iSixSigma

Combining Cpk values

Six Sigma – iSixSigma › Forums › Old Forums › General › Combining Cpk values

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • February 21, 2008 at 2:08 pm #49409

    Koray Alpaslan
    ★ 10 Years ★
    Participant

    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?

    0
    February 21, 2008 at 3:11 pm #168824

    Mikel
    ★ 10 Years ★
    Member

    Why 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.

    0
    February 21, 2008 at 4:14 pm #168825

    Heebeegeebee BB
    ★ 10 Years ★
    Participant

    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.

    0
    February 21, 2008 at 6:23 pm #168830

    Bower Chiel
    ★ 10 Years ★
    Participant

    Hi KorayYou might consider using multivariate process capability indices – try a Google search and you’ll find references.Good LuckBower Chiel

    0
    February 22, 2008 at 6:29 am #168842

    Koray Alpaslan
    ★ 10 Years ★
    Participant

    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?

    0
    February 25, 2008 at 12:18 pm #168951

    Fontanilla
    ★ 10 Years ★
    Participant

    Koray, re-read Stan’s message.  You’re wasting your time.

    0
    February 25, 2008 at 2:41 pm #168956

    Ron
    ★ 10 Years ★
    Member

    Stop doing that ! There is no mathematical logic behind adding Cpk values.

    0
    February 26, 2008 at 9:53 am #169005

    k.bhadrayya
    ★ 20 Years ★
    Participant

    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
     

    0
    February 26, 2008 at 12:11 pm #169009

    Koray Alpaslan
    ★ 10 Years ★
    Participant

    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
  • Author
    Posts
Viewing 9 posts - 1 through 9 (of 9 total)

The forum ‘General’ is closed to new topics and replies.