iSixSigma

3 sigma six sigma etc etc

Six Sigma – iSixSigma Forums Old Forums General 3 sigma six sigma etc etc

Viewing 17 posts - 1 through 17 (of 17 total)
  • Author
    Posts
  • #41126

    melvin
    Participant

    I have recently undertaken a green belt course at a local University, it was an excellent course.
     
    Going over the notes, I think I either missed a session or the facilitator was confused.
     
    A simple question, why is six sigma a better performance than say 3 sigma, I know it is but can’t reconcile 6 std deviation variance to be better than 3 std deviation variance.  I KNOW what I have just written is wrong but thats what my notes say.   Thanks for the assistance in advance.
     

    0
    #128523

    Mikel
    Member

    Wow, sounds like an excellent course to me.

    0
    #128524

    Art
    Participant

    Sure Does Stanley! – Maybe your highness could even volunteer to be a guest speaker!!  If u don’t have anything nice to say…….you know the rest.
     
     
    Hi to the cats!
     
    Peace out,
     
    Art

    0
    #128525

    Mikel
    Member

    Hi Artey,
    Long time.
    Maybe you want the guys question
    What, you don’t know the answer?
    Figures.

    0
    #128526

    Art
    Participant

    Well Stanley my boy, perhaps it has something to do with number of defects or somthin’ like that?  Your smarter than the rest of us, we all know that.  Good luck in the dungeons and dragons tourney and with the courtship of your on line girlfriend.
     
    PS – I am a huge fan! 

    0
    #128546

    HF Chris
    Participant

    Stan,
    If you had 99% of the people who are attracted to you in one room, how many rejects would be there? If you increased your measurement area to a higher sigma which placed more people in the room that are attracted to you, how many rejects would there be?
    Chris 

    0
    #128547

    Mikel
    Member

    Good to see you know how to spell attact.
    Good question. I suspect they are all rejects.

    0
    #128549

    CT
    Participant

    Bob,
       I see you have been caught in what appears to be a lovers quarrel here. So I’ll answer your question. Sigma level represents Parts Per Million Scrap. 6 Sigma represents 3.4 PPM, while 3 Sigma Represents 66,807 PPM. You must have missed day one in your class, or your professor shouldn’t be teaching.
     GOOD LUCK.
    CT

    0
    #128550

    Mikel
    Member

    CT,
    Bob  thinks he attended an excellent course bt doesn’t understand the basic premise. He missed more than a day or it wasn’t excellent.
    He needs more than the understanding of a z value of 1.5 and 4.5.

    0
    #128554

    CT
    Participant

    Stan,
       Agreed, just gets a little tiring sometimes reading the exchanges yourself, JoeBB, and others leave on here. I would like to think it is all in good fun, but I really dont think it is, I really dont think this is the place for it. Sarcasim is one thing, and everyone here obviously likes a good heated debate, but the name calling, bashing, and belittling of posters is purely adolescent. But I know you guys dont think much of me either, so I leave it at that.
    CT

    0
    #128555

    Mikel
    Member

    Agreed.
    have I ever said anything negative about you?

    0
    #128556

    Mike Carnell
    Participant

    Bob,
    Let me give you the theory. It doesn’t mean I am endorsing it or not endorsing it.
    If I set up a process where I have six std dev between the closest spec and the mean in a short term measurement then the process will run in the parts per billion. The controversial 1.5 sigma shift kicks in over the long term and may reduce that capability to 4.5 sigma or 3.4 ppm.
    That is the theory. I have no interest in getting into any discussion about the 1.5 sigma shift, if I do or do not believe it, etc. It is a waste of time.
    I do agree with Stan – this is a basic piece of understanding for the methodology. This is not a piece they should have missed.
    Just my opinion.
    Good luck

    0
    #128560

    CT
    Participant

    Stan,
    Not that I recall, directly, no. I was just pointing out that Bob, although had a pretty elementary question, didnt deserve the exchange that was posted.
    CT

    0
    #128579

    Mikel
    Member

    CT,
    Good. I am usually selective in those I overtly offend and I could not recall thinking you were one I needed to offend.

    0
    #128588

    CT
    Participant

    Stan
    As much as I question your motives, I cant help but admire your sense of humor.
    CT

    0
    #128590

    Ben Royal
    Participant

    Bob,
    Sigma represent the variation in a process in relation to the customer’s specs. If your process performs at a six sigma level, its variation is so small that the product (or service) is 99.9997% defect free.
    Your class handouts may include a sigma conversion chart that shows the defects per million for 1, 2, 3 etc. sigma. My chart I shows that a 3 sigma process has a yield of about 93% defect free.
    If you have a six sigma process, it means you can squeeze 12 standard deviations in between the customer’s upper and lower spec limits (six SDs from the mean to the closest spec limit). The more standard deviations you can squeeze in, the less likely it is that your process will make product that falls outside the customers spec limits, especially if the process mean is centered in the middle of the specification.
    But even if the process isn’t centered, with 12 standard deviations to play around with, it is very unlikely it will make bad parts. So in this case, six is better than three.
    Hope this helps some.

    0
    #128616

    Anakin
    Participant

    Bob,
    1. Draw a normal distribution (bell curve).
    2. Draw a line down the middle (the mean).
    3. Draw two lines, one 2 sd above and one 2 sd below the mean.
    4. Within this range is 95% of the data (this represents your successful items).
    5. Now draw two lines, one 3 sd above and one 3 sd below the mean.
    6. Within this range is 99.73% of your opportunities (again, your successful items).
    From the “Mortimer-Inchobod-Marker” school of statistics this picture should convey the idea as well as any.  Please pay no attention to the language regarding long-term, and short-term variation on this site or the speak of any shift.  From your description of your instructor it would seem that these concepts are foreign (as they probably should be).

    0
Viewing 17 posts - 1 through 17 (of 17 total)

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