MSA in DMAIC: Gage R&R and the Measure tollgate
Before you can trust your baseline data, you must verify that the measurement system itself is capable. This is one of the first tasks in the Measure phase. A flawed gauge can make a good process look bad, mask a real improvement, or send you down the wrong root-cause path.
For detailed Gage R&R math, see the CQE page on Gage R&R →. This page focuses on the DMAIC framing and exam-tested decision rules.
MSA components (CSSBB exam scope)
| Component | What it measures | Typical method |
|---|---|---|
| Repeatability (EV) | Variation when one operator measures the same part repeatedly with the same gauge (equipment variation) | Crossed Gage R&R study |
| Reproducibility (AV) | Variation between operators’ averages when measuring the same parts with the same gauge (appraiser variation) | Crossed Gage R&R study |
| Gauge R&R (GRR) | Total measurement system variation: GRR = √(EV² + AV²) (variances add, not standard deviations) | ANOVA or average/range method |
| Part-to-part (PV) | Actual process variation captured by the study | Same study |
| %Study Var | GRR as a percentage of total study variation | %GRR = (GRR / TV) × 100, where TV = √(GRR² + PV²) |
| Bias | Systematic offset of measurements from true value | Bias study: repeated measurements of a reference part with a known value |
| Linearity | Bias variation across the measurement range | Multiple reference standards |
| Stability | Gauge drift over time | Control chart of a master or reference part measured over time |
Acceptance criteria (%GRR, AIAG MSA 4th ed.)
| %GRR result | Decision |
|---|---|
| < 10% | Measurement system is acceptable |
| 10–30% | May be acceptable depending on the importance of the application, cost of the gauge, cost of repair and similar factors; improve if possible |
| > 30% | Measurement system is unacceptable; must improve before collecting baseline data |
Number of Distinct Categories (NDC)
NDC = 1.41 × (PV / GRR), truncated to a whole number. The NDC tells you how many distinct groups the gauge can reliably distinguish.
- NDC ≥ 5: recommended by AIAG; measurement system is adequate for process improvement work
- NDC = 2 to 4: only coarse distinctions; NDC = 2 is no better than go/no-go (attribute) data
- NDC = 1: the gauge cannot distinguish one part from another — not useful for understanding variation
What happens if MSA fails?
If the measurement system fails, you cannot proceed to the Measure tollgate. You must:
- Identify the dominant source of variation (repeatability vs. reproducibility).
- Fix it: recalibrate the gauge, train operators, standardize measurement procedure, or replace the gauge.
- Re-run the MSA study before collecting baseline data.
Attribute agreement analysis
For go/no-go and visual inspections, use attribute agreement analysis with kappa (Cohen's kappa for two raters or rater vs. standard, Fleiss' kappa for three or more). See Attribute agreement & kappa → for the math. The DMAIC framing: kappa above 0.75 indicates good to excellent agreement and below 0.40 indicates poor agreement (AIAG MSA 4th ed.); above 0.75 is generally acceptable for classifying parts into pass/fail categories used in a baseline defect rate.
Try 5 free CSSBB practice questions on MSA and the Measure tollgate.
Take the free CSSBB diagnosticMore: DMAIC overview · Project charter · VOC & CTQ · SIPOC · MSA in DMAIC · Process capability · Data collection planning · Graphical analysis · Root cause analysis · Multi-vari analysis · Hypothesis testing · Regression · DOE · FMEA · Define tollgate · Change management · Financial benefits · Team dynamics · Improve tollgate · Piloting & scaling · Poka-yoke (BB) · Solution selection · Control plan · SPC in DMAIC · Sustaining gains
CSSBB Quest by Bizarre Act. Original study material; not affiliated with ASQ. CSSBB is a certification of ASQ. Terms · Privacy