KaizumiKaizumi™

Gage R&R

Measurement System Analysis — Crossed Study

1/6

What is a Gage R&R Study?

Before you trust your data, you have to trust your measurement system. A Gage R&R study measures how much of the variation you see comes from the gage and the people using it, rather than from the parts themselves. R&R = Repeatability (same operator, same part, different result?) plus Reproducibility (different operators, different results?).

If the measurement system is bad, every chart, capability study, and improvement decision built on its data is bad too. Run the R&R before the improvement project, not after.
Study Info (optional)
Operators
Trials
Standard study: 5–10 parts spanning the process range, 2–3 operators, 2–3 trials each.
Measurements
PartOp A · T1Op A · T2Op B · T1Op B · T2
#1
#2
#3
#4
#5
R&R
Fill every cell of the measurement grid
Or load the example — a 5-part, 2-operator, 2-trial study

Want an editable copy?

Download the free Gage R&R template — the standard crossed-study layout with EV, AV, PV, %GRR, and ndc computed automatically from your measurements. Opens in Excel and Google Sheets.

Why measurement system analysis comes first

Every improvement decision rides on data, and every data point rides on a measurement. A Gage R&R study asks the uncomfortable question: if we measure the same part twice, do we get the same answer? When the answer is no, the "process variation" on your charts is partly an illusion manufactured by the gage — and chasing it wastes improvement effort on a measurement problem.

This is why MSA sits at the start of the Measure phase in a DMAIC project: before the control chart and before the capability study, you validate the instrument that feeds them both.

Repeatability vs. reproducibility

The split matters because the fixes differ. High repeatability error (EV) means the gage itself is inconsistent — worn fixtures, loose probes, resolution too coarse for the tolerance. High reproducibility error (AV) means different people use the gage differently — clamping pressure, reading angle, technique — and the fix is usually a better standard method and training, which is far cheaper than a new gage.

Frequently asked questions

What is a Gage R&R study?
A Gage R&R (repeatability and reproducibility) study quantifies how much of the variation in your data comes from the measurement system rather than the parts being measured. Repeatability is variation when the same operator measures the same part repeatedly — the gage itself. Reproducibility is variation between operators measuring the same parts — method and training. Together they are the measurement system's contribution, expressed as %GRR of total variation.
What is an acceptable %GRR?
By AIAG MSA guidance: under 10% of total variation is acceptable; 10–30% is conditionally acceptable depending on the importance of the characteristic and the cost of improving the gage; over 30% is unacceptable — the measurement system needs to be fixed before its data can be trusted for decisions.
What is ndc and why does it matter?
The number of distinct categories, ndc = 1.41 × PV ÷ GRR, is how many distinct groups of parts the measurement system can reliably distinguish across your process spread. AIAG guidance requires ndc ≥ 5 for the system to be useful in process analysis. A gage can look acceptable against a wide tolerance yet have ndc of 1–2 against actual process variation — meaning it cannot tell your parts apart at all.
How many parts, operators, and trials do I need?
The standard crossed study uses 10 parts, 3 operators, and 2–3 trials, but 5 parts × 2 operators × 2 trials is a workable minimum this tool supports. What matters most is that the parts span the true range of process variation (including borderline parts), that operators are the people who normally use the gage, and that measurements are taken in random order, blind to previous readings.
What is the difference between the average-and-range method and ANOVA?
The average-and-range method — used by this calculator — estimates the variance components from ranges with tabulated constants (K1, K2, K3), and is the classic AIAG approach that can be computed by hand on the shop floor. The ANOVA method partitions variance statistically and can additionally separate an operator-by-part interaction. ANOVA is slightly more precise; average-and-range is more transparent. For most studies they reach the same verdict.
Is this Gage R&R calculator free?
Yes. It runs in your browser with no sign-up. Enter your crossed-study measurements to get %GRR with a verdict, the EV/AV/PV breakdown showing whether the gage or the operators drive the error, ndc, and a print-ready MSA report. A free Excel template for recording the study is also available.

Related tools