What MSA/GR&R targets are acceptable for aerospace parts?

For aerospace CNC parts, a GR&R result below 10% is normally the preferred target for critical CTQ features; 10–20% may be accepted only with documented risk justification, and above 20% normally requires measurement-system improvement before inspection data supports release. These targets are not a substitute for the drawing, customer specification, or control plan. AS9100 expects controlled and traceable measurement processes, while common MSA practice gives the statistical language for judging repeatability and reproducibility. Buyers should ask which feature was studied, which gauge was used, how many operators and trials were included, and how the result affects part acceptance.

Understanding MSA and GR&R

MSA evaluates the complete measurement system, including the gauge, fixture, operator method, software setting, temperature condition, part handling, and inspection sequence. GR&R is the part of MSA that separates repeatability error from reproducibility error. Repeatability shows whether the same operator and gauge can repeat the reading. Reproducibility shows whether different operators or stations obtain comparable readings. In precision CNC machining, a weak measurement system can hide true machining variation or create a false nonconformance. The failure mode is simple: the shop may adjust a stable process because the gauge is noisy, or release a drifting process because the inspection method cannot see the drift.

GR&R Acceptance Criteria for Aerospace Applications

The practical GR&R decision range should be read as a risk filter. Less than 10% is suitable for critical dimensions such as close-fit bores, bearing seats, sealing diameters, and datum features that drive assembly. A 10–20% result needs written justification, usually for a less critical feature, a wider tolerance, or a feature where the control plan uses extra confirmation. A result above 20% means the measurement system may consume too much of the tolerance band. Corrective action may include a better fixture, clearer operator method, gauge repair, probe strategy change, or environmental control. The same rule applies across CNC machining processes, CNC grinding services, and multi-axis machining when inspection evidence is used to release tight functional features.

Key Influencing Factors

Several conditions decide whether a GR&R target is realistic. The part must be stable during inspection, the datum setup must match the drawing, and the gauge resolution should be appropriate for the tolerance being judged. Superalloys such as Inconel 718 and Rene 80 can require temperature control because heat from machining or handling changes measured size. Hard coatings like PVD coatings or heat-treated surfaces can change probe contact, burr behavior, and edge definition. High-performance alloys such as titanium Ti-6Al-4V or stainless steel SUS316L need stable clamping during measurement. Reflective aluminum 7075 surfaces may need optical settings or contact inspection confirmed with traceable calibration records.

Integration with PDCA and SPC

MSA and GR&R should feed directly into the PDCA quality cycle and SPC (Statistical Process Control) system. During Plan, the supplier selects the gauge, feature, operator group, sample parts, and acceptance rule. During Do, the study is run under the same handling and environmental conditions used for production inspection. During Check, GR&R results are compared with tolerance risk, control-chart behavior, and inspection history. During Act, the supplier updates the fixture, work instruction, inspection program, training record, or sampling frequency. This connection matters because SPC charts are only useful when the measurement system can detect the variation being charted.

Industry Expectations

High-risk buyers in aerospace and aviation, power generation, and nuclear sectors usually expect measurement evidence that can survive audit review. A low GR&R result does not guarantee part reliability by itself. It shows that inspection error is small enough to support the release decision for the studied feature. Buyers should separate three questions: whether the part was machined within tolerance, whether the gauge can prove that result, and whether the supplier has a reaction plan when the measurement system trends worse.

For an RFQ or supplier audit, request the GR&R study report, feature drawing callout, gauge type, calibration status, number of operators, number of parts, number of trials, study date, environmental condition, and acceptance rule. Also ask whether the study used production parts or representative samples. The strongest answer connects the GR&R target to the control plan, first-article inspection, SPC charting, and nonconformance containment. If the report omits tolerance width, datum setup, fixture repeatability, or part temperature, the number is hard to compare across suppliers. If the report was run on a convenient feature instead of the critical feature, request a new study before approval. The review should also state who can approve a marginal result and when the study must be repeated after gauge repair, fixture change, inspection-room relocation, or a new inspection shift. Ask for the planned repeat interval during production. If a supplier quotes a tight tolerance but cannot explain measurement-system capability, the buyer should treat the inspection plan as an open risk before pilot-lot, ramp-up, or production approval.