The Evaluate the Measurement Process Study (EMP Study) in Minitab
Capability Statistics Unpacked
If you’ve done or plan to do process improvement work, then you’ve probably asked yourself whether you’re really getting precise enough measurements. You must trust your data before using it to make critical decisions about adjustments and resource allocation. Measurement systems analysis (MSA) broadly refers to procedures that estimate and evaluate the amount of variation in a measurement system. One type of is an EMP study, also known as Wheeler’s Method. EMP stands for Evaluate Measurement Process.
If you’ve done or plan to do process improvement work, then you’ve probably asked yourself whether you’re really getting precise enough measurements. You must trust your data before using it to make critical decisions about adjustments and resource allocation. Measurement systems analysis (MSA) broadly refers to procedures that estimate and evaluate the amount of variation in a measurement system. One type of is an EMP study, also known as Wheeler’s Method. EMP stands for Evaluate Measurement Process. The EMP study evaluates two sources of measurement variation:
Repeatability: The variation that is observed when the same operator measures the same part many times, using the same gage, under the same conditions.
Reproducibility: The variation that is observed when different operators measure the same part many times, using the same gage, under the same conditions.
Based on the repeatability and reproducibility, the EMP study classifies measurement systems ranging from the best rating of First Class to the worst rating of Fourth Class. In practical terms, these classes explain how well the measurement system detects a shift in the process mean of at least 3 standard deviations. If the measurement system can detect shifts like these, then the measurement system should be useful for other process improvement activities. For example, many control charts use a subgroup mean more than 3 standard deviations away from the overall mean as a signal that a special cause affected the process.
For example, a manufacturer of consumer foods monitors fill weights of cereal boxes. The manufacturer wants to make sure that the variation from different measurements is small enough that they can use other process improvement analyses. The results from an EMP study help to determine if the measurement system is acceptable and how to improve the measurement system.
Is the measurement system acceptable?
The EMP statistics provide the measurement system with its classification. In these results, the classification is First Class. The team can be confident that the measurement system will be good enough to use for other process improvement activities.
How can the measurement system improve?
The EMP study also includes information that you can use to decide when to prioritize improvements to the measurement system. The Analysis of Mean Ranges (ANOMR) and the Analysis of Main Effects (ANOME) show where the reproducibility is low relative to the process variation. In this ANOMR, operator B is less consistent than the other two operators. Improving the consistency for operator B will improve the measurement system.
In this example ANOME, different operators tend to measure higher or lower than each other. Improvements that bring the mean measurements of different operators closer will improve the measurement system.
Trust Your Data
To act on your data, you need to trust that the data are right. The EMP study in Minitab Statistical Software gives you the power to see whether your measurement system is acceptable and how to improve your measurement systems. When you assess the precision of your measurements, you can have confidence that everything that follows is built from data that you can trust.
Original blog post by Cody Steele for Minitab