Combining Cpk values
Six Sigma – iSixSigma › Forums › Old Forums › General › Combining Cpk values
- This topic has 8 replies, 7 voices, and was last updated 14 years, 3 months ago by
Koray Alpaslan.
-
AuthorPosts
-
February 21, 2008 at 2:08 pm #49409
Koray AlpaslanParticipant@Koray-AlpaslanInclude @Koray-Alpaslan in your post and this person will
be notified via email.In our company, we are trying to put Cpk values for different characteristics of a manufacturing process together (e.g. length, weigth, thickness of a metal block). The aim is to get a statistically acceptable but single value based on the Cpk values of these process characteristics, which will allow us compare different batches for an annual evaluation. Is it possible to get a single, combined Cpk value for a process or are there other ways you may suggest for such an evaluation?
0February 21, 2008 at 3:11 pm #168824Why in the world would you waste resources on this?Decide minimally acceptable Cp and Ppk value for your CRITICAL
processes and those process’s CRITICAL measures and have
reviews monthly (annual tells everyone it’s not important right now)
on what is being done on those measures that do not meet the
minmum requirement. Done. So simple and your processes will improve which is the real
objective, not coming up with some meaningless number.0February 21, 2008 at 4:14 pm #168825
Heebeegeebee BBParticipant@Heebeegeebee-BBInclude @Heebeegeebee-BB in your post and this person will
be notified via email.I gotta say, I’m in 100% agreement with Stan. You have individual Cpk’s…Rack and stack, then execute.
If you must aggregate/amalgamate, I would look at Rolled Throughput Yield.0February 21, 2008 at 6:23 pm #168830
Bower ChielParticipant@Bower-ChielInclude @Bower-Chiel in your post and this person will
be notified via email.Hi KorayYou might consider using multivariate process capability indices – try a Google search and you’ll find references.Good LuckBower Chiel
0February 22, 2008 at 6:29 am #168842
Koray AlpaslanParticipant@Koray-AlpaslanInclude @Koray-Alpaslan in your post and this person will
be notified via email.Thanks everybody. After a quick search in Google, there are various topics on “multivariate process capability index”. Is there a widely accepted model for cases with 4 or 5 specifications? Mostly, there’s info on bivariate models, but which model should be picked for cases with multiple ones?
0February 25, 2008 at 12:18 pm #168951Koray, re-read Stan’s message. You’re wasting your time.
0February 25, 2008 at 2:41 pm #168956Stop doing that ! There is no mathematical logic behind adding Cpk values.
0February 26, 2008 at 9:53 am #169005
k.bhadrayyaParticipant@k.bhadrayyaInclude @k.bhadrayya in your post and this person will
be notified via email.dear sir,
will you give me your mail id sotaht i acn share with you a case study with all calculation to get an over all Cpk from individual values.
k.bhadrayya
0February 26, 2008 at 12:11 pm #169009
Koray AlpaslanParticipant@Koray-AlpaslanInclude @Koray-Alpaslan in your post and this person will
be notified via email.To K.Bhadrayya:
My e-mail is [email protected] I would be glad if you could share the case study on multivariate process capability index calculation. Thanks.
To all other colleagues:
Thanks again for your help. Any suggestion on multivariate process capability indices will be greatly appreciated.0 -
AuthorPosts
The forum ‘General’ is closed to new topics and replies.