Gauge R&R in Simple Terms
If a gauge gives different readings for the same part, you cannot trust any decision based on it. Gauge R&R splits measurement variation into two parts:
- Repeatability (equipment variation): the same person measuring the same part several times with the same gauge.
- Reproducibility (appraiser variation): different people measuring the same parts.
It is one part of Measurement System Analysis, described in the AIAG MSA reference manual.
How It Works
A common study uses 10 parts, 3 appraisers and 2 or 3 trials each, with parts measured in random order. Results are analysed with the average and range method or ANOVA. AIAG guidance for percent GRR:
| % GRR | General guidance |
|---|---|
| Under 10 percent | Generally acceptable |
| 10 to 30 percent | May be acceptable depending on the application, cost and risk |
| Over 30 percent | Generally unacceptable; improve the measurement system |
The number of distinct categories (ndc) should be 5 or more.
Example
A study on a bore gauge gives 24 percent GRR, with most variation coming from reproducibility. The team finds that operators position the gauge differently. A fixture and a short training session bring GRR down to 8 percent.
Why It Matters
A poor gauge can reject good parts, accept bad ones and make Cp and Cpk results meaningless. Gauge R&R is expected for gauges in the control plan under IATF 16949. For variable data use Gauge R&R; for pass/fail checks use attribute agreement analysis. OperaHive calibration management software manages Gauge R&R and MSA records alongside calibration.