Articles

Did you know that all data is prone to arbitrary variation? Did you also know that individual differences and uncertain measurements are the two main causes for variation? Gage R+R helps uncover the extent of variation in measurement systems and helps identify the primary source of the problem. It can also help you avoid costly errors in measurement that can start to add up unexpectedly. If you’re still learning or already familiar, our Gage R+R guide will be useful either way. It might even teach you a few things you didn’t know!

What is Gage R+R?

Standing for gage repeatability and reproducibility, Gage R+R works by measuring variables of production measuring processes. The aim is to determine whether the variation is excessive or not, before acting on the result. If the variation is extensive, then corrective action must be taken to fix the problem. Every production measuring process in use requires Gage R+R assessment, though not for each individual gage.

Who Should Do It? 

A group of three people who perform measurements in production typically carry out Gage R+R. Your team can consist of people on the production line, quality inspectors, or lab technicians. We recommend you select a good calibration technician to collect your data, someone familiar with the process and a source of information for you. Gage R+R software can be overwhelming if you have no relevant training, which is why it’s best left to your technician to use it.

Before You Begin

When performing measurements, your team should be familiar with Gage R+R measurement concepts. Team members should also employ preventative measures to ensure they never become aware of the parts they measure, which would invalidate your results. Numbering the parts (temporarily) in such a way that they can avoid knowing which number is measured is the best way to work, though any awareness at all can reduce results, even if you actively try to remain unaware.

Read About Step-by-Step Guide to Gage R&R

Setting Up and Collecting Your Data

First, you need an appropriate sample size of 10 parts, with three operatives, and three tests, which should total 90 individual measurements. If necessary, a smaller sample size can also be used, such as when you are running short on staff or have a limited number of trials. To begin the process, the data-collector presents the parts in a random sequence, recording each measurement according to its (temporary) part number. GAGEtrak software then organizes the data before selecting a calculation method.

Calculation and Evaluation

There are several methods by which to calculate Gage R+R results, though the most common method, known as average and range, is by far the most effective. Also known as long AIAG, average and range makes assumptions based on the equation: appraiser x part interaction = zero. If the equation turns out to be incorrect, this proves your calculations unreliable. When evaluating your results, it is best to do so asking the question: do the parts meet tolerance? If they do, then you can assume relative reproducibility and repeatability for your production parts. Evaluating your results will allow you to select alternate gages for specific purposes too. All so you can reduce process variation wherever you find it!