SC QUALITY GUILDGage R&R Calculator
SCQUALITYGUILD.COM/CALCULATORS/GAGE-R-AND-R.HTML
Free gage repeatability and reproducibility calculator. Enter appraiser and part measurements, get %GRR and the number of distinct categories, with what each result actually means.
AVERAGE AND RANGE METHOD · %EV, %AV, %R&R, ndc · FREE · NOTHING YOU TYPE LEAVES YOUR BROWSER
Q-BotClick for an explanation of this calculator.Gage R&R
AVERAGE AND RANGE METHOD · %EV, %AV, %R&R, ndcReading %R&R as a pass mark. The bands are a convention, not a rule: a system at 25 percent may be fine for a loose characteristic and useless for a tight one. Look at ndc and at which component dominates before deciding anything.
For educational use. This page is a study aid. The method behind it has far more depth and detail than one page and one tool can cover, and what you see here is not everything the method offers or everything this site offers. For full understanding and application, use the governing standards and texts, and seek further education, training or qualified advice. Results depend on the data you enter, so check them before relying on them.
What a gage R&R study measures
A gage R&R study has several appraisers measure several parts several times each, then apportions the total observed variation between the measurement system and genuine part-to-part differences. The question it answers is not "is the gauge accurate" but "how much of what I am seeing is the gauge rather than the parts".
Repeatability is the same appraiser getting a different answer twice on the same part. Reproducibility is two appraisers getting different answers on the same part. The first is usually the instrument or the fixturing; the second is usually the method, or the absence of one.
Reading the %GRR result
%GRR is measurement-system variation expressed as a percentage of total variation (a ratio of standard deviations, so %EV, %AV and %PV do not add to 100). The commonly cited bands are under 10% acceptable, 10% to 30% conditional depending on the criticality of the characteristic and the cost of improvement, and over 30% not acceptable until the measurement system is improved.
Those bands are guidance, not a standard. A 25% result on a safety characteristic is worse than a 25% result on a cosmetic one, and the decision belongs to somebody who knows which it is.
Number of distinct categories
The ndc states how many separate groupings the system can reliably tell apart across the range of parts measured. Five or more is the usual expectation. Below that, the measurement is sorting parts into fewer buckets than the resolution of the display suggests.
A gauge reading to four decimal places with an ndc of two is telling you "big or small" while appearing to tell you rather more than that.
Where these studies go wrong
Parts chosen from a narrow range. %GRR is a ratio, and if the parts barely differ from each other the denominator collapses and the measurement system looks far worse than it is. The parts should span the range the process actually produces.
Appraisers who know which part they are measuring. If the parts are numbered and visible, you are measuring memory as much as method.
Studying a system nobody uses that way. If production measures once and the study measures three times with a fixture that is not normally there, the result describes a process that does not exist.