iSixSigma

Product Report vs Process Report for Non-normal Data

Six Sigma – iSixSigma Forums Old Forums General Product Report vs Process Report for Non-normal Data

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #34069

    Donald Knox
    Participant

    Currently I have been having a disagreement with several BBs and MBBs around what report should be used to baseline non-normal data. My position is that non-normal data should be baselined using the product report. I am sure this is right but I am not sure why. I think it has something to do with the process report overstressing the outliers. Can someone please give me a good argument stance to take on this? I would like to know which report is proper to use and why.

    0
    #93959

    Marc Labrie
    Participant

    PROCESS Report is used with Continous data that follow a bell curve distribution. Calculations are based on an assumption of normally distributed data. 
    while PRODUCT Report apply to Discrete data and therefore can be used for all type of distribution. Calculations are based on the numbers of defects per opportunity.

    0
    #93962

    Mikel
    Member

    The process should be baselined with actual data with an appropriate description of central tendancy and variation based on the underlying distribution. A rational thought process of whether the distribution makes sense or is there some other things to be understood goes a long way as well. Forget the canned reports, understanding the process behavior is the only thing of interest.
    Go get some good training for your BB’s and MBB’s.

    0
    #93972

    Arend
    Participant

    Dear Donald,
    I think the solution for how to do a baseline with non-normal data depends entirely on what sort of non-normality you are talking about. Is it a matter of many outliers ‘on top of’ a further normal distribution, or is the distribution essentially non-normal like for instance lognormal or exponential or even bimodal.
    It could also depend on what aspect of the process you are interested in. For instance, I have recently dealt with a process that gave many outliers on top of a distribution that was normal (and very narrow and stabile). I decided to leave the normal distribution for what it was, because my interest was in solving the outliers. The baselining was done on the frequency of occurence of outliers.
    If your process is continuous but essentially non-normal, a transformation might work very well to make it normal and do baselining in the conventional way.
    kind regards,
    Arend

    0
    #93989

    Vikas Jalan
    Member

    Hi,
    The reason why we can’t use “process report ” for non-normal continuous data is explained below :
    – Process report generates a normal distribution by using Mean and std deviation of the data
    – By using this normal distribution, probability of defects at USL and LSL is computed
    – Depending upon the yield ( 1- Total probability of defects ), Z values are reported out.
    If the data is not normal, taking mean and std deviation to generate a normal distribution will lead to error in computing probability of defects….hence we say go thru product report method.
     
    Hope this helps.

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

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