iSixSigma

Selection of Instruments for Control Charts

Six Sigma – iSixSigma Forums Old Forums General Selection of Instruments for Control Charts

This topic contains 4 replies, has 2 voices, and was last updated by  Gabriel 15 years, 1 month ago.

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #34874

    AT
    Participant

    We are in the business of Manufacturing precision Automotive components.We monitor critical parameters using X bar R Charts.My Question is,while selecting instruments for this purpose is there any limit for maximum discrimination.For Example if i want to measure a parameter having tolerance of 0.1 mm  & my process is having a precision of 0.001 mm,if I use an instrument based on my tolerance(Say 0.01 LC),i will end up with a unstable process as my instrument is not able to measure the minute process variations.What is your advice in this case under the following Conditions
    a) I can not use a precision instrument as the tolerance does not calls for it.Though the process is not stable,there is a wide gap between process variation & Specification limits.
    b) I can not change the machine,as I have spare capacity.
     

    0
    #96735

    Gabriel
    Participant

    If you have a tolerance of 0.1, the discrimination (I am assuming it’s the effective discrimination, not the nominal one) of the instrument is 0.01 and you still cannot detect the process variation because it is 0.001… what’s the problem? It is the process of my dreams!
    1) You are so far to put the quality in risk.
    2) Even if you can’t detect the process variation due to common causes, any special cause that generates a variation wich is of any practical significance will be immediatly detected with that instrument.
    3) Switching to an instrument with a precision that is better than the process, like 0.0001mm, wil meak it unnusable exept in very controlled laboratory conditions, and it would me much more expensive to buy, mantain and calibrate. And it would add virtually no value.
    4) Swtching to a worse process is nosense if no economic gain is made in the process, as you say it’s your case.
    5) Intentionally worsening the process is just nonsense!
    6) If any auditor ever tells you that you are not meeting the requirement of having an intrument woith an r&R better than 10% of the process variation, he is right, that requirmenet is not met. But if he makes any problem because of that, in this context, that auditor does not have a brain, just a standard inside his head.

    0
    #96736

    Gabriel
    Participant

    Oh, I forgot. If many of your ranges are zero, you won’t be able to use an Xbar-R or IX-MR chart. But I would not care about of that.

    0
    #96796

    AT
    Participant

    Mr .Gabriel
    Thank you for your clarification.I understood that you are a knowledgable person in this area.Please let me know your email ID for further discussions.Thanks.

    0
    #96812

    Gabriel
    Participant

    ethelygabi at accessplus com ar
    (It is intentionally coded to avoid spam. Just replace “at” by “@” and put the missing dots)
    In the meantime, read the following thread in this forum:
    https://www.isixsigma.com/forum/showmessage.asp?messageID=26524
    It is closely related to your question.

    0
Viewing 5 posts - 1 through 5 (of 5 total)

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