(Arbitrarily linked to LO29884 by the author.)
Another point to remember about Six Sigma is what it assumes:
- an adequate metric of a repeatable process,
- a well-defined "ideal" value of the metric,
- and that it makes sense to compare measurements made over time.
I recently attended a talk on "Software Applications of Six Sigma". The
main usefulness of the Six Sigma "toolkit" was the provision of standard
quality improvement tools and processes, and the attempt to define useful
metrics. Typically, by the time a team had gone through the exercise of
trying to define the metrics, they understood enough about the system to
make significant quality improvements without trying to "play Procrustes"
and force fit the process into the statistical variation bed.
This isn't to say that the exercise was a waste of time; it provided a new
perspective on continuous improvement that added considerable depth to the
participants' understandings. Fortunately, the software organization that
did this was at a fairly high maturity level (in the SEI CMM framework)
and knew enough to use the tools wisely.
--Don Dwiggins "Experience is what you get when d.l.dwiggins@computer.org you were expecting something else." -- Seen on an office wall
Learning-org -- Hosted by Rick Karash <Richard@Karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>
"Learning-org" and the format of our message identifiers (LO1234, etc.) are trademarks of Richard Karash.