Meaning
Statistical tool used to assess the precision and consistency of a measurement system. Grr study evaluates the amount of variation in a measurement process that stems from the equipment and the operators. It decomposes the total observed variance into repeatability, which is the equipment error, and reproducibility, which is the human or setup error.
Measurement Variation
Repeatability represents the variation seen when the same technician measures the same part multiple times with the same gauge. A grr study quantifies this internal equipment noise to ensure the tester is stable. High repeatability error suggests the fixture is loose or the sensor is drifting.
During
Reproducibility captures the difference in the average of the measurements made by different people using the same equipment. During a grr study, multiple testers perform the same sequence on a set of known good and known bad boards. Measurable differences between operators indicate a need for better training or more automation in the alignment process.
Acceptance Threshold
Results are typically expressed as a percentage of the total process tolerance or total variation. A grr study result below ten percent indicates that the measurement system is acceptable for the intended application. Values between ten and thirty percent might be acceptable depending on the application and the cost of the device.
If the value exceeds thirty percent, the measurement system is considered unreliable and requires redesign before production can resume. Redesign follows a failed study.