Re: Knowledge and Variation: Some thoughts on Deming and Fisher

[Jonathan Siegel <jmsiegel.info.research@worldnet.att.net>]

I appreciate your comments.

I'm suggesting that Deming's approach involved much more than a doctrine of
continuous improvement towards fixed goals -- "on target with minimum variation" --
under stable conditions in things like product metric characteristics.

I think Deming's comments about carburetors vs. fuel injectors are a good example of
this. A company that focuses on continuously improving the carburator can get put out
of business when a new invention discontinuously replaces the whole thing. I think
Deming was suggesting that the world is like that; that this is usually what
eventually happens. Continuous improvement under stable conditions is only a tempory
kind of change. System dynamics are not  always continuous. Like a genetic mutation,
a new invention involves a discontinuous adaptive change. No matter how sound the
theory behind it, it inevitably involves a leap into the  unknown that carries a risk
of failure or unexpected consequences. I believe that Deming sought to foster safer
and more effective "new invention" kinds of change as much as he sought to foster
better "continuous improvement" change. A company which is so focused on carburetors
it knows nothing else will have no intellectual resources with which to grapple with
the challenge of an altered world.

Although his remarks about continuous change have found a more receptive audience,
Deming talked quite explicitly about discontinuous change. Continuous improvement can
itself precipitate unanticipated discontinuous change -- as occurred, for example,
when trade surpluses wrought by Japanese quality efforts had effects on exchange
rates, or when well-drilling and bovine health measures resulted in overgrazing and
encroaching desert in sub-Saharan Africa. We and our customers exist in a context. We
are not free to set our aims and ignore all else as if we acted in
isolation. The more interrelated the world becomes, the more discontinuous
consequences become the rule rather than the exception.

Jonathan Siegel