Is there a Cpk requirement for automotive industry stated in TS16949?

[keatwoon]

Hi everyone. Is there a Cpk requirement for automotive industry stated in TS16949? I couldn't find it in the manual. Maybe 1.67 is still the standard practice?

[Wes Bucey]

My memory is that this is a customer specific requirement, not TS16949 general requirement Typically, they make a distinction among critical characteristics, key characteristics, and significant characteristics as well as between "capability" shown by PPAP and actual during real production.

Here's a sample clause from a customer specific requirement:

Characteristic mmmmmmmmmmmPPAPmmmmmmmmOn-going production

CriticalmmmmmmmmmmmPp/Ppk > 2.0 n=50mmmmmCp/Cpk > 1.67
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmm& Continuous Improvement

Key mmmmmmmmmmmmPp/Ppk > 2.0 n=40mmmmmmCp/Cpk > 1.67
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmm& Continuous Improvement

Significant mmmmmmmmPp/Ppk > 1.67 n=30 mmmmmCp/Cpk > 1.33
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmm& Continuous Improvement

[D.Scott]

Originally Posted by keatwoon

Hi everyone. Is there a Cpk requirement for automotive industry stated in TS16949? I couldn't find it in the manual. Maybe 1.67 is still the standard practice?

TS16949 - 8.2.3.1 states "The organization shall maintain manufacturing process capability or performance as specified by the customer part approval process requirements"

Most of our customer requirements state 1.67 for the PPAP. Others state a minimum of 1.33. Some have no specified requirement.

Dave

[Trolle]

Originally Posted by Wes Bucey

My memory is that this is a customer specific requirement, not TS16949 general requirement Typically, they make a distinction among critical characteristics, key characteristics, and significant characteristics as well as between "capability" shown by PPAP and actual during real production.

Interesting this! Just the other day I was at an interview being asked just this very same question!!

My answer was in line with the quotation above, unfortunately I am not as capable as y in quoting from the standard. Now I do know that in real life (eg business life) things are a tad more complicated. But the interview setting wasn’t the place to elaborate on such practical issues.

However the MD doing the interview insisted that it was not the customer but more the tools themselves, those that you use in production that decides issues of capability.
Why would he say this? Could it be something that is/was true in the world of the old QS 9000??
And please could someone help me with this? Another issue at the interview concerned the relation between the old QS and ISO TS. I was under the impression that QS is to be phased out and replaced by TS. The MD however was very adamant about the fact that TS standard is only there to enhance the QS and that they would continue to coexist. I really would appreciate if someone could tell what goes here.

Cheers!

[Howard Atkins]

Originally Posted by Trolle

However the MD doing the interview insisted that it was not the customer but more the tools themselves, those that you use in production that decides issues of capability.
Why would he say this? Could it be something that is/was true in the world of the old QS 9000??

The capability that you achieve is based on the tolerances and the tools.
You must build the tool in a more accurate manner to achieve a higher capability as it must be in the middle of the tolerance. The capability is based on how much the MD is prepared to pay for the tooling.
The customer decide you need the equipment to comply

Originally Posted by Trolle

And please could someone help me with this? Another issue at the interview concerned the relation between the old QS and ISO TS. I was under the impression that QS is to be phased out and replaced by TS. The MD however was very adamant about the fact that TS standard is only there to enhance the QS and that they would continue to coexist. I really would appreciate if someone could tell what goes here.

Cheers!

You are right at the end of 2006 QS 9000 ceases to exist, the confusion could be that for the customer requirements the handbooks for PPAP, APQP,SPC,FMEA and MSA can remain the same

[D.Scott]

There is also the confusion of the 2 standards "co-existing" until December of 2006. Both QS and TS are active systems until that time.

Dave

[bobdoering]

Don't forget the little caveat found in AIAG PPAP 4th edition (2.2.11.5, NOTE) that Cpk's cannot be calculated if the distribution is non-normal, such as precision machined diameters and lengths. For more information, see Statistical process control for precision machining

[Murphy's Law]

Let's get back to basics.

My understanding of SPC and Cpk < 1.67 was that if you do not meet it, then 100% testing is required.

If you are 100% testing, then provided you understand the influence of GR&R and you factor in an appropriate guardbanded test limit, you should be ok. (There should be a test and/or yield hit !)

Provided you then do seperate QA sample which should be to a clipped distribution, AOQ should be zero. Provided there is no test escape possibility proven over a large sample size, your customer should be ok with that.