Cp/Cpk on Gages for MSA (vs. Cg/Cgk)

D

dumasbeef

I have an auditor asking me if we do Cp/Cpk on gages for MSA. I've explained that we do Cg/Cgk for MSA but not Cp/Cpk, so I am getting a non-conformance. Has anyone ever heard of doing Cp/Cpk on a gage? I've checked the MSA manual, and it discusses relationships of observed process variation, and variance in the measuring system. It does not mentions Cp/Cpk on an actual gage. Am I off base or missing something here. I always thought that measurement system capability indices were Cg/Cgk. I'm lost on this one, and could really use some help. Any assistance would be greatly appreciated.
 

howste

Thaumaturge
Trusted Information Resource
Hi Dumasbeef, and welcome to The Cove forums. :bigwave:

Since this is not posted in a standard-specific forum, what is the standard the auditor is auditing to? What is the requirement the auditor is quoting in the nonconformity?

I agree that Cp/Cpk are generally product-related, not gage-related.
 
D

dumasbeef

We are being audited to IATF 16949. I'm just trying to get my mind wrapped around where she is coming from. Any help would be most welcome.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
Did they actually point to the requirement for the requirement this analysis? (If so, I would like to know where it is specified as required). Even though most gage error is normally distributed - not all is - especially on physically limited characteristics like roundness, flatness, etc. Cp and Cpk ONLY apply to normal variation and centered target variables PER THE AIAG PPAP BOOK! So, it can not apply to all gages, even if it applied to any.
 

howste

Thaumaturge
Trusted Information Resource
We are being audited to IATF 16949. I'm just trying to get my mind wrapped around where she is coming from. Any help would be most welcome.

Since you're being audited to IATF 16949, let's look at what it requires for MSA:
Statistical studies shall be conducted to analyze the variation present in the results of each type of inspection, measurement, and test equipment system identified in the control plan. The analytical methods and acceptance criteria used shall conform to those in reference manuals on measurement systems analysis. Other analytical methods and acceptance criteria may be used if approved by the customer.

Records of customer acceptance of alternative methods shall be retained along with results from alternative measurement system analysis (see Section 9.1.1.1).

Note Prioritization of MSA studies should focus on critical or special product or process characteristics.
I don't see anything in there that requires Cp/Cpk for gage studies. It says you analyze the variation in the gage systems but doesn't specify how.

Let's look a the AIAG MSA Manual. Here's the stated purpose in the manual:
The purpose of this document is to present guidelines for assessing the quality of a measurement system. Although the guidelines are general enough to be used for any measurement system, they are intended primarily for the measurement systems used in the industrial world. This document is not intended to be a compendium of analyses for all measurement systems. Its primary focus is measurement systems where the readings can be replicated on each part. Many of the analyses are useful with other types of measurement systems and the manual does contain references and suggestions. It is recommended that competent statistical resources be consulted for more complex or unusual situations not discussed here. Customer approval is required for measurement systems analysis methods not covered in this manual.
First, the MSA manual is a guidance document, not a requirements document. Second, if you read the entire manual (I have) you'll find that using Cp/Cpk for gage studies is not covered in the manual and per the last sentence in the quote you would need to have customer approval to even use that method. Unless you have a customer-specific requirement that specifies that method, I don't see where they are coming from.

The short story is that the auditor is trying to write a nonconformity to a requirement that doesn't exist. The ISO definition of nonconformity is "non-fulfillment of a requirement." The auditor must specify a requirement if they write a nonconformity, which is why I asked what they quoted as the requirement. My guess is that they haven't quoted one.

You didn't mention what kind of audit this is. If it's a certification body audit my suggestion is that you appeal the finding with the certification body. Every certification body is required to have an appeals process using personnel independent from the auditor. Findings like this is what it's for. If the auditor is still there, please ask them to show you where the requirement is and I suspect that the NC will magically disappear.
 

Miner

Forum Moderator
Leader
Admin
Another reason that Cp/Cpk for a gage makes no sense is you could easily obtain a Cp > 2, yet still have a bad gage that is totally unacceptable. To avoid this you would have to have specification limits for the gage that are different from that of the product. Where would you obtain these?

What does this auditor think that Cp/Cpk for a gage would provide that is not already provided by % Tolerance?
 
D

dumasbeef

This is a IATF 16949 readiness audit. I have read the manual as well, and really couldn't find much. The only item that I could find was on pg. 20 & 21. It discusses relationship of the standard deviation of the measurement system, and the standard deviation of the measurement system. My problem is if I have to calculate Cp (Tolerance Range/6sigma), which standard deviation should I use, "Total GRR" or "Total Study Variation"? In all my years, I have never broached the subject of Cp in a gage study. Thank you for your help.
 

Miner

Forum Moderator
Leader
Admin
It discusses relationship of the standard deviation of the measurement system, and the standard deviation of the measurement system.
???

These are the same. Did you mean relationship of the standard deviation of the measurement system to the standard deviation of the study/process variation?

If so, that is what % Study Variation/% Process Variation tell you. No need to calculate a Cp index.
 
D

dumasbeef

I'm sorry, perhaps I should have been more clear. We use Minitab, and it gives "Total Gage R&R", and "Total Variation". "Total Variation" includes the part to part variation in the study. Where as "Total Gage R&R" only includes the gage/operator interaction and variation. If I am to calculate the Cp for MSA, I want to know which standard deviation to use. On a side note, all things point to an auditor attempting to dictate corporate procedure via interpretation of a standard.
 

Attachments

  • mtb___sessionwindow.pdf
    604.4 KB · Views: 508
Top Bottom