Re: Semantics and 6 Sigma

[GrantBlair@aol.com]

In a message dated 10/12/2000 12:58:46 PM Eastern Daylight Time, 
mtribus@home.com writes:

<< Shewhart's control chart plays the same role as entropy.  When a 
 process is in steady state, it will yield control charts which obey 
 very simple rules, it becomes predictable.  However, if the process 
 is not in a steady state but is perturbed by various 'causes' (known 
 and unknown) the chart will not follow the simple rules, it becomes 
 harder to make useful predictions >>
I've seen the terms "Predictable" and "Unpredictable" used before regarding
control charts, but it appears to me the definitions: 
common cause=predictable,
special cause=unpredictable 
are backwards, providing we are talking about a shift in the mean and not an
increase in common cause (short-term) variation...this is why Range and 
Moving Range charts are so important.
Since we're not in a classroom, you'll have to pretend we're in one, and I 
have
a flip chart, pen, and two dice (1-12).
Suppose I were to roll the dice seven times and plotted these points. Now, I 
say: "Quickly... take my pen and plot the next point" ...It would be very 
unlikely
you could pick the next number I would roll . (If you could do this for the 
next 7 points, we would both fly to Vegas the next day };->
Now, suppose I plotted the following numbers on my chart, 1.0, 2.0, 3.0, 4.0, 
5.0,
6.0, 7.0 and said "Quick..take my pen and plot the next point". EVERY person
in the classroom would immediately plot the point at 8.0.
Now, suppose I rolled the dice, plotted the result as 20 and asked  what 
everyone would do if this were on a control chart. They would examine the 
dice, then ask
me if I really did get this number. (I would reply.."yes, after I added 10 to 
the dice").
Now the point of this exercise is:
1. Common cause variation is unpredictable. It behaves the same as Shewhart's
    Bowl
2. Special Cause is predictable, even within control limits. This is why 
trend rules 
    work.
3. Special Cause is also a signal something is going on. We can PREDICT, with
    high certainty, a search will find the condition. (or adjust, worst
    case when you see 20 is to subtract 8, then see what happens)
Now, it isn't this easy in the real world, because Special Cause sits on top 
of
Common Cause...this creates uncertainty in our action on the process. 
But, in a practical sense, It also means that if I see a special cause on a 
chart
measuring toothpicks made from matchsticks, it is certain I will not find a 
2x4 causing the problem...This is why 6 sigma (Personally, I'm a 9 sigma
person };-)..works so well.
Where does my thinking go wrong here?

GrantBlair@aol.com
Ninety Six SC