iSixSigma

why 6 sigma? why its not called 5 or 3 sigma?

Six Sigma – iSixSigma Forums Old Forums General why 6 sigma? why its not called 5 or 3 sigma?

Viewing 18 posts - 1 through 18 (of 18 total)
  • Author
    Posts
  • #44734

    SACHIN UMRIKAR
    Member

    hi to all
    may i knw plz why its called 6 sigma? why not 5 or 4 sigma?
    Thanks
    sachin

    0
    #143927

    Marlon Brando
    Participant

    Because  6Sigma means 3.4 DPMO (Defects-free  products  and  services) 

    0
    #143932

    Brit
    Participant

    Please browse the blue menu on the left – click on new to six sigma. It will help explain a lot.  If you have more questions after, we encourage it!

    0
    #143947

    Orang_Utan
    Participant

    Because Motorola quality was stink in late 70s. Their marketing people felt pretty “sick”. :-O Actually Japanese semicon companies were operating at 4 to 5 sigma level at that point of time. Motorola must set a higher target in order to beat their competitors. So, 6 sigma is a logical and rational choice.

    0
    #143953

    Mike Costello
    Participant

    There actually was a very logical reason for setting the goal at 6 Sigma.  The issue has to do with the difference between common cause and special cause variation (short term vs long term). 
    Over the short term a process that has three standard deviations betwen the mean and the specification limits will produce about 2.7 defects per thousand opportunities.  If we increase the capability so that the specification limits are at 4.5 standard deviations away from the mean, you would expect about 3.4 defects per million opportunities in the SHORT TERM. 
    At 4.5 sigma, your not making many defects, but if the proccess shifts a little bit over time, you start making more defects.  At this level of capability there isn’t much room for medium to long term drift.
    Years ago, Motorola did a study to understand the difference between short term and long term variation.  What they found was that in most processes, the special causes added variation over the medium to long term which increased the total variation by between 1 and 2 standard deviations.  On average, the difference between short term and long term variation is about 1.5 sigma. 
    So if you wanted a process that created almost no defects regardless of the presence of special causes, you would want to reduce the variation to the point where the spec limits are about six standard deviations away from the mean. 
     

    0
    #143956

    Marlon Brando
    Participant

    Clever Reply:hitting  the  target

    0
    #143984

    Mike Carnell
    Participant

    Sachin,
    To provide a diffinitive answer on something that happened so long ago and at a level that nobody on this forum was privey to at the time makes everything conjecture. The opinions are pretty much driving their own agenda or whatever ocb thing they have going on.
    Motorola did lose its television business in the 70″s. That really drove the push to do something. Nobody really knew what.
    The piece that people seem to miss is a chart that floated around Motorola in the 80’s and 90’s. It is simple enough to create if you are interested. Take the sigma levels of 2 through 6. Calculate rty on a process that has 1,2,3 ……..100 (you don’t have to do each number to see the effect). Having a six sigma process capability in a process that is complex will keep the rty at something pretty decent. At that time the automotive industry was driving a Cpk of 1.0 and transitioning to 1.33. It may work on the outgoing parts you are willing to receive but from the business standpoint it is still a process with plenty of opportunity.
    Just my opinion.
    Good luck.

    0
    #143990

    Saherngu
    Participant

    When will the idiots here learn to read … Mikel’s theory states “long term” is 25 samples … that is about 1 day for hourly samples.

    0
    #143992

    Orang_Utan
    Participant

    A white paper written by a former Motorola VP mentions that the historical long- term shift of 1.4 to 1.6 std dev was found, not 1 to 2 sigma as you highlighted.3.4 DPMO is a long-term goal, not short-term. Short-term goal shall be 2 DPB(billion)O.

    0
    #143994

    Orang_Utan
    Participant

    A bunch of ex-Motorola staff who involved directly in Six Sigma since 80s are still around. I am still talking to many faces show in an old photo at Mario’s website.Motorola Malaysia ex-GM, Roger Bertesen published a white paper before he retired. That paper is a first hand account about the birth of six sigma.

    0
    #143995

    Saherngu
    Participant

    Read your history !!!  
    It was Motorola’s Bill Smith who saw “sudden shifts” not “long term” ones.  Who knows why he didn’t recognise them as special causes ?

    0
    #143996

    Orang_Utan
    Participant

    Unfortunately Bill Smith did not leave any trait about your shift story. Many conjectures about shift are not from him.

    0
    #143998

    Mike Carnell
    Participant

    Your point is?

    0
    #144001

    Orang_Utan
    Participant

    My point is I trust more on Motorola ex-GM’s written record on six sigma historical facts than those self-proclaimed Motorola six sigma insiders.

    0
    #144007

    Cynic
    Participant

    It is sad to see there is no clarity of thoughts among the “self proclaimed six sigma experts” here…who dares to call others Idiots but cant provide a clear thought. Even when someone tries to clear doubts others come up and make it complicated with jargons. Who cares what Bill Smith said and What Motorola’s VP wrote…its important to understand the Issue!!!…The initial question, I think, was why Six Sigma and why not 5 or 3 Sigma.
    Can anyone try to give a simple, logical response to this. If one doesnt know the answer, it would be better to keep quiet and stop trying to act knowledgable.
    I dont know the answer…..but very keen to know it…..
    Regards,
    Cynic
     
     

    0
    #144011

    Mike Carnell
    Participant

    Cynic,
    If you do the rolled throughput yield exercise it makes it very clear where the leverage comes from in terms of what sigma level you need to keep aprocess from wasting a great deal of money on rework.
    Good luck

    0
    #144013

    Brit
    Participant

    Cyninc:
    Orang-Utan and Mike’s comments answered it in an earlier post.  The Japanese were smacking the US in many american markets.  Based on the sigma measure, our competitors were running at around 4-5 sigma.  Motorola chose 6 sigma as a goal to challenge their operations to beat them (and everyone else for that matter).  Orang_utan expresses this and Mike follows up with the capability justification.
    Not sure where you’re missing the “clear thought”
    After the 4th or 5th post, the rest is commentary based on history, which, in itself can be valuable if you care. If you don’t, then you probably can move on to another thread.  I hope you will find value – many posters try to do their best.

    0
    #144038

    Orang_Utan
    Participant

    Seems you have no real capability to comprehend information in the public forum.Do you own research at your uviversity library or approach Motorola directly if you are serious about this matter.

    0
Viewing 18 posts - 1 through 18 (of 18 total)

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