RE: System vs. Individual - B Coop

["Prevette, Steven S" <Steven_S_Prevette@rl.gov>]

John Bruman said: How can we do this?

By focusing on the inner workings and the interactions of all the
variables within our process of driving home every day. Yes human
fallibility can be overcome. Real people CAN "Do It Right Every Time".

It however requires in depth study of the internals of the process
rather than simply examining outputs and reacting to them.

Does this make sense to anyone else but me?

Or am I the only one that is confused?

======

I would suggest that yes, that could be done.  If I was interested in
the results of flipping a coin, I could exert an exact amount of force
on an exact location on the coin, monitor all of the interactions of the
coin with air molecules and the floor and predict whether the coin will
land "heads" side up or "tails" side up.  Very expensive, but
conceptually do-able.  But what would be the benefit?

Dr. Shewhart posed SPC in the same idea - it is an issue of Economics.
We could indeed to in depth studies of EVERYTHING.  We would quickly run
out of time and money, and likely not find any results that could be
applied such that the benefits would exceed the cost.

By treating process results as a random process, and separating out
non-randomness from randomness we likely will find things that can be
done whose benefit will exceed the cost.  We can target those areas ripe
for improvement with methodologies set to find common causes when
stable, or special causes when not stable.  By the way, even stable
processes can be improved, but requires us to step back away from
monitoring the internals, and monitoring the interactions between the
stable process and the rest of the system over a long run.


Steve Prevette
Occupational Safety and Health
Fluor Hanford, A Fluor Government Group Project
ASQ Certified Quality Engineer
steven_s_prevette@rl.gov
509-373-9371