Locked lesson.

Upgrade

About this lesson

Gage R&R studies can be done with automated test equipment to be certain that they are providing accurate and precise results. These are known as Type 1 Gage studies. These studies can be used with any type of automated checking or testing system, not just manufacturing systems. A satisfactory Type 1 Study is often needed to convince stakeholders to have confidence in automation.

Exercise files

Download this lesson’s related exercise files.

Type 1 Gage Studies .xlsx
10.6 KB Type 1 Gage Studies - Solution.xlsx
136 KB

Quick reference

Type 1 Gage Studies

Type 1 Gage Studies are an analysis of the test or inspection measuring equipment.  Its purpose is to determine if the equipment provides results that are accurate and repeatable.  The analysis is simple, but very powerful for ensuring the measurement equipment is capable of performing measurements that can be trusted.

When to use

This analysis is used when the test or inspection measuring equipment is complex or is being newly introduced.  The analysis does not consider either operator variation or part to part variation, so its focus is strictly on equipment variation.  It is often used as a preliminary analysis before a Gage R&R study is conducted.

Instructions

This analytical technique is simple but powerful.  It follows a pattern that is similar to the process capability indices of Cp and Cpk.  Like those ratios the numerator is based upon the part tolerance and the denominator is based upon the standard deviation of the measurement. In addition, the target value for both ratios is similar to one often used with process capability and that is 1.33 or greater.

To conduct this analysis, one operator uses one part and measures that part a statistically significant number of times, I recommend at least 40 times.  Each of those measurements are recorded.  The data set of the measurements is then analysed to determine the mean value and standard deviation.

The Cg ratio is focused on repeatability.  It compares the variation based upon the standard deviation with a percentage of the tolerance band that is allowed for the feature being measured.  Normally the percentage of the tolerance being used is 20%.  And the denominator is six times the standard deviation – essentially form minus three sigma to plus three sigma.

The Cgk ratio builds on the Cg ration but includes a bias effect.  It includes an effect for how close the measured value is to the true value.  A different independent and highly accurate measuring system must be used to determine the true value.  In this calculation, we first break the Cg calculation in half to only consider the side of the tolerance where the bias is occurring.   Then the numerator is reduced by magnitude of the difference between the mean of the measured values and the true value.  This is subtracted from half the tolerance percentage and then again divided by half of six times the standard deviation – or three standard deviations.  If the equipment is “spot on” providing a mean that is exactly the true value, then Cg and Cgk will be equal.  However, if there is any difference, then Cgk will be less than Cg.  Never-the-less the goal for both ratios is to exceed 1.33.

Minitab can also do the analysis.  First conduct the measurements and place them in a column in Minitab.  Then the study can be initiated by selecting “Stats”, “Quality Tools”, “Gage Study”, and finally “Type 1 Gage Study.”

Minitab brings up a form where you select the data column, enter the true value and the tolerance spread.  Then the calculation can be performed.  If for some reason you want to change the default tolerance percentage from 20% or the denominator number of standard deviations, you can make those changes in the Options window.

Minitab will calculate Cg and Cgk plus provide a Run Chart that shows the data values as they occur within the allowable tolerance spread so that you can visualize how the equipment is performing in addition to calculating the effect.

