Nonstatistical improvement

[Kromkowski@aol.com]

Anton noted:  >>My point was to echo Deming's point that really we are 
talking about two issues:  1) process and system improvement, and 2) quality 
management.<<

This is where I disagree.  There is only one issue in my opinion and I think 
in my reading of Deming -- "Thoughtful Management".

>From p. 20, Chapter 2, OOTC:

"1.  The central problem of management in all its aspects, including 
planning, procurement, manufacturing, research, sales, personnel, accounting, 
and law, is to understand better the meaning of variation, and to extract the 
information contained in variation."

>From p. 309, Chapter 11, OOTC:

"The central problem in management and in leadership, in the words of my 
colleague Lloyd S. Nelson, is failure to understand the information in 
variation."

I can't understand why Deming would put this in twice, if he wasn't sure by 
what was meant by the word "central".

I have absolutely doubt about the soundness of Deming's statement at p. 26,  
"Top management should publish a resolution that no one will lose his job for 
contribution to quality and productivity", but let me assure you that 
notwithstanding a state of maximum input from employees, it remains the 
responsibility of Management to decide whether, to what extent, and how any 
"great idea" gets tested and put into practice.  This is purportedly why they 
get salaries (although I cannot make any sense of it at all except the 
preservation of the rich elite ruling class who are certainly not 
meaningfully smarter than the rest of us), which are uniformly outside the 
"system" of pay for the rest of the employees in the company.

Deming was not about tearing down the tyranny of management, he was about 
tearing down the tyranny of bad management.

>>If you can shift the way management thinks about people, about systems, and 
about aim, purpose, etc. *then* statistical improvement is a valuable tool in 
the armamentarium of the manager/executive.<< 

Most poor "thinking about people" is driven by a failure to understand 
variation -- a) that everyone is different and b) much of the variation is 
not worthy of treatment as a special cause.

Understanding variation is the first opportunity to understand "a system".  
First, the control limits operationalize the system, they suggest a boundary. 
 The search for "common causes" must lead one to thinking about interactions 
and components to variation, and to what is downstream and what is upstream.  

Understanding variation leads one to the idea of the possibility of 
prediction (which is what the theory of knowledge is about).

You say if A (we can shift management thinking about certain things), the B 
(understanding variation is a good tool).

I believe that D's writings suggest something different:

If B (management understands variation), then A (there will follow a 
different way of thinking).

and 

if not A, isn't at least B better than not B?

and

How is again that we are going to "shift the way management thinks about 
people, about systems, and about aim, purpose, etc." as you put it?

JDKromkowski
======================================================================