and free-text fields are stripped before any aggregation pipeline runs.
- Failure modes, detection controls and findings are normalised against a controlled vocabulary, never the raw user input.
- Action Priority is recomputed from Severity, Occurrence and Detection using the AIAG-VDA 2019 lookup — guaranteeing cross-customer comparability.
- Cells below the suppression threshold (n < 10 organisations or n < 30 rows) are dropped and replaced with a notice.
- Methodology and any code paths used for the calculation are published alongside the report.
Benchmark categories
The research roadmap is split into eight benchmark categories. Each plugs into the same data pipeline and renders through the same components — so as soon as one dataset reaches publication threshold, the report goes live without any redesign.
Industry benchmarks — dataset not yet published
Data collection is in progress. Qhubio only publishes verified, anonymised aggregates — no statistics appear here until a dataset reaches publication threshold.
Planned reports
These reports are explicitly on the roadmap. Each card shows the dataset's current status. No card shows fake numbers — if a value isn't measured yet, it isn't shown.
Failure Mode Trends
Year-over-year shifts in the most common failure modes across process families.
Industry Action Priority Distribution
Share of Low / Medium / High Action Priority outcomes by industry under AIAG-VDA scoring.
Most Common Detection Controls
Catalogue of the detection controls most frequently used across manufacturing PFMEAs.
Manufacturing Risk Trends
Quarterly view of where manufacturing risk concentrates across machining, molding, assembly, welding and electronics.
Automotive FMEA Benchmark
Reference distributions for Severity, Occurrence, Detection and AP across IATF 16949 suppliers.
Medical Device FMEA Benchmark
ISO 14971-aligned FMEA reference distributions for medical-device manufacturers.
Supplier Audit Finding Trends
Aggregated supplier-audit findings reported through the Qhubio Audit module, normalised per finding type.
Trend reporting — placeholder
Trend charts will appear here once at least two reporting periods of verified data are available. Until then, the chart renders empty rather than illustrative — Qhubio does not draw fictional plot lines.
Failure-mode prevalence — automotive
Will track year-over-year shift in the ten most common automotive failure modes.
No published datapoints yet. Qhubio will plot this trend once at least two reporting periods of verified anonymised data are available.
Action Priority distribution — manufacturing
Will track quarterly Low/Medium/High AP shares across opted-in manufacturing cohorts.
No published datapoints yet. Qhubio will plot this trend once at least two reporting periods of verified anonymised data are available.
How organisations can participate
Participation is opt-in, per project, and reversible at any time. Qhubio customers will be able to enable anonymised data contribution from the project settings panel — and disable it with one click without losing any of their own data.
Opt-in tooling for organisations ships alongside the first published report. To register early interest, contact info@qhubio.com.
Why original research matters
The FMEA literature is saturated with explanations of the same method. What the industry lacks is comparable data: how automotive PFMEAs actually distribute Severity across processes, how often supplier audits cite missing process flow diagrams, how Action Priority outcomes shift after teams retire RPN. By publishing that data — anonymised, methodologically sound and free — Qhubio raises the floor of the entire FMEA discipline, not just its own software.
Help us improve Qhubio. Analytics help us understand which FMEA features are useful, which pages need improvement, and where users encounter problems. No advertising. No selling personal data. See our Cookie Policy.