Inspec Insights

Gage R&R How and Why


by James Lakin; Vice President, Sales

Gage Repeatability & Reproducibility (GR&R) quantifies the robustness of an entire dimensional measurement system.  It does not measure accuracy, accuracy is determined by the calibration of the measurement tools you are using.  A Gage R&R is not only used for gages, it’s used for: Instruments, People, Parts, Specifications and Methods.  It’s a combination of all these things and more, it quantifies the measurement precision.    

Performing a Gage R & R study is a systematic approach to improving in your measurement system and to ensure the quality of your (or your vendor’s) in-process measurements results.  Because of this, you need not worry, if your GR&R results are not acceptable on your first attempt.  You can make adjustments in your measurement system and repeat the GR&R to confirm the improvements. 

Let’s begin by defining the two R’s.  The first “R”, according to Wikipedia, Repeatability - is the variation in measurements taken by a single person or instrument on the same or replicate item and under the same conditions. 

The second “R” – is the variation induced when different operators, instruments, or laboratories measure the same or replicate specimen.   In other words, the first “R” is meant to represent tool/instrument variation, when the second “R” is meant to represent "People" \\\t variation. 

These two variations can be used to calculate the GR&R result using either one of two methods: X-bar & R-chart, which can calculated by hand or spreadsheet but is generally considered less accurate, or ANOVA, which will require specialized computer software, such as Minitab, but is considered more accurate.

Your first step in beginning a GR&R study is determining the size of the project.  There are trade-offs, naturally, between time/money and utility.  Larger data sets produce more accurate data, but may require more time and resources. 

You will want to begin by quantifying the number of parts (including the number of features to be measured on each part), the number of operators, as well as the number of measurements (replicates) for each operator and each part.  Typical project sizes are: 5 x 2 x 2 (20 data points) or 10 x 3 x 3 (90 data points). 

Next you will want to construct a physical or electronic data sheet to record and organize all the data.  Once the organizational foundation is complete, you can begin collecting data.  After you’ve collected all the data, you can begin calculating the GR&R results.   GR&R results are reported in percent, and are typically graded as: 0 – 20% is acceptable, >20% to 30% is marginal and anything over 30% is unacceptable.

If your GR&R results are acceptable, you’re done with this exercise.  If your GR&R result is anything but acceptable, you’ll want review the data to decide where the measurement weaknesses are and how they might be made more robust either by making the measurements more consistent or lowering the variation.  Then you’ll need to repeat the GR&R to verify the measurement system improvements.  Once the improvements are confirmed, they should be properly documented in your in-process inspection work instructions, so that all future measurements can be made properly.

Following this process will give you, your company and your customer confidence in your in-process measurement system and minimize waste in your production system due to unacceptable measurement variation.

Leave a comment