iSixSigma

what does six mean in six sigma

Six Sigma – iSixSigma Forums Old Forums General what does six mean in six sigma

Viewing 27 posts - 1 through 27 (of 27 total)
  • Author
    Posts
  • #41816

    sneha
    Member

    hi
    pls let me know what six mean in six sigma. Normally what I can see is +3 or -3 sigma on each side of target.
    Thanks. 
     

    0
    #131558

    RDogg
    Participant

    Six represents the positive integer that is greater than 5 but less than 7

    0
    #131561

    Brit
    Participant

    This is a very basic description, so please, all the stat nerds out there cut me some slack (I’m one too). There are much longer responses. Sounds like you are fairly new to 6sigma, so I don’t want to get into issues like sigma shift or other various distribution/probability explanations.  Please search the isxsigma site for these if you want more – it provides more detailed information.
    When a process is being measured, we can normally expect to see values fall within 3 standard deviations of the mean and a control limit.  For six sigma, you are improving a process to meet a target more often, either by shifting the mean toward the target and/or reducing the standard deviation to meet the target more often.  Sigma in Six Sigma is a measure of probability that you will hit a target/goal. When we talk about sigma level, we are talking about target and spec, not mean and control limit.  A common mistake is to confuse the sigma level to the engineering specs of the normal curve (e.g., x-bar +/- (3*std dev)). 
    If you can fit 6 standard deviations between the target and an upper specification (one side), you are operating at a 6 sigma level.  At this level, you would expect only 3.4 defects per million opportunities to occur outside of one of your spec limits.
    For example, a process has a target of 10, an upper spec limit of 16,  and a std deviation of 2.  Say your process mean is meeting your target of 10.  In this instance, you could fit 3 std deviations between 10 and 16 (i.e., 10+(3*2)=16). The process is operating at 3 Sigma.  If you reduce the std deviation to 1 (and the mean stays the same), you would be operating at a 6 sigma level (10+(6*1)= 16).
    Again – very basic explanation.  My guess is that you will get a few responses to my text, arguing how I’m off the mark a bit.  The explanation above is the simplest I can make.  Good Luck!

    0
    #131566

    Shinjini
    Member

    Hi,
    It was found from statistical studies that in a normal distribution graph(a bell shape one) M +/- ó (where M denotes the mean) covers 66% of data points between USL and LSL.
    M +/- 2ó covers 96% of data points between USL and LSL
    M +/- 3ó covers 99.73% of data points between USL and LSL
    So taking a safety factor of 2 , it was emperically found that there will only be 2 data points ( out of a million ) if we consider M +/- 6ó thereby accomodating 12 standard deviations (ó) between the USL and the LSL…
    This is how the “6” came historically.
    Above, USL means Upper Specification Limit and LSL means Lower Specification Limit which are given by the customer.
    Thanks & Regards,
    Shinjini

    0
    #131592

    Kumar
    Participant

    Sneha,
    As you told, since we have +3sigma in one side and -3sigma in the other side, totally the spread is 6 sigma and hence we are calling as 6 sigma. If you need further info, please contact me in [email protected]

    0
    #131593

    Bruce Hudson
    Participant

    You must be working for Microsoft since you provided accurate but usesless information. 

    0
    #131595

    Kumar
    Participant

    Sneha,
    Sorry. it is +6sigma on one side and -6sigma on another side. Hence the name as 6sigma.

    0
    #131597

    Darth”s Sister
    Participant

    Wrong  Interpretation:It  should  6  sigma at  both  sides (please revise  your  knowlege again ,it  is  +/- 6  sigma  and  not +/- 3  sigma)??

    0
    #131598

    Kumar
    Participant

    Thanks for your correction. But the correction is already posted by me.

    0
    #131604

    Deepa Vegesina
    Participant

    hello ppl,
    its very easy to get confused with the six….
    well for the start its not 6 sigma on both sides….
    plz do remember the sigma level is determined by the no.of sigmas one can fit between the mean and the specs
    now having said that… let me get down with the nitty gritty of it….
    1. a process can have two specs the USL & LSL ‘or’ just one spec limit – either USL or LSL.
    2. when there r 2 spec limits, its a usual tradition to take(or want) the mean to be at the centre of these specs.and when that happens you have 3 sigmas to either side of the mean which makes up the 6 sigmas.
    3.when there is only one spec limit all the sigmas (6)are one side of the mean.
    well comments r welcome
    adios
    deepa

    0
    #131605

    Flying Sigma Circus
    Participant

    You are correct – it is easy to get confused and you are.  The following quote is taken directly from this website and it took about one minute to find.
    “For a process with only one specification limit (Upper or Lower), this results in six process standard deviations between the mean of the process and the customer’s specification limit (hence, 6 Sigma). For a process with two specification limits (Upper and Lower), this translates to slightly more than six process standard deviations between the mean and each specification limit such that the total defect rate corresponds to equivalent of six process standard deviations.”

