Measurement System Analysis in Six Sigma Black Belt Training
MSA — Measurement system analysis in Six Sigma Training is the statistical tool used to validate the measurement system. It helps to check the variation that is caused in the process due to measurement. Variation exists in all processes and the variation could be due to the true process or due to the measurement issue.
Total Variation = True process variation + Measurement Variation
Total Measurement Variation = Appraiser Variation + Equipment Variation
Measurement system analysis is one of the critical tests to be performed prior to any other statistical tests. It enables to identify and eliminate the variation caused due to the measurement issue. If the variation is only due to the measurement system, the objective of the project will be met without proceeding with further analysis or tools implementation. That is the beauty of MSA Measurement System Analysis.
Key components in Measurement System Analysis:
- Instrument used for measurement
- Appraiser or the operator, one who performs the measurement
- Process for undertaking measurement
- Environment and the time factor
Measurement tools
Broad categories of the tools used for conducting measurement is Mechanical, Pneumatic, Electronic, Light technologies and Electron systems
Few Examples:
Caliper, Micrometer, Spring dividers, telescope gage, height gage, sine bar, precision protractor, Ring gage etc..
Common Issues with Measurement System Analysis:
Precision — Is the possibility of getting same or similar repetitive result or the proximity between test results or individual measurement results

Accuracy — Indicates the presence of the result around target consistently, could also be determined by how close is the agreement between a measurement result and the true value

Measurement System Analysis in Six Sigma Training
Resolution — Ability to measure to the closest decimal point or magnitude, lesser the measurement unit the better it is.

Resolution Measurement System Analysis in Six Sigma Training
Stability — Variation in the measurement over a period of time, indicates the change in bias of the measurement system

Stability in Measurement System Analysis
Linearity — Bias in range of measurements, there will be constant bias in the measurement system that has good linearity irrelevant of the magnitude of the measurement

Linearity in Measurement System Analysis
Repeatability — Variation in the measurement results when test is performed by same operator on the same part, also represented as variation within appraisers
Reproducibility — Variation in the measurement results when test is performed by different operators on the same part, also represented as variation between appraisers

Reproducibility in Measurement System analysis in Six Sigma Black Belt training
Measurement System Analysis Tests:

Different kind of Measurement System Analysis in Six Sigma Black Belt Training
Attribute Agreement Analysis:
Conditions — Repeatability, Reproducibility and Accuracy, all three needs to be greater than 90% for the Measurement System Analysis to be successful
Anova — Gage R&R:
Conditions –
- No of distinct categories should be greater than 5
- % Contribution variation of total gage R&R should be less than part to part variation
- % Study variation of total gage R&R should be less than 10%, 10–30% is marginally accepted and above 30% is rejected
source from : http://www.niqcgroup.com/measurement-system-analysis/