It's not difficult to read Deming's intentions about what he always calls
the Shewhart cycle (although he admits that it is widely known as the
Deming cycle.) For example, in Out of the Crisis (p. 88) he describes it
this way:
"1. What could be the most important accomplishments of this team? What
changes might be desirable? What data are available? Are new
observations needed? If yes, plan a change or test. Decide how to use
the observations.
2. Carry out the change or test decided upon, preferably on a small scale.
3. Observe the effects of the change or test.
4. Study the results. What did we learn? What can we predict?
5. Repeat Step 1, with knowledge accumulated.
6. Repeat Step 2, and onward."
Later he writes: "Step 4 of the Shewhart cycle (study the results; what
did we learn from the change?) will lead (a) to improvement of any stage,
and (b) to better satisfaction of the customer for that stage. The
results may of course indicate no change at all, at least for now."
By the way, the word "stage" is used because he's describing the Shewhart
cycle in conjunction with a discussion about how "Every activity, every
job is part of a process. A flow diagram of any process will divide the
work into stages." He continues to write: "At every stage there will be:
Production -- change of state, input changes to output. something
happens to material or papers that come into any stage. They go out in a
different state.
Continual improvement of methods and procedures, aimed at better
satisfaction of the customer (user) at the next stage."
A final comment: I endorse José Antonio Espín's interpretation of the
Shewhart/Deming cycle as including, via Step 4, double-loop learning. In
fact I believe that's the main point of the cycle.
And just in case it's still not clear from Deming's own words:
"Do" (Step 2) means carry out a "test or change" (i.e., do an experiment)
"Act" (Step 4 -- the double-loop learning step) means make a decision
based on observing what happened when you ran the experiment. This "Act"
can only be one of four things: (a) implement a change based on what you
learned from testing (usually on a wider scale; (b) go back to doing
things the way you have been doing them; or (c) try another test (a
different change, or the same change under different conditions.)
John
--"John Gunkler" <jgunkler@sprintmail.com>
Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>