    0
    #131606

    Brit
    Participant

    Please go to the blue bar on the left and click New to Six Sigma.  There is a section called the Statistical Definition of Six Sigma.  Unfortunately, someone asking for new information in a discussion forum is getting incorrect answers.  I hope some people responding aren’t black belts.
    The value is having 6 standard deviation s fit between the mean/target and a spec limit (either spec limit).  Acheiving this would provide a probability of a good product at 99.99966%, much better than the competition.  If the process has 2 spec limits, you still call it a six sigma goal, however, acheiving it would show 12 sigmas on either side of the mean (i.e., 3.4 defects per million or 99.99966% good).
    If you have 2 specs on either side and your process measures fall within +/- 3 sigmas on both sides of the mean, you are running at 99.73% good or roughly 2300 DPMO – long term (look at any normal distribution comparison of +/- 1, 2 &3 sigma).  A process running between +/- 3 standard deviations would be running a 4.28 Sigma Level (including the 1.5 sigma shift accounting for long term variation), not 6 sigma.  Try this with the sigma calculator on isixsigma (27 defects out of 10,000 opportunities).

    0
    #131938

    Nageshchandraa Adhav
    Participant

    I’ll be thankful if you could kindly inform me about the subject matter at your earliest. I look forward to recieve the same.
    Kind regards,
    Nageshchandra Adhav.
     
     

    0
    #131939

    RDogg
    Participant

    I believe that says something about the question…

    0
    #139413

    Dilton Davila
    Participant

    Six Sigma intent to aply a value of defects this can change on multiples numbers and values for example a (6 sigma)= .000001 errors on the procecing or aplication of the systems or it will be an equal of 3.4 of a million units, six sigma messures defects or non compliace : example performing at a level of 3.4 defects per millon oportunities. the value will change depending on the first value before the sigma the lower the number the higher the defects allowed in the system.

    0
    #139414

    Heebeegeebee BB
    Participant

    Amateur!This thread is only six months old…The record stands at a 4-year old thread response!Sigma relates to SD.   6 Sigma = 3.4 DPMO=Cp 1.5=Cpk 2.0=99.9666666% yieldThen there is that pesky “shift”.

    0
    #139415

    Dilton Davila
    Participant

    thanks for that but I all ready know it and sorry if I mised your intellectual, was only a brief on the subject but if you are looking for more depth on it, email me and we put it to the text.

    0
    #139416

    Heebeegeebee BB
    Participant

    All your base belong to us!

    0
    #139429

    technicality
    Member

    Well, technically that’s not correct.   The land occupied by forward placed bases still belongs to the country in which they are placed.   There is however a practical application of ownership in that, while legally still a long-term tenant with an open-ended lease, said tenant is equipped and ever increasingly willing to drop large bunker-busting bombs on your bunker.  See, it’s the yin and yang of befriending the mighty.  

    0
    #139453

    Heebeegeebee BB
    Participant

    Nice one technicality…
    Also, I botched the line…
    It should read,
    “ALL YOUR BASE ARE BELONG TO US…”

    0
    #139456

    Ken Feldman
    Participant

    HeBe, I will check for the full moon tonight.

    0
    #139460

    Heebeegeebee BB
    Participant

    Darth,
    No doubt…Man it has already started.. didja see what Anderjad (sp?) posted about time studies???
    Someone is hittin’ the crack pipe a bit too much…
    OOOOOOOWWWWWWWWWWWW—OOOOOOOOOOO!!
    Full moon, indeed!

    0
    #139465

    BTDT
    Participant

    YOU ARE ON THE WAY TO DESTRUCTIONCheers, BTDT

    0
    #139478

    Ruddy
    Participant

    Mr. HeeBee “Wanna Be” – I rarely read these threads… seems you and others have managed to take a once quality-laden site and brought it down to a level close to your collective IQs.
    It’s a shame that those attempting to better themselves should need to put up with your demeaning and racist comments.  As I asked of you several months ago, why don’t you have the intestinal fortitude to use your real name. 
    Believe me, you’re only hiding your name, not your ignorance.
    Oh, and by the way… 6Sigma=Cp=1.5=Cpk=2???  Since when?  You’re riding a wave that’s about to crash – enjoy it.

    0
    #139617

    Heebeegeebee BB
    Participant

    Please, by all means, show us the light Michael!   Where is your data?
    If you whip out the race card, you better have the data to back it up…more than one lawsuit has been generated as a result of a baseless, unsupported slur posted on this very website.   The Webmaster logs and records IP addresses, so make sure you have some concrete proof.   Libel and slander suits could be very costly to all those involved.   Yes, I am being soberingly serious.  
    As for using a “nom de plume”, the internet can be a very scary place, fraught with peril and the like.   Having been the victim of identity theft, it is in my best interest to limit the extent of info on the web.   Call me chicken, or call me coward, whatever makes you feel better about yourself…

    0
    #139619

    Heebeegeebee BB
    Participant

    Oh me-oh-my, Michael did manage to smack me down on the Cpk/Cp values I posted earlier in the thread.   I did have a dyslexic moment and swapped the values…my apologies!
    As for the Race card slur, put up or shut up Mikie.  

    0
    #139994

    howe
    Participant

    You make idiotic comments, too… but I never accused you of being an idiot. Bring on the lawyers, pal.
    I’ll check back in a year or so.
     
    Have a nice Christmas.
     

    0
Viewing 27 posts - 1 through 27 (of 27 total)

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