The Elsmar Cove Business Standards Discussion Forums More Free Files Forum Discussion Thread Post Attachments Listing Elsmar Cove Discussion Forums Main Page
Welcome to what was The Original Cayman Cove Forums!
This thread is carried over and continued in the Current Elsmar Cove Forums

Search the Elsmar Cove!

Wooden Line
This is a "Frozen" Legacy Forum.
Most links on this page do NOT work.
Discussions since 2001 are HERE

Owl Line
The New Elsmar Cove Forums   The New Elsmar Cove Forums
  Statistical Techniques and 6 Sigma
  Cpk and Ppk calculations for TE

Post New Topic  Post A Reply
profile | register | preferences | faq | search

UBBFriend: Email This Page to Someone! next newest topic | next oldest topic
Author Topic:   Cpk and Ppk calculations for TE
mibusha
Forum Contributor

Posts: 60
From:Royal Oak, Michigan USA
Registered: Nov 98

posted 06 June 2000 11:14 AM     Click Here to See the Profile for mibusha   Click Here to Email mibusha     Edit/Delete Message   Reply w/Quote
Canât avoid it any longer. Must do something with CpK and PpK to satisfy QS 90000 TE requirements and to facilitate continuous improvement efforts. Iâm in charge of registering 3 separate but related companies. Process capability and performance calculations are not my strong point. Three scenarios with three questions:

1)This company makes wood, plastic and metal core box patterns that are used to produce crankshafts or cylinder heads at the Ford foundries. We make anywhere from 2 to 6 at a time following Fordâs specific instructions (usually taken over the phone). All dimensions are checked to ensure they get what they asked for but no final dimensions are actually recorded anywhere. Plus they are always exact because if they arenât, the Pattern Maker keeps working on it until it is correct. Question: What can I do to calculate CpK and PpK ?

2)Machine slides. We produce 2 to 6 at a time (total of 10 to 30 per month). All slides have a ăprint specä to which we final audit for flatness, parallelism, surface fit, etc. We donât have ăbatchesä or enough volume to obtain samples in order to calculate CpK and PpK . We do keep all final audit reports that state the USL/LSL and measured values. Question: Is there a way to use these measurements to come up with a valid CpK and PpK ?

3)Automated Machines. These projects take from 3 to 8 months. Usually they are not full turn-key (missing the controls so wonât run). Question: How does one calculate CpK and PpK for this type of manufacturing? We canât even do the Machine Qualification Runoff Testing (another requirement of TE) because controls are not in place and the customer does not require or want it done.


I would really appreciate help on these issues from knowledgeable folks.

IP: Logged

Laura M
Forum Contributor

Posts: 299
From:Rochester, NY US
Registered: Aug 1999

posted 08 June 2000 01:09 PM     Click Here to See the Profile for Laura M   Click Here to Email Laura M     Edit/Delete Message   Reply w/Quote
I think I've digested the scenarios and refreshed my memory enough to hopefully shed some light. I'm open for others interpretations, as my experience is somewhat limited.

First, reference the introduction ăimplementationä section·.ähowever it is recognized that not all of these requirements are applicable to every organizationä ö unfortunately they use 4.4 ö Design ö as the example which is fairly obvious in regards to applicability. Having said that Îcontinuous improvementä is obviously applicable, but how you go about accomplishing it may or may not involve Cpk/Ppk.

Since you mentioned facilitating continuous improvement efforts, Iâm assuming you are referring to 4.2.5, and page 19 of QS, which says if you have variables data, that you need to optimize characteristics and parameters toward a target value. Your process should first be identified as ăstableä ö hence, SPC would apply. In the first 2 scenarios, short run SPC would seem to make sense ö reference T/E supplement, page 14. Since you mention that the pattern maker works until the parts are exact, Iâm not sure you even have an SPC application in the first scenario, but I would start writing down the data, and look for other CI possibilities ö like first time quality, delivery timeliness, I am not personally familiar with any ăcustom workä scenario which would allow for SPC and Cpk to be used ö but am willing to learn if anyone else is.

The Phase 1,2,and 3 process performance referred to in the T/E supplement would be regarding the product produced with the tooling. It doesnât appear that you have the capabilities to do this. Anyway, once stability is proved, use data from your shortrun SPC charts for the process capability. Make sure you note process changes.

