Gage R&R Sample Selection - Can we include NG Samples?

[WEAVER]

Can we include NG samples when conducting GR&R?

 

[Bev D]

Re: GR & R Sample Selection: Can we include NG Samples?

yes. you should include samples that span the full range of observed variation. (you should not create bad samples)

 

[Miner]

Let me rephrase what BevD said because it could be interpreted multiple ways.

There are three possible scenarios for gage application(s):

1. Part inspection only
2. Statistical uses (i.e., Statistical Process Control (SPC), capability studies, DOEs, etc.) only
3. Both 1 and 2

For scenario 1, part selection is discretionary because the %Tol metric does not use the study variation as part of the calculation. If linearity studies have or will been done, it truly does not matter how the parts are selected. If linearity studies have/will not be done, it is a good practice to spread the samples out over the full range of tolerance and even beyond.

For scenario 2 (and 3) part selection is absolutely critical if %SV (Study Variation) or ndc (number of discrete categories) will be used. The parts absolutely MUST reflect actual process variation. If the parts are deliberately selected to be a larger spread than the actual process, study variation will be inflated and the %SV or ndc will be very misleading. Doing such with full knowledge of the implications is unethical. A better approach is to use the results of a capability study to calculate the %PV (Process Variation) metric. %PV is more accurate than %SV.

 

[Satellite]

If the gage is an Attribute gage, you absolutely have to have parts out of print or the formulas don't work well.

That is unless it is impossible for your process to produce parts that are bad. Usually the impossible is due to custom tooling. You would have to make a bad tool to get the bad parts.

As long as the parts can be made from an offset, I include bad parts even in a variable gage R&R. After all you want to make sure that the gage will correctly find the bad parts.

 

[WEAVER]

How can I make sure that the samples reflect the full process variation?
We use %Tol but we also consider the NDC (should be greater than 5). Do I need to get the same type from different machines/production runs?

 

[Miner]

How can I make sure that the samples reflect the full process variation?
We use %Tol but we also consider the NDC (should be greater than 5). Do I need to get the same type from different machines/production runs?

How is this gage used? For inspection only, SPC only or both?

 

[WEAVER]

We use this for inspection but we do not keep record for SPC-- at least not yet.

 

[Miner]

As long as you are only using the gage for inspection then use only %Tol as the metric. ndc only applies when you will use the gage for SPC.

The %Tol equation only uses the tolerance and the R&R variation, so the method of collecting samples will have no impact on the results. This is not true of ndc or %SV, so it you ever start using the gage for SPC, beware of this.

 

[sunituni]

Dear all,
Anyone can explain what's the difference of '% of Tolerance analysis', '% of study variation', '% of process variation' & '% contribution'....

 

[Miner]

Here is some information in my MSA Blog.

• % Tolerance - This is the percentage of the combined repeatability and reproducibility variation (R&R) as a percent of the tolerance. It is used to assess the suitability of a gage for inspection to a tolerance.
• % Study Variation - This is the percentage of the combined R&R as a percent of the variation of the parts selected for use in the study. These parts are intended to be representative of the actual process variation, but are often not very representative, so the metric is often higher or lower than reality. It is used to assess the suitability of a gage for Statistical Process Control (SPC) or for statistical studies.
• ndc - Number of Distinct Categories. Used for the same purpose as % Study Variation.
• % Process Variation - Similar to % Study Variation, but the variation used is from an actual capability study instead of the parts used for the R&R study. This is a much more reliable metric than % Study Variation or ndc.
• % Contribution - This metric is based on the variance rather than the standard deviation as is % Study Variation. It is mainly used to prioritize the major source of R&R variation for improvement.