iSixSigma

potential within vs overall capability

Six Sigma – iSixSigma Forums Old Forums General potential within vs overall capability

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #50024

    MB
    Participant

    Hi,
    Im am doing a cpk assesment and my results for potential within to overall vary enormously. Cpk – 4.72 / Ppk 0.92.  Can anyone explain to me why this is the case?  I have one data set so have put subgroup size of one in monitab.
    Any help would be appreciated.
    Thanks
    MB

    0
    #171807

    Taylor
    Participant

    Just from the 30000 ft level ,Your short term capability is extremely off center from the target (Cpk), this is giving you a long term capability skew (Ppk). Or you have calculated incorrectly.
     
     

    0
    #171903

    Brian M
    Participant

    First question… is the process in control?I doubt it. With a sub-group of one I take it you are using a moving range and the part to part variation is low within the sub-groups which is resulting in an artificially high est st dev giving you a good Cpk… but the overall variation it large driving your calculated st dev down and killing you Ppk. If the process is not in control, then all of this is meaningless anyway.

    0
    #171919

    Ron
    Member

    Cpk estimates use an approximation of SD due to the extremely small sample sizes involved.
    Ppk calcualtion use the classical SD calcualtion and an actual standard deviaiton.

    0
    #171922

    walden
    Participant

    Brian,
    I agree with you that, with the huge disparity between Cpk and Ppk, I can’t believe the iX-mR charts are in control. If they aren’t in control, Cpk should not be calculated or analyzed since the process is not predictable.
    You have it backwards in your explanation though. Low variation between the subgroups will lead to a low estimate of the standard deviation, resulting in a high Cpk. Sudden changes in level (to name one possible cause) would result in the RMS calculated standard deviation to be large, resulting in a low Ppk.
    Without seeing the data or control chart, we’re all just speculating as to what caused this disparity.
    Chris

    0
    #171940

    Brian M
    Participant

    Chalk it up to a brain fart. Thanks for the correction, thats what I meant… Low est st dev=high Cpk

    0
    #171961

    Forrest W. Breyfogle III
    Member

    I have some theories; however, would really need to look at your data and ask a few questions about the process from which the data are collected.
    If you would like for me to look at your data and share my thoughts, contact me directly at [email protected] and/or 512-918-0280. 
    Forrest Breyfogle
     

    0
    #171965

    sandrine berthet
    Member

    How to purchase your new books?
    thanks and regards

    0
    #171967

    Forrest W. Breyfogle III
    Member

    Per your request — a listing of my four new book-volumes will appear when you do a search in Amazon.com for “Integrated Enterprise Excellence” or “Forrest Breyfogle”

    0
Viewing 9 posts - 1 through 9 (of 9 total)

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