Statistical Design of Experiments
Statistical Design of Experiments
Sometimes in industrial experiments
Sometimes in industrial experiments
we have reason to believe that
we have reason to believe that
higher order interactions don’t
higher order interactions don’t
exsist or are negligible. In such a
exsist or are negligible. In such a
or are negligible. In such a
case they aren’t necessary to
case they aren’t necessary to
evaluate.
evaluate.
Fractional Factorial DOE
Fractional Factorial DOE
A Fractional Factorial Experiment is
A Fractional Factorial Experiment is
a Full Factorial Experiment in which
a Full Factorial Experiment in which
all treatment combinations of the
all treatment combinations of the
factors are NOT tested, and thus not
factors are NOT tested, and thus not
all interactions can be evaluated
all interactions can be evaluated
If there are many factors, a Full
If there are many factors, a Full
Factorial Experiment requires all
Factorial Experiment requires all
possible combinations of levels of
possible combinations of levels of
the factors which will result in a
the factors which will result in a
large number of tests (and
large number of tests (and
resources).
resources).
 |
 |
 |