= X1
•X
2
• X
2
4
5-1
fraction
is called a
resolution V
design.
Looking at
the
confounding pattern
we see
that
1
1
→
345
and
1
12
→
12
+ 345.
Thus main
effects are confounded
with four-factor
interactions, and two-
tor interactions withthree-factor interactions.
a design of
resolution
R is one
in
which no
p-factor effect
is confounded
with
any other effe
factors.
resolution
of
a
design
is denoted
by the
appropriate Roman lette
as a
subscript.
of resolution
R = III
does not confound
main effects with
one another
but
does
effects with two-factor
interactions.
of resolution
R = IV does not
confound
main
effects and
two-factor interactions but
two-factor
interactions with other two-factor interactions.
of resolution
R = V
does not
confound
main
effects and
two-factor
interactions with
but does confound two-factor
interactions with
three-factor interactions,
and
so
on.
the
resolution
of
a
two-level fractional design
is
the length of the shortest word in the definin
Fractional
Designs
For K Variables And N Runs
Represent Replication).
3 – 1
= 1 2
3
4– 1
V
5 – 2
6 – 3
7 – 4
± 5 = 1 3
± 6 = 2 3
± 5 = 1 3
± 7 = 1 2 3
± 6 = 2 3
= 1 23
5 = 1 3
3
4
5 –1
V
6– 2
V
7– 3
V
8– 4
V
9 – 5
1 0 –6
1 1 –7
±
±
±
±
±
± 6 = 23 4
± 7 = 13 4
± 8 = 12 4
± 6 = 23 4
± 7 = 13 4
± 6 = 13 4
12 3
23 4
12 3
12 4
1 23 4
12 3 4
13 4
12 4
1 23 4
11
3
4
5
6 –1
V
7– 2
V
8– 3
V
9– 4
V
10 – 5
V
11 – 6
V
±
±
± 7 = 1 23 5
2 34 5
12 3
1 34 5
1 24 5
1 23 4
12 4
1 24 5
1 34 5
= 1 23 4 5
1 23 4
2 34 5
= 1 23 4 5
2 34 5
3
4
5
6
7– 1
V
8– 2
V
9– 3
V
10 – 4
V
11 – 5
V
±
±
±
± 8 = 1 34 6
1 23 4
1 23 4
1 35 6
1 24 5
7 = 1 2 34 5 6
1 25 6
3 45 6
1 23 5
3
4
5
6
7
8– 1
V
9 –2
V
1 0– 3
V
1 1– 4
V
±
1 23 7
= 1 34 6 7
2 34 5
= 2 35 6 7
1 34 6
=
↑
1?
? ?
?
↑
1?
4?
?
↑
1?
? ?
↑
1
?
16?
7