Calculate GRR percentage from repeatability and reproducibility components. Evaluate measurement system adequacy for quality decisions.
Measurement System Analysis (MSA) quantifies the variation contributed by the measurement system itself. Before you can trust process capability studies or SPC charts, you must verify that your measurement system is adequate. If measurement variation is too large relative to process variation or tolerance, you cannot make reliable quality decisions.
The GRR (Gage Repeatability and Reproducibility) study is the most common MSA method. Repeatability measures variation when the same operator measures the same part multiple times. Reproducibility measures variation between different operators measuring the same parts. The total GRR combines both into a single percentage of total observed variation or tolerance.
This calculator takes repeatability and reproducibility standard deviations along with total variation or tolerance to compute %GRR. It classifies the measurement system as acceptable, marginal, or unacceptable per AIAG MSA standards.
This analytical approach aligns with lean manufacturing principles by replacing waste-generating guesswork with efficient, fact-based processes that directly support value creation and cost reduction.
Without MSA, you cannot trust your data. A measurement system contributing 30% of observed variation makes Cpk studies meaningless. MSA ensures that the variation you see in data is real process variation, not measurement noise. It is a prerequisite for meaningful SPC and capability analysis. Having accurate figures readily available streamlines reporting, audit preparation, and strategic planning discussions with management and key stakeholders across the business.
GRR = √(Repeatability² + Reproducibility²) %GRR (vs total variation) = (GRR / σ_total) × 100 %GRR (vs tolerance) = (GRR × 6) / Tolerance × 100 ndc = 1.41 × (σ_part / GRR) — number of distinct categories
Result: 28.3% GRR
GRR = √(0.015² + 0.008²) = √(0.000225 + 0.000064) = √0.000289 = 0.017. %GRR = 0.017 / 0.06 × 100 = 28.3%. This is in the marginal zone — acceptable only with caveats.
Every quality metric — capability indices, control charts, defect rates — relies on measurement data. If the measurement system adds significant noise, these metrics are inflated and unreliable. MSA is not optional; it is the foundation upon which all data-driven quality decisions rest.
When %GRR is too high, decompose it into repeatability and reproducibility components. If reproducibility dominates, standardize procedures and train operators. If repeatability dominates, the gage itself is the problem — improve resolution, fixturing, or replace the instrument.
Attribute measurement systems (pass/fail gages, visual inspection) also require MSA. Attribute Agreement Analysis evaluates whether operators agree with each other and with a known standard. This is critical for go/no-go gages, visual inspection, and manual sorting operations.
Per AIAG MSA guidelines: below 10% is acceptable, 10–30% is marginal (may be acceptable depending on application), above 30% is unacceptable. For critical characteristics, target below 10%.
Repeatability is within-operator variation (same person, same gage, same part, multiple trials). Reproducibility is between-operator variation (different people measuring the same parts). Together they form the GRR.
Ndc estimates how many groups the measurement system can reliably distinguish within the process variation. Ndc ≥ 5 is required for process control decisions. Ndc < 2 means the gage cannot distinguish parts at all.
If the measurement system is used for process control (SPC), evaluate against total variation. If used for product acceptance (pass/fail against specs), evaluate against tolerance. Report both when possible.
Conduct MSA at gage qualification, after calibration or repair, when operators change, and at least annually for critical gages. Many quality standards require MSA as part of control plan validation.
Improve the measurement system: upgrade to higher resolution gage, improve fixturing, standardize measurement procedure, train operators, or change to automated measurement. Re-run the study after improvements to verify.