Measurement System Analysis (Gage R&R) Calculator

Variances add, standard deviations do not, so every combination here is a root-sum-square. Percent of study variation and percent of tolerance answer different questions and frequently disagree: the first depends on how representative the specimens were, so a study run on deliberately similar samples flatters the gauge and one run on extremes flatters it further. Percent of tolerance is the criterion that matters when the gauge is used to accept or reject against a specification, and it is independent of the specimens chosen. The ndc figure uses the AIAG 1.41 constant and should be at least 5; below 2 the gauge cannot distinguish parts at all and is only counting. None of this detects bias — a systematically wrong instrument can be perfectly repeatable and perfectly reproducible. Calibrate against a traceable standard as a separate exercise.