MSA — Measurement System Analysis

MSA evaluates the measurement systems used to collect product and process data, quantifying how much of the observed variation comes from the measurement system itself versus the actual process. The key MSA study is Gauge R&R (Repeatability and Reproducibility).

Why companies use it

·Invalid measurements lead to false accept/reject decisions, hidden defects, and wrong process adjustments
·Required by IATF 16949 and PPAP before SPC data can be trusted
·Quantifies whether a measurement tool is fit for purpose before investing in process improvement
·Identifies whether measurement variation comes from the gauge (repeatability) or the operator (reproducibility)

What hiring managers look for

·Understanding MSA prevents the common engineering error of "improving" a measurement noise problem rather than a real process problem
·A %GR&R below 10% is typically required before SPC can be implemented — engineers must know this threshold
·Experience designing and running a Gauge R&R study is a practical skill in any manufacturing quality role
·Ability to interpret ndc (number of distinct categories) shows depth of knowledge

Typical interview questions

What is the difference between repeatability and reproducibility in a Gauge R&R study?

Your Gauge R&R study shows 28% measurement variation. What does this mean and what do you do?

When would you use an attribute MSA (Kappa study) instead of a variable Gauge R&R?

How many parts, operators, and replicates are needed for a standard Gauge R&R study and why?

Describe a situation where an MSA study revealed a measurement problem that was masking a real quality issue.

Common mistakes

·Using a measurement system with >30% GR&R for SPC — the control limits will be inflated and meaningless
·Running the Gauge R&R study on a narrow range of parts — you need parts that represent the full process spread
·Not blinding the operators to each other's measurements during the study, introducing bias
·Ignoring the appraiser variation (reproducibility) — if operators measure differently, training or clarity on measurement method is needed
·Confusing %GR&R to tolerance (used for inspection decisions) with %GR&R to study variation (used for process analysis)

Real engineering example

A medical device manufacturer found that their SPC chart for implant surface roughness showed excessive variation. An MSA study revealed that 34% of the total variation was from the measurement system — specifically, operators were positioning the stylus at different angles. After creating a jig to fix stylus position, %GR&R dropped to 8%, revealing that the actual process was already in control.

Topics covered

MSAGauge R&RrepeatabilityreproducibilitymeasurementqualityIATF 16949