iSixSigma

Understanding Gage RR

Six Sigma – iSixSigma Forums Old Forums General Understanding Gage RR

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #41560

    RonJ
    Member

    If a gage study gives 10% Total Gage R&R, can it be said that when using that measurement system to estimate standard deviation, 10% of the estimate is due to the measurement system?
    I suspect things aren’t that simple. Thanks for any help.

    0
    #130761

    sumpat
    Member

    yes you are right

    0
    #130784

    Anuragi Raman
    Participant

    It is not standard deviation but variace (square of standard deviation).
    In the example given, the interpretation will be: 10% of the observed variation is contributed by the measurement system. Minitab should give 10% contribution (of VarComp).

    0
    #130791

    AB
    Participant

    Standard deviation would be calculated and not estimated using a measurement system.

    0
    #130819

    Hola
    Participant

    If I understand your question right, then the anser is – NO!
    If you use the measurement system to measure a bunch of parts to get the standarddeviation for the process, then the contribution from the measurement system is much smaller! First when the Gage R&R gets up in the area of >25% you can say that it has a significant effect on the standard deviation.
    It’s late now. I will get back to you tomorrow.
     

    0
    #130857

    RonJ
    Member

    Thanks Hola,
    Yes, I’m measuring a bunch of part to estimate the stddev of the process. If the stddev is 10 and the gage R&R is 10% then 0.1 of the stddev is due to the measurement system?

    0
    #130859

    RonJ
    Member

    Sorry, it should read: “If the stddev is 10 and the gage R&R is 10% then 1.0 of the stddev is due to the measurement system?”

    0
    #130928

    Hola
    Participant

    If gage R&R is 10%, and your standard deviation from the process is 10, then the standard deviation from the gage is approx 3,16.
    Use the formula for adding standard deviations:
    Observed standarddeviation * observed standard deviation = (R&R standard deviation * R&R standard deviation) + (real process standard deviation * real process standard deviation)
     3,16*3,16 + 9,48 * 9,48 = 10*10
    Best regards,Hola

    0
    #130936

    RonJ
    Member

    Great, that is exactly what I’m looking for.
    Thanks again
     

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

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