In the 3rd case, the Machine qualification section would apply. Since the machines are missing the controls, I would ask if you run parts manually? The SPC/CPk, PPk, would apply to the product produced by the machines. I worked with a T/E client that in some cases was just material handling equipment, no product was produced. Their qualification was the 50/20 dry run and MTBF, MTTR data. Without controls, Iâm wondering how you verify your equipment can produce parts within specification today. Bottom line is the table on page 6 of T/E says, ăAs required by the customer.ä Also take a close look at the ăintentä stated in the bullets on page 6 of the T/E supplement. BTW ö when do controls get installed? At the customer?


Finally, note that section 4.2 ö Maintaining Process Control, in the T/E supplement refers to high volume tooling ö if you are not required to ămaintain process controlä would imply that you shouldnât be required to use Cpk analysis ongoing for continuous improvement either.
As stated on page 18 of QS ö the continuous improvement philosophy shall be deployed throughout the organization. They couldnât eliminate the product Cpk/PPk aspect for the T/E supplement because you do have high volume machine tool suppliers that this section could apply to.

With my training and experience to date, thatâs the best I can do, and Iâm open for discussion. Would love to hear from others that have been registered to T/E or the T/E registrars out there.

Now we can all probably use a refresher on Short-run SPC!

IP: Logged

mibusha
Forum Contributor

Posts: 60
From:Royal Oak, Michigan USA
Registered: Nov 98

posted 09 June 2000 04:04 PM     Click Here to See the Profile for mibusha   Click Here to Email mibusha     Edit/Delete Message   Reply w/Quote
Very thourough response Laura, thanks. I think continuous improvement statistics will be the key here. Try to get data on MTTF, MCTF, etc. from the customer and then attempt to improve on it. Also count on our own improvements in delivery, cost, R&M predictions, etc. Might be able to get something from the customer on process capability but usually they don't want to be bothered. Could also do something in terms of turnaround for our repair work, and, as you suggested, first time quality. As far as Mach Runoff Qual Test-I've developed procedures for it and everyone has been trained on the requirements but...until we also install the controls it is not possible to run even one part for CPk/Ppk or 50/20hr data. Perhaps I'll see if I can follow the machine to the customer's site and participate next time around. Believe it or not, the picture is a bit clearer now.

IP: Logged

Laura M
Forum Contributor

Posts: 299
From:Rochester, NY US
Registered: Aug 1999

posted 09 June 2000 05:15 PM     Click Here to See the Profile for Laura M   Click Here to Email Laura M     Edit/Delete Message   Reply w/Quote
So you have a way of verifying your equipment to the customer specs before you ship?

IP: Logged

mibusha
Forum Contributor

Posts: 60
From:Royal Oak, Michigan USA
Registered: Nov 98

posted 14 June 2000 04:58 PM     Click Here to See the Profile for mibusha   Click Here to Email mibusha     Edit/Delete Message   Reply w/Quote
Working on that now. Apparently we do in fact perform 20 hr runs although I haven't seen any records of it!! The machine slides are put through a rather extensive "2nd audit" to ensure that they are to customer spec. Also I was able to do some Target average and range charts to show capability and process performance.

My challenge is to document all this better than in the past. Thanks for the help.

IP: Logged

Laura M
Forum Contributor

Posts: 299
From:Rochester, NY US
Registered: Aug 1999

posted 20 June 2000 08:47 PM     Click Here to See the Profile for Laura M   Click Here to Email Laura M     Edit/Delete Message   Reply w/Quote
Thought someone else might have chimed in by now? Are there any T/E certified companies that can add to my comments?

IP: Logged

Geoff Cotton
Forum Contributor

Posts: 34
From:Staffordshire, England
Registered: Jun 2000

posted 26 June 2000 08:26 AM     Click Here to See the Profile for Geoff Cotton   Click Here to Email Geoff Cotton     Edit/Delete Message   Reply w/Quote
Could the 20hr run data be collected every hour to form part of a Ppk study?

Geoff
(not familliar with T/E)

IP: Logged

Laura M
Forum Contributor

Posts: 299
From:Rochester, NY US
Registered: Aug 1999

posted 21 July 2000 04:29 PM     Click Here to See the Profile for Laura M   Click Here to Email Laura M     Edit/Delete Message   Reply w/Quote
Mibusha,
How did this discussion go for your audit?

Laura

IP: Logged

All times are Eastern Standard Time (USA)

next newest topic | next oldest topic

Administrative Options: Close Topic | Archive/Move | Delete Topic
Post New Topic  Post A Reply Hop to:

Contact Us | The Elsmar Cove Home Page

Your Input Into These Forums Is Appreciated! Thanks!


Main Site Search
Y'All Come Back Now, Ya Hear?
Powered by FreeBSD!Made With A Mac!Powered by Apache!