iSixSigma

The 1.5 Shift – Time For A Paradigm Shift?

Years ago (almost ten now!) when I was going through Black Belt training, I remember seeing the famous slide describing what a three-sigma world would look like. The presentation slide described how three-sigma aircraft landing performance would mean two long or short landings per day, and that 20,000 articles of mail would be lost per day at a three-sigma level. After completing the presentation, an astute participant in the class asked why 3.4 DPMO was described as six-sigma performance… to him, it seemed like a high level of defects for a true six-sigma process.

All statistical purists know this is the case, but the instructor started describing how the 1.5 shift and drift effect degrades performance over time, and that this number was used based on historical performance, etc….

Since I’ve been in a position to coach various individuals in six sigma concepts for some time now, I have to admit that each time I describe the 1.5 shift it gets more and more frustrating. Here’s why:

  • The 1.5 shift doesn’t really hold over time in all cases, so it can be a poor approximation. I’ve seen long-term performance influenced by much less than a 1.5 shift and drift factor (sometimes maybe more like 0.5). Likewise, I’ve seen much worse (maybe at 2 or 2.5).
  • I can’t honestly say that from my experience the shift factor distribution is normal, so therefore I can’t predict it ;).
Handpicked Content :   A Spoonful of Sugar

Based on this, here’s my proposal:

  • Get rid of the 1.5 shift factor in training, and explain only what long term shifts in the processes do to process performance.
  • As a second step, explain how to calculate a short-term and long-term sigma value from a process.
  • Lastly, make a “continuous improvement” metric out of getting the long-term process sigma value closer to short-term sigma performance levels (minimizing process shifts).

At the very least, the above methodology should make the concept of “long-term shift” more understandable to new practitioners, and make it a practical tool as well.

You Might Also Like

Comments 13

  1. Sowmyan

    Beyond a point conventions take over to make expressions from every one consistent.

    The 1.5 sigma shift concept enables a discussion point. It gives an opportunity to discuss long term and short term sigma.

    Likewise, we also have the alpha factor of 5% being applied for a wide range of hypothesis testing based methods. Depending on the nature of the problem and a desire to "fail safe", people may even choose an alpha value that is higher. But 5% is convention / default value for most analysis and discussion.

    Since these conventions have got established over a long time, but are not misleading, they may remain.

    0
  2. Ramaswami Viswanathan

    I stopped teaching 1.5 sigma drift 6 years ago. I say to the class (1) what you see in Minitab or Excel is the Z. do not add or subtract. (2) long term is when you have seen all possible X factors, and all combinations of such Xs. which you would never know, theoretically. (3) Practically if a term (time period, say 1 year) looks like another (statiscially that is), then it is the long term. (4) long term could be different for different industries, and within the same industry, different for different firms, and within the same firm, different for different processes. for example, a week looks like (statistically) another for a hospital (in terms of patients arriving or cycle time) while days within the week do not, so perhaps, a week is a long term, and a day short term, assuming there are no month-end factors or month to month variation (5) all processes need not drift. if you continuously improve a process, why would it shift & drift?

    0
  3. Kosta Chingas

    Hi Sowmayan….you do have points here. I agree that the discussion needs to be had between long and short term sigma. I have seen students become confused becuase they start thinking that they need to see a 1.5 long term shift in performance based on the initial definitions given in the training. However the processes they are dealing with may not have more or less of a shift factor, thus the students become confused, asking "why am I not seeing a 1.5 long term shift factor here?"…

    The 5% alpha rule is an adjustable guildeline that we can modify based on risk tolerance. The shift factor is a characteristic of the process that can be influenced with process controls and variation reduction.

    0
  4. Kosta Chingas

    Hi Ramaswami – my opinion on #5 is that in most cases long-term variation will be greater than short-term. Most processes are not only complex, but have multiple supplier inputs that are subject to shift and drift, thus shifting and drifting the overall process.

    From a continuous improvement standpoint, after going through an optimization activity, the process shift and drift may be drastically reduced, but it won’t be eliminated for the long term. The continuous improvement opportunity is to reduce the drift to the point where long-term equals short term (no drift at all).

    0
  5. Vinod Bhardwaj

    It is indeed very difficult to explain the 1.5 sigma shift in a lecture of 6 sigma unless you have a fairly good experience of running and controlling a continous process in an industry. Good knowlege of SPC and on hands experience is a must to understand the phenomenon. After all Motorola had some logic in prescribing the shift.

    0
  6. Kosta Chingas

    Hi Venod –
    Absolutely…100% agree, but I’d like to do what Ramaswami did, and get rid of the 1.5 number completely…you’re right about the SPC..that background would help.

    0
  7. Marty Y.

    Here, here. . .

    I resolve to eliminate the Sigma shift from future training! I can’t count the number of times this has bogged down the conversation and moved it into the trivial realm that only reduces learning and correct application of the concepts.

    0
  8. Kosta Chingas

    Marty….let’s do it. How do we standardize that?

    0
  9. Vinod Bhardwaj

    Hi Kosta,
    you have asked a million dollar question. standertization is a big challange. I have faced the situation Marty has expressed in my training sessions. How should we standertize?

    0
  10. Kosta Chingas

    Vinod/Marty

    Sounds like a topic that needs more discussion for sure….it would be really advantageous if there was a "six sigma" standardization group of some kind.

    Maybe we should set one up!

    Kosta

    0
  11. Marty Yuzwa

    Kosta,

    I would think that the first step would be to create a few slides and charts showing the sigma level without the shift. Then we could go from there.

    Here’s my e-mail.

    I’m in Linked In. Let’s go ahead and connect, and we can talk more.

    marty

    0
  12. Marty Yuzwa

    And Here’s the e-mail I forgot to post:

    yuzwam@sbcglobal.net

    0

Leave a Reply