iSixSigma

Six Sigma – attacking variation or mean and variation

Six Sigma – iSixSigma Forums General Forums General Six Sigma – attacking variation or mean and variation

Viewing 14 posts - 1 through 14 (of 14 total)
  • Author
    Posts
  • #53412

    rams
    Participant

    Literature emphasizing how six sigma focusses on reducing variation. However there are a number of tools that focus on improving the capability by moving the mean of the process. Brainstorming tools like QFD and Fishbone, help us to identify solutions that can alter the mean of the process. 2 t tests and anova are also used to see if the mean of the process has changed. Given this is it not incorrect to say DMAIC Six Sigma focusses on variation?

    0
    #190021

    Hody
    Member

    In the lore of Six Sigma, to improve the capability of processes, sometimes you need to reduce the variance, sometimes you need to shift the mean, sometimes you need to fix both. Mikel Harry stated from the start that “Process Variation” may be caused by poor “accuracy” (mean) and/or poor “precision” (variance). It’s a bit of a poor phrasing since process “variation” is not meant to be exactly the same as “variance”…

    I hope that helps…

    0
    #190022

    Murugan
    Member

    Hi,

    Which one would you think will be more beneficial “reducing the variation” or “moving the mean towards the target” ?

    0
    #190026

    Mikel
    Member

    Taguchi has it right. An off target mean generates a lot of variation. Read about Taguchi’s loss function.

    0
    #190035

    KKN
    Participant

    You have to look at both the effectiveness and the cost of improving the customer experience. Sometimes you can get there by shifting just the mean and sometimes just reducing the variation. It all comes down to what has the best business case.

    0
    #190036

    rams
    Participant

    Dear Sensei,

    Thank you for clarifying the difference between variation and variance. So if a process is precise but not accurate, in six sigma parlance this would still be called variation. Got it.

    0
    #190046

    Mike Carnell
    Participant

    Rams,

    If you want to understand the concept of ‘Breakthrough” then you need to go to the original literature which is Juran. In his book “Managerial Breakthrough” he explains the difference between Breakthrough and Control. Breakthrough is by definition dynamic change and the model he depicts shows shifting the mean to the target (it would be completely stupid to focus on variation reduction if your mean was off target). This makes Juran and Taguchi (as Stan said) consistent with each other.

    Control by definition is lack of change and that is variation reduction.

    https://www.isixsigma.com/index.php?option=com_k2&view=item&id=1221&Itemid=1&Itemid=1

    At the end of the day who care who wrote what. A six sigma project is about fixing the process. If that process isn’t on target are you going to leave it that way because some wrote something in a book that said Six Sigma was variation reduction? When you see the comment that SS is variation reduction that is your first clue that that person doesn’t have a clue. SS is about results.

    Just my opinion.

    0
    #190050

    Leader
    Participant

    I would recommend reducing the variation first; then you will have a good idea on how inaccurate you are. Moving a tightly controlled process to correct for bias should be easier than trying to correct an imprecise process for bias, the reducing variation.

    0
    #190056

    Murugan
    Member

    Thanks All, I have got a fair idea now on the priorities

    0
    #190092

    James Considine
    Participant

    Given the choice, go after variation first – generally your mean will shift automatically

    0
    #190093

    Mikel
    Member

    What nonsense.

    0
    #190094

    MBBinWI
    Participant

    Oh, Stan, tsk, tsk, you know better than that.

    0
    #190099

    Mikel
    Member

    You are right, I forgot about those magic means that center themselves on target.

    Part of that six sigma magic. No real work, just a bunch of theoretical blow hards teaching classes.

    0
    #190103

    Craig
    Participant

    Improved Cpk from my experience has been achieved by minimizing variation.
    This is done through process characterization and optimization, controlling input variables, etc. No simple task!

    Although a mean shift towards target can improve Cpk, I have rarely found processes with these types of quick fixes. (Although one of my projects where this was true was the highest impact change I have implemented)

    Regardless, Six Sigma focuses on both.
    (reducing variation and shifting the mean where necessary)

    0
Viewing 14 posts - 1 through 14 (of 14 total)

You must be logged in to reply to this topic.