Hints & tips

  • If there is a special setup required to do the measurement, do not take the equipment apart and set it up for each measurement, that should be done the Gage R&R study.  In this case, leave the equipment setup and just keep measuring the same part again and again.
  • If you get the same answer each time, that is OK, we would hope that would be the case because then there is high repeatability. 
  • You may have to send your part out to be measured by some other measurement system to determine the true value. 
  • This study can often be done very quickly, so take advantage of the insight and use it.
Login to download
  • 00:01 Hi, I'm Ray Sheen.
  • 00:06 Now, there's another type of MSA study that we should discuss.
  • 00:09 This is called a Type 1 Gage Study.
  • 00:12 And it's a critical study for ensuring that the equipment in your tests and
  • 00:16 inspection are capable of giving reliable results.
  • 00:21 >> So let's review this analysis and when we would use it.
  • 00:25 This study is focused on the equipment.
  • 00:27 It doesn't look into variation due to operators or parts.
  • 00:31 We want to ensure that the basic equipment variation is a very small component
  • 00:36 of the variation that we will measure.
  • 00:38 Ideally, only a small fraction of the allowable tolerance variation.
  • 00:43 A typical time when you would use this study was prior to conducting
  • 00:47 a full Gage R&R study, especially one that uses complex equipment.
  • 00:53 This study will make sure that the equipment does not introduce a bias effect
  • 00:57 into the measured value and that there is a high degree of repeatability.
  • 01:02 In essence, we want to be sure that when the measuring equipment is being used by
  • 01:06 the same person to measure the same part repeatedly,
  • 01:10 the measurement value is consistent and correct.
  • 01:14 In addition to a preparatory step to a Gage R&R study, I strongly recommend using
  • 01:19 this type of study whenever introducing a new piece of test equipment or
  • 01:23 a brand new test method with an existing piece of test or inspection equipment.
  • 01:28 And since we are only concerned about the equipment variation,
  • 01:31 this study is very simple to execute, one person measures the same part many times.
  • 01:37 And by many, I would say you should do at least 40 times
  • 01:41 to ensure that the standard deviation has very little uncertainty.
  • 01:45 This type of measurement study relies on two ratios, Cg and Cgk.
  • 01:52 If you've already studied the process capability, you will see that these
  • 01:57 ratios follow the same pattern is that to choose with Cp and Cpk ratios.
  • 02:02 Our goal for both ratios is to be greater than or equal to 1.33.
  • 02:07 And I recommend that you do the Cg calculation first, because if you
  • 02:12 can't reach the value of Cg, you certainly can't get there with Cgk.
  • 02:16 The Cg ratio focuses on repeatability, it will be a ratio of the percentage
  • 02:21 of the tolerance associated with the feature being measured,
  • 02:26 divided by a multiple of the standard deviation of the measured value.
  • 02:31 In particular, we normally use 20% of the tolerance in the numerator.
  • 02:36 And just like with the Cp ratio,
  • 02:38 we use 6 times the standard deviation in the denominator.
  • 02:43 Cgk ratio is a bias measure, start with the ratio similar to Cg.
  • 02:48 But now, we need to adjust for the case when the measurement system
  • 02:53 is recording everything as too big or too small.
  • 02:56 So we will split the ratio into 2, dividing both the numerator and
  • 03:00 the denominator of the Cg ratio by 2.
  • 03:03 Then adjust the numerator value by subtracting the bias in the system.
  • 03:08 This is calculated by taking the absolute value of the difference between
  • 03:13 the average of the measured value and the true value of what is being measured.
  • 03:18 So if the equipment has no bias error, Cg and Cgk values will be the same.
  • 03:24 But if there is a bias error, the numerator for
  • 03:27 the Cgk will be less than for Cg.
  • 03:31 Just for clarity, I've defined all the terms in the formulas.
  • 03:34 In some cases, you find that an organization wants to change the tolerance
  • 03:38 percentage range from something other than 20%.
  • 03:41 The calculations are easy to do by hand, but
  • 03:45 you may also want to use Minitab to assist in your analysis.
  • 03:50 To start the study, select Stats, then select Quality Tools.
  • 03:54 Within that pulldown menu, select Gage Study,
  • 03:58 and then you can select Type1 Gage Study.
  • 04:01 Just a quick reminder, Minitab works with columns of data, so when you decide to do
  • 04:06 this study, first put all of the measured data values in a column in Minitab.
  • 04:12 Now, we're ready to have Minitab do the calculations.
  • 04:15 If you did the previous selection,
  • 04:17 you should have this form appear on your screen.
  • 04:21 Put your cursor in the measurement data field,
  • 04:23 then highlight the column of data from the list, and click on the Select button.
  • 04:28 You also need to enter the true value for the item that you're measuring.
  • 04:33 And then enter the spread of the tolerance on the attribute you're measuring.
  • 04:37 This is the spread from the minimum to the maximum allowable value.
  • 04:43 If for some reason you want to change the tolerance percentage from the default of
  • 04:46 20%, or the number of standard deviations in the denominator from the default of 6,
  • 04:51 you can do this by selecting the Options button.
  • 04:54 Once you've made your selections, click the OK button, and
  • 04:57 Minitab will calculate the Cg and Cgk ratios.
  • 05:02 >> The Type 1 Study is often a quick and easy study to complete.
  • 05:06 But unfortunately, I can speak from experience, that if you fail to do it,
  • 05:11 you can waste a lot of time and money doing a major Gage R&R study
  • 05:15 with equipment that is unable to provide measurements that you can trust.

Lesson notes are only available for subscribers.

PMI, PMP, CAPM and PMBOK are registered marks of the Project Management Institute, Inc.

Gift this course