Gage R&R

The Gage R&R is one of the main tools used in the measure phase of a Lean Six Sigma project.

The Gage R&R is a very underrated tool. Many people seem to dismiss this step and sometimes skip the measure phase altogether as they do not feel their measurement system needs any work. This may be the case, but unless there is no possibility of error in the measurement system, it should always be verified.

Before you go ahead and create experiments and analyze any data, you want to make sure that the data is measured properly and that you can actually trust the data. This is the tool you need to use to test the capability of your measurement system.

It is important to note that this tool is used for variable measurements (time, distance, length, weight, temperature, etc.) and not attribute measurements (category, error type, ranking, etc.). To test the capability of an attribute measurement system, you need to perform a Kappa Study.

The Gage R&R tests two main characteristics of a measurement system - its REPEATABILITY, and its REPRODUCIBILITY. Yes, that is what the R&R stands for.

What is repeatability? When we are doing a measurement system analysis, we want to find out how accurately a measurer can repeat their measurement. The measurer is usually a person, but sometimes it could also be a machine or gage. Basically, the question we are asking here is - If I measure the height of a product today, and I come back and measure the same piece of product next week…will I get the same result?

For most people, that is a strange question. Most people will say - "Of course you will get the same result!" This is not always the case; especially when you start getting down to millimeters and further. If I am using a measuring tape to measure the height of a product, and the reading I get is 100 mm. It is very possible that I may measure it again and get a reading of 101 mm due to measurement error. No, the product hasn't changed. It is most likely that you did something slightly different when measuring this time. Perhaps you are measuring under a different light. Or your eyes are measuring from a different height. There could be a number of factors.

What is reproducibility? Reproducibility looks at how well a measurer can reproduce a measurement performed by another measurer. Again, the measurer here is usually a person, but could be a machine or gage. Basically, the question we are asking here is - If Tony measured the height of this product, will I get the same result as he did when I measure the same piece of product?

Again, there could be multiple factors for why the measurements each person took may come out a little different. The goal is to have as little measurement error as possible and to be able to repeat and reproduce measurements accurately every time.

The Gage R&R is set up like an experiment. Samples are randomly chosen for multiple operators to measure. Each operator will also measure each sample randomly multiple times. The results of each measurement are then run through a Gage R&R analysis (very easily done through statistical software like Minitab or Sigma XL). This then tells us how good our repeatability and reproducibility are.

So what if my measurement is a millimeter off? What's the big deal? The Gage R&R will tell you whether it is a big deal or not.

This is determined by comparing the measurement variation to two things. Firstly, it compares the measurement error to the tolerance in the specification. This is called the P/T ratio (or P to T ratio). It addresses what percent of the tolerance is taken up by measurement error.

For example, if your tolerance is +/- 1 mm and your measurement variation takes up most of that tolerance, then you need to find ways to improve your measurement system. Otherwise, you may be rejecting good product or accepting bad product because of a bad measurement system.

On the other hand if the tolerance on the product is +/- 30 mm and your measurement variation is just a small part of that, your measurement system should be fine.

The second thing that the measurement variation is compared to is the variation in the product itself. This is called the %R&R. This addresses what percent of the total variation is taken up by measurement error.

Similar to the P/T ratio, you want your measurement variation be small compared to the product variation. The lower the P/T ratio and the %R&R, the better. As a rule of thumb, if you have a P/T ratio lower than 30% and a %R&R lower than 28%, you can consider the measurement system to be usable.

During the measure phase of your project, if you find that the measurement system used to measure your inputs or outputs is not up to the mark through the Gage R&R, you will have to investigate the causes for the measurement variation and eliminate them. You will then have to perform the Gage R&R again to see if the system has improved. Once your Gage R&R metrics look good, you can start using the measurement system to collect data for analysis.

Leave "Gage R&R" and go back to "Lean Six Sigma Tools"