TUESDAY, NOVEMBER 25, 2014
Font Size
Topic 6Sigma vs. Shainin Red X

6Sigma vs. Shainin Red X

Home Forums Old Forums General 6Sigma vs. Shainin Red X

This topic contains 6 replies, has 1 voice, and was last updated by Profile photo of BC BC 7 years, 5 months ago.

Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
    Posts
  • #68325

    Can anyone help me about identifying the differences between Shainin RedX Methodologies and Six Sigma ?

    #127905

    The samples in Six Sigma DMAIC must be statistically relevant or stable. This almost always a large sample size. The conclusion from this data are driven by the statistics and not the engineering or physics of function.
    The Shainin FACTUAL requires contrast. sometimes with only one part WOW and the design intent BOB. This is based on the singular function it does not perform. The simplification of the statistics allows for a 95% confidence with small sample sizes in confirmation. The logic is driven by the engineering and physics this actually makes the problem easier.

    #127912

    Man! You really took a big hit of the Shanin koolaid, didn’t you?
    You don’t have a clue what you are talking about. All of Shainin’s techniques are statistically valid and require the same size sample as you would with SS given the same assumptions.

    #127963

    I think it is you that does not know what your talking about. Shainins techniques leverage the ability to distinguish BOBs and WOWs with as little as 2 samples, those who accomplish the most with the least, the fastes…win!  Again start with two parts: BOB and WOW rather than a larger sample size which could be wasteful.  If you fail to get diserable feedback with those two you do not need to continue that route.  Most things we attempt to control do not have much influence over what the customer cares about.

    #127965

    Christian,I am not going to get into a pixxing contest with you. You are
    passionate about your approach.BOB’s and WOW’s assume huge differences, when you go look up a
    sample size needed for differences such as that, you will find the
    sample size is 1.I knew Dorian. I also had the opportunity to be part of
    conversations where satisticians tried to debunk his methods, but
    what the statisicains learned as the conversations progressed is
    that Dorian’s methods were well founded in stats, he just did not
    tell people his assumptions. What I object to is the way Shainin is taught as a dogma – we are
    the way, we are the only way – and the insistance on the strange
    lanuage around the method.You work in the automotive world of OEM’s and Tier 1’s – what we
    both know about this world is the whole PPAP, process cability,
    gauge capabilty, … is a joke. It is not a joke because the ideas are
    not right, but because most of what is done there is a lie. Your
    method, my method, Taguchi’s methods, Kepner Tregoe, whatever
    will work if supported in your environment and they aren’t.That is what is important, why don’t we talk about that with the
    leaders who are driving these key industries deeper and deeper
    into the ground?

    #127970

    ok, ok, I hear ya and agree.  I posted a msg agreeing with you on shainins approach in our 5 other threads. :)
    Your point is heard on up front engineering. Too many times do we have to validate the design and process in the assembly plant.  I guess that waste keeps us reactive problem solvers in business.

    #127985

    Stan,
    I too live in the world of automotive Tier 1’s and OEM’s.  Wanna hear a good joke?
    I was actually asked, what is the sigma level if my Ppk is 3.8?
    In my best English, “ain’t no way”.  Unless, of course, we can apply the 8.5-sigma shift.
    BC

Viewing 7 posts - 1 through 7 (of 7 total)

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

Login Form