Bill wrote:
> America's problems with quality were not due to ignorance of Deming. The
> problems came from a conscious decision by upper management to maximize
>profits at the expense of quality.
I confirm what Bill has written. In the early 1960's as a very young
graduate from U of Michigan I worked in a research lab and tried to employ
very basic statistics in the designing of experiments and process control on
production lines. I was advised that such extra cost was totally unnecessary
because the products were being purchased as they were. The concept of being
able to have the products be uniformly good within a predetermined range
seemed absolutely ridiculous those with whom I spoke. The didn't believe it
could be done or if it could be done it would cost too much to do it.
While an undergraduate at the U of Minnesota, I had had the experience of
working in a small factory as a summer job.While there I analyzed and
eliminated essentially all of the production failures from a process machine
central to all of the factory production of which I was the operator. After
I had done this the plant no longer had to work overtime on Saturdays to
meet production goals -- we finished early on Fridays. After a few weeks it
was apparent that the factory did not need all of the workers it employed.
The foreman, being a shrewd person, decided to fire me. The following week,
having eliminated the guy who could get 98% yield from the key machine, the
factory was once again was working overtime and everyone on the factory
floor was happy.
In later years I was fortunate enough to have lunch with W. Edwards Deming
and told him this story. Before I got to the punch line he finished it for
me --- apparently mine wasn't the only factory where the workers had a
vested interest in inefficiency.
It wasn't until the exceptional quality consistency of Japanese products
began to be recognized by the US public that US companies decided that the
market demanded improvements in their quality.
It should be remembered that "quality" is defined in term of requirements,
not perfection. If the products meet the requirements more than 99.9 % of
the time they are probably being correctly quality controlled (1 lemon per
thousand units). For certain things five nine's is critical (internet
routers, etc.), but for regular life three nine's is often as much as we are
willing to pay for.
Obviously, the critical thing is the "requirements" with respect to product
characteristics --- expected product lifetime of use, consistency between
units, frequency of failure, achieving published performance specifications,
etc.
Tod, KØTO
--
FAQ on WWW: http://www.contesting.com/FAQ/towertalk
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
|