Have you heard all the buzz surrounding Measurement System Analysis (MSA)? Probably not. This methodology doesn’t get as much praise as it should compared to all the other Statistical Process Control (SPC) and Six Sigma methodologies. But if you want your Six Sigma and SPC projects to be effective, you’ll need to learn about it and master it.
Why MSA Matters
MSA can be used to determine how much variation in a measurement process affects or factors into process variability. This mathematical method is often used to find out if a given measurement system process is accurate. Because process measurement systems can have significant implications for important business decisions, it is critical that these systems be reliable. Otherwise, you risk your Six Sigma and/or SPC projects falling flat.
What MSA Investigates
There are 5 core parameters investigated during MSA:
With bias, you are investigating the difference between the true value of a given part and the value that has been observed. Linearity measures bias through the range of test values of interest. To calculate linearity, you must find the difference in the observed bias values through the range of measurement. Stability looks at how the accuracy and precision of the system (more on that below) operate over a given period of time. This is especially important as MSA looks at three categories of measurement system errors: accuracy, precision, and of course, stability.
Next, we have repeatability and reproducibility. Repeatability can be defined as the variation that occurs in a value when one item is measured by one person multiple times. In this case, everything, from the person, to the time, to the conditions, are the same. It is only the number of times the item is measured that changes.
Reproducibility is defined as the systematic variation that occurs in a value when two or more people are measuring the same item. Here they may be measuring the item at different times and in different ways.
While there is a lot of attention given to calibration (an important component of measurement system reliability), too little attention is given to measurement repeatability and reproducibility (Gauge R&R). R&R error, also referred to as precision, is the statistical sum of repeatability and reproducibility error. A test instrument may be perfectly calibrated but due to poor repeatability and reproducibility is useless and unreliable. The error rate for R&R error can be many times the calibration error.
As you work to determine the reliability and accuracy of your measurement system, make sure to rely on the MSA method for optimal results. Don’t miss out on NWCPE’s SS 504 Measurement System Analysis course. Register today.