Measurement system analysis (IATF only)
Plain-language summary
Knowing your gauge is calibrated is not enough — MSA asks whether the whole measurement SYSTEM (gauge + operator + method + environment) can actually distinguish good from bad on each control-plan measurement.
What the clause is really asking
Statistical studies (GR&R, bias, linearity, stability, attribute agreement) are required to analyse variation in the results of each measurement system named in the control plan. Methods and acceptance criteria should follow customer-referenced manuals (typically AIAG MSA) unless the customer approves alternatives.
What auditors look for
Auditors take the control plan and ask for the MSA study behind each measurement: current GR&R for variable gauges, attribute agreement for visual/go-no-go checks, study repeated after gauge change or repair. Results exceeding criteria with no action is a standard finding.
Typical evidence
GR&R studies; attribute agreement analyses; MSA planning linked to control plan; reaction records when studies fail criteria.
How to comply — recommendations
Plan MSA off the control plan: every measurement listed gets a study type and frequency. Re-study after repair, relocation or method change. When GR&R fails, fix the method or gauge — do not quietly widen the tolerance.
Common nonconformities
Control-plan measurements with no MSA; GR&R >30% accepted without action; visual inspections never given attribute studies; MSA not repeated after gauge replacement.
Related clauses
Builds on ISO 9001 7.1.5 (see Example row)
In the book
Chapter 7 — Support — People, Kit and the Truth About Documents treats this clause family in full: what it is really asking, what auditors look for, the evidence that satisfies, how to make it work on a real floor, and where companies stumble. This IATF requirement is covered in the Automotive Edition, in the same six blocks as the ISO clause around it.
Resources for this clause
- ArticleMSA and Gage R&R explained: type 1 studies, ANOVA vs average and range, %GRR vs %tolerance, ndc and attribute agreement
- ArticleControl plans explained: what goes in one, how it links to the PFMEA, and a free template
- ArticleCp vs Cpk vs Ppk: process capability without the confusion
- Free downloadMSA / Gage R&R Study TemplateProve the gauge before you trust the number — a crossed Gage R&R worked by ANOVA and by average and range side by side, plus a type 1 gauge study, all as live formulas.
- Free downloadCpk CalculatorPaste your measurements, get Cp, Cpk and predicted yield instantly — prove the process is capable.
- Free downloadControl Plan TemplateThe production rulebook as a working file — a blank control plan, a filled worked example, and a 12-point health check for the plans you already have.
- Toolkit · $29SPC Control Chart ToolkitPick the right tab, fill the yellow cells, and watch control limits and charts calculate live — no SPC maths, no macros, no setup.
- Free masterclass · ~18 minControl Charts MasterclassStatistical process control in plain English — common vs special cause, choosing and building the right chart (X̄-R, X̄-S, I-MR, p/np/c/u), reading the signals, and proving capability with Cp/Cpk.
Get the free MSA / Gage R&R Study Template
Free, instant download. We'll only email you the occasional practical quality update — no spam, unsubscribe anytime.
Qlause provides interpretive guidance only and is not a substitute for the standard. Refer to your licensed copy of the relevant standard for the authoritative text.