Measurement System Analysis may employ any of several metrics to assess the suitability of a measurement system. These include:
(1) Repeatability,
(2) Reproducability,
(3) Bias,
(4) Linearity, and
(5) Stability.
The combination of (1) and (2) is commonly referred to as Gauge R&R (GR&R).
Measurement System Analysis is a component of Metrology and Measurement Uncertainty Analysis, which includes Uncertainty Budgets.
Values that are calculated during an MSA analysis, such as Repeatability and Reproducibility, are test statistics that have both a point estimate and an interval estimate (based on uncertainty due to sampling) because they are estimations of populations parameters based on sample statistics. As such, they have both point and interval estimates. The focus in most textbooks and in most practical application is on the point estimate.
The point estimates are what are usually presented in most textbooks on the topic.
Each component also has an interval estimate based on the standard error of the sample statistic. This is covered thoroughly in Montgomery's MSA book.
References and resources include:
Montgomery section 12-2 (page 519) for an example of GR&R based on a random effects ANOVA model using variance components.
Hicks section 6.9 for a similar example of GR&R based on variance components.
Searle text for more on the mathematics behind Variance Components models.
METHODS:
Other methods:
NOTE - this section is primarily a holding area for prototype techniques for non-standard MSA / GR&R situations. As a method is developed, it should be given its own web page in this section.
See article attached below "ASQ Alternate Method (using PCA) Can Validate Measurement Systems.pdf". Also available from ASQ.