Let’s look at Measurement System Analysis, and how it fits into continuous process improvements for a business. Since measurements and the collection of data are important to continuous improvement programs, we should have an interest in Measurement System Analysis.
We use measurements to guide our decision on improvements and should be interested in the errors there are in measurements so we can make better decisions. If there are errors in our measurements, there could be errors in our decision. So let’s look now at Measurement System Analysis as it attempts to show the quality of measurements and how accurate and precise they are.
What it is
Measurement System Analysis is a method of looking at variation in a measurement process. Some groups such as the Automotive Industry Action Group (AIAG.org) provide guidelines on what percent error is acceptable, such as less than 10 percent. 10 to 30 percent error could be too high for some applications. Over 30 percent error could be unacceptable for all applications.
When you complete a Measurement System Analysis, you perform an experiment to show measurement variation. You should make sure you understand the data you collect could have errors. Measurement System Analysis can help you figure out how many errors.
Measurement System Analysis is a component of Six Sigma approaches that attempt to reduce or eliminate defects and errors.
How the analysis can help
When you complete such an analysis you can better select what measurement you want to use when you go through a process improvement.
Imagine you work with a group that is trying to increase the number of usable parts they make in a day. A team determines to collect data on the variables they feel could help produce more parts. No matter what data they collect, they cannot find a link of variable to result. A Measurement System Analysis could show that the data you are collecting includes measurements that have errors. With this example you can see how a Measurement System Analysis could be helpful.Register For a Course Near Me
Some variations to consider
Accuracy is a key quality in data to consider. Bias is another word used to describe accuracy. As an example, if we took a scale that is off by 1 pound for every 100 pounds measured, for a 1000-pound measurement, the error would be 10 pounds. If you are measuring something that weighs even more the error would be proportional and possibly out of the range where it could be of use to you in a continuous improvement process.
Stability is another key quality of data to consider. If you are taking measurements over time you could see numbers that do not make sense. For example, you could be taking a measurement of weight that varies and does not make sense. Consider that the calibration of your scale could be drifting. You may need to calibrate more often.
This is a simplified way of explaining what variations could impact the quality of the data and measurements you collect.
A manufacturing industry tool
Concerning a Six Sigma improvement process, a standard tool is the Gage R&R study. R&R refers to repeatability and reproducibility. The tool uses an Analysis of Variance (ANOVA) and is sometimes called an ANOVA Gage R&R study. While the tool applies to gauges it can apply to other measurement instruments too.
Here are variables to consider where you could have variation and use a Gage R&R study.
1. The gauge itself
2. The people performing measurements
3. The method of recording
Repeatability is variation where one person takes a measurement and you compare that with another person taking the same measurement. Reproducibility is variation with different people, instruments, or locations taking the same measurements.
Gage R&R looks at the precision of a measurement system. It is a tool often used in Six Sigma projects.
Contact us at 6Sigma.us and find out how we can positively change the culture and operations of your manufacturing unit. We offer Green Belt and Black Belt training programs, as well as a Master Black Belt program.