Re: How To Compare Two Processes?

[Kromkowski@aol.com]

In a response to a message from WMack...

OK, here we have a case study.  

AIM: Surely, the DEN can collectively come up with along with WMack, 
something that would "look like" it could have come from OOTC or TNE as an 
example of how good management would proceed.

How would we know we have succeeded with the above AIM?:

I am speaking in broad brushes here, so let's not get so picky that we fail 
to help WMack.  It seems to me that a fulfillment of the stated AIM would 
have the following things, that are generally present in WED's case studies:

1.  (STATEMENT OF THE ISSUE) A concise statement of the general issue 
understandable by a layman. 

2.  (STATEMENT OF THE AVAILABLE DATA) A statement of the available data in 
relation to the issue. 

3.  (CONTROL CHARTS USED AND CONCLUSIONS DRAWN THEREFROM) A presentation, 
where possible, of a control chart or charts, 
which are (a) (WHICH CHARTS AND WHY) appropriate for the kind of data,  
which (b) (CALCULATION OF LIMITS) show how the limits are calculated, 
which (c) makes a determination of whether or not 
    (i) (STATISTICAL STABILITY AND EXPECTABLE VARIATION) there is (what might 
be reasonable concluded to be) a generally stable system or process which is 
producing the data
    (ii) (SPECIAL CAUSES OF VARIATION) and which provides are basis for 
telling us whether there are signals suggesting that one look for a special 
cause of variation.

4.  (COMMON CAUSES OF VARIATION) An identification either in a general list 
or by fishbone chart or in a linguistic equivalent to a fishbone because of 
the limitations of the DEN, of the likely common causes for the variation in 
the data, and the upstream sources or producers of inputs to the system or 
process at issue.

5.  (EFFORTS TO REMOVE OR EXPLAIN SPECIAL CAUSES OF VARIATION) If the search 
for special causes is warranted, a proposal (theory) and actual action for 
removing them from happening again if feasible, and evidence that they have 
been removed, or an explanation (if not feasible to remove, e.g., if the 
special cause was an earthquake or the like).

6.  (THE RELATION TO DOWNSTREAM CUSTOMERS AND LARGER SYSTEMS) An 
identification of the downstream customers or effects of the process or 
system producing the data as well as some statement as to the effect on the 
larger systems within the reasonable control of the people with power to do 
anything about it.

6.  (THEORIES FOR SYSTEMIC IMPROVEMENT) A theory, in light of the list of 
common causes, that will lead to a plan that will lead to better centering of 
the data with less variation.

7.  (TESTING THEORIES FOR SYSTEMIC IMPROVEMENT) A test of the theory of 
improvement, preferable on a small scale. 

8.  (POST-TEST DATA) Collection of post-test of improvement data, as evidence 
for support or rejection of theory.

9.   (DECISION-MAKING) Determination whether the theory of improvement was 
correct and if so whether it makes economic sense to implement theory of 
improvement and if not, a new theory of improvement or a decision to accept 
stable system/process producing the data (i.e., leaving things in control 
alone and not tampering.)

10.  (FUTURE PLANS) A plan for future evaluation improvement or preventative 
improvement which might likely include a determination of whether 100% 
inspection or sampling is best method of future data collection,(see OOTC).

******************
If we can generally agree on above, and as the "price" for DEN help, WMack 
agrees to provide the data and an ongoing account of the efforts using for 
clarity purposes the section headings I have proposed, then we will have 
something that is valuable for us. 

As such, I hope the moderator and the DEN will be vigilant in seeing that 
each of the 10 items have been documented and answered and collected and that 
WMack re-presents it to the DEN as one complete document.

I'd answer the questions, but I am sure others will and I am more interested 
in encouraging an approach that will have future usefulness.

John David Kromkowski
Kromkowski@aol.com

[Moderator's Note:  I certainly will support and process this line of thinking.
I can also be vigilant.  The balance is up to the subscribers.  It *would*
be a good test of the system we have here to collaborate in this manner.
Thanks for the challenge, JD.
=============================================================================