iSixSigma

MSA and Tolerance

Six Sigma – iSixSigma Forums Old Forums General MSA and Tolerance

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #31370

    Serge
    Member

    Hi
    Can anybody explain why, for attribute data, the measurement system variation should be less than 10% of the overall variation while, for continuous data, it should be less than 30%?Rgds

    0
    #82672

    Never Mind
    Participant

    Where did you take that from?

    0
    #82676

    Hemanth
    Participant

    Hi
    Attribute data doesnt give you much of strenght in analysing variation. Secondly attribute data is subjective in many cases and thats why the norms in general are more strict for attribute data.
    Hope this helps

    0
    #82681

    Tierradentro
    Participant

    I am with Never mind – where did you get this stuff?
    And no Hemanth, it does not help. There is no relation between the question and your answer. There is no such criteria.

    0
    #82693

    Ravi Khare
    Participant

    The AIAG MSA manual recommends attribute data to be analysed by
    The Cross Tab method – that evaluates interrater agreement . An equivalent of Reproducibility studies in Variable data. An interrater agreement of 70% or above is desirable.
    The signal detection method – that evaluates the zone of uncertainty (or grey zone) of an attribute gage where all appraisers across trials do not agree.
    The Gage performance curve – that evaluates and plots the probability of acceptance across a given range of sizes. It computes the repeatability and the bias of the attribute gage.
    Ravi Khare
    http://www.symphonytech.com

    0
    #82833

    Andhale
    Participant

    attribute data is only 64 % accurate as continous data hence for accepting the gage we have this stricter norms

    0
Viewing 6 posts - 1 through 6 (of 6 total)

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