Click here for a more complete video explanation
Walker Runn is quality manager for Color in a Can,
a paint processing company with facilities in three states. He is
pleased to be assigned to the plant that lies furthest from company
headquarters, hoping that an "out of sight, out of mind"
sensibility will keep the home office from pestering him about things
like statistical process control, or getting more training to do
his work, or a host of other things that are pebbles in his shoe.
The company sends a vice president to audit his quality systems
each year, but Walker has been able to sustain a facade of competence
because the vice presidents that the home office sends have generally
known little about statistical process control, so he can simply
throw around a few impressive terms and charts and they leave satisfied.
This time, however, the company has chosen to send Dan Druff, a
highly competent (though somewhat flaky) statistician.
Thumbing through an old statistics textbook, Walker
Runn decides that the only way to convince Dan Druff that the plant
is producing consistently high quality paint, is to wow him with
hypothesis testing. Since he has never done hypothesis testing himself,
Walker finds that he must actually skim the chapter to learn some
terminology. A section entitled "Differences between Means"
draws his interest. In the associated case study, a plant has two
identical production lines that produce identical products--not
unlike the way Color in a Can is set up, with two lines that produce
the same product.
As part of the corporate quality effort, each production
line is rated on a quality index each day. Data for the two lines
Walker collects the data below and asks one of his
employees to enter it into a statistical software program so it
can be presented to Dan Druff in a Major PowerPoint presentation
that he has planned.
As he shares the presentation, he points to the
high t value of 2.834. "Aha," he says. This demonstarates
how different the lines are from each other, he points out to Dan
Druff, who nods, and then asks Walker what alpha value might be
Oh-oh. "Selecting the Alpha" was in the
part of the chapter that Walker had not skimmed. The only connection
with 'alpha' that came to his mind derived from the sports car his
neighbor just bought, an Alfa Romeo 164. Fishing for a response,
he blurted out "12," since the sum of
and is 12. Not
fancy, but a fast calculation, he thought to himself.
Note: I am his neighbor. This is my Alfa Romeo 164.
(What do you think of my license plate?)
Was Dan Druff impressed? Should he have been?
a) No. Walker Runn
should have done more than brush up on his terms.
b) Yes. Dan Druff went
back to company headquarters praising Walker Runn's grasp of statistical
quality control as well as the plant's production process.
Copyright 2006 PQ Systems.
Please direct questions or problems regarding this web site to the Webmaster.