iSixSigma

sigma calculation

Six Sigma – iSixSigma Forums Old Forums General sigma calculation

Viewing 9 posts - 1 through 9 (of 9 total)
  • Author
    Posts
  • #51430

    nancy
    Participant

    I am doing a roll – up sigma calculation for my organisation . There are 20 processes in all in my organisation doing different kind of work . My boss has asked me to find out the sigma level at which the organisation is working at .
    How should I get to sigma level of the organisation ? In most of these 20 processes it’s quality / TAT which is the customer CTQ on which sigma is being measured . However the no. of data points available are not the same in each of these 20 processes .

    0
    #178229

    Ron
    Member

    Don’t fall into that trap! Sigma values are defined for a specific process and only make sense for the process you are working on.
    If you want to rate your company you can calculate your PPM level on outgoing quality as reported by your customers and convert that to a sigma level and refer to it as the outgoing quality sigma level.

    0
    #178231

    benjammin0341
    Participant

    I concur.

    0
    #178264

    nancy
    Participant

    Thanks for your response but what is the trap and why should I choose PPM and not DPMO ?

    0
    #178267

    nancy
    Participant

    Request a reply for the same

    0
    #178269

    nancy
    Participant

    Request someone to look on  this . My boss is after my life on this and I need to provide an explanation

    0
    #178270

    Jered Horn
    Participant

    The “roll-up sigma calculation” for your organization is not uncommon.  Done right, it can be a value adding exercise.  I would refer you to chapter 9 in Breyfogle’s “Implementing Six Sigma”.  Here’s an exerpt from the end of that chapter…
       … 
    A variety of responses can be expected when many are asked their perception of Six Sigma.  Among other things, responses will depend upon a person’s background and position within an organization.  One possible response involves the unique metrics of Six Sigma.  Another possible response is that Six Sigma can offer a methodology to integrate meaningful metrics with effective process improvements.
      I have seen organizations get very frustrated and then drop a Six Sigma initiative because management tried to drive improvement through the metrics.  Another common problem with the implementation of Six Sigma is creating Six Sigma metrics using very limited data that do not represent the population of interest.  In addition, the comparison of suppliers by examining only reported Six Sigma metrics without an understanding of how the metrics were originated can lead to distorted results because the data might not have been similarly obtained.
      It is difficult, expensive, and time-consuming to collect good data and create a metric that is meaningful.  This task can be even more difficult if metrics are forced into a one-size-fits-all Six Sigma format.  Metrics are important; however, some formats are more beneficial for a given situation.
    … 

    0
    #178278

    Old MBB
    Participant

    Hi Nancy,
    A “Six Sigma” organization IS NOT one that generates 3.4 DPMO watching product move from start to finish.  Rather, it is an organization where ALL of its processes individually function at 3.4 DPMO.  The space shuttle may have individual processes operating at 8s (with redundancy) but the vast number of processes creates an RTY that makes failures possible…
    So…  What many organizations do is the following:
    1) Collect data from all of the processes that you can
    2) Calculate a yield value for each (either with DPMO or e^(-dpu))
    3) Calculate the geometric mean of all of those yields (the nth root of the n-way RTY product)
    4) Translate the resulting “normalized yield” to a Sigma using std math…
    What you now have is the Z-score of your organization’s combined processes.  It is useful directionally to see which processes are “below sea level” or even to compare different facilities.
    Good Luck,
    Bob

    0
    #178286

    Tim
    Member

    ppm vs dpmo is a smokescreen, the point is that you cannot simply rollup the metrics to get a sensible number as you’ve no view on the significance of each process to the overall defect rate for the end customer. So the suggestion is that you measure that defect rate directly.

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

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