Taguchi Methods
Genichi Taguchi has been identified with the
advent of what has come to be termed quality
engineering.
The goal of quality engineering is to move quality
improvement efforts upstream from the production
phase to the product/process design stage (off-line).
As his loss function demonstrates, his main concern
is deviation of a characteristic from its nominal
value. Uncontrollable factors (noise) are often
responsible for this deviation and, therefore,
Taguchis approach to experimental design has as
its goal the design of products/process that are
robust to these noise factors.
Taguchis three stage design
process
System Design - create prototype product and
process to produce it.
Parameter Design - find settings of process and
product parameters which minimize variability.
Tolerance Design - tradeoff between loss to
consumer and manufacturing costs
Signal to Noise Ratios
In the parameter design stage Taguchi makes use of
designed experiments and signal to noise ratios to
determine the optimal parameter settings.
The signal to noise ratios are derived from the
Taguchi loss function.
While Taguchi has proposed a large number of
signal to noise ratios three are the most widely
used:
Nominal is Best:
Larger is Better:
Smaller is Better:
|
|
.
|
\
|
=
2
2
log 10
s
y
SN
N
|
|
|
|
.
|
\
|
=
=
n
y
SN
n
i i
L
1
2
1
log 10
|
|
|
|
.
|
\
|
=
=
n
y
SN
n
i
i
S
1
2
log 10
Experimental Design
Taguchi has designed a number of orthogonal
arrays to aid in the development of experiments
These arrays are essentially balanced fractional
factorial designs.
He suggests using two array matrices for each
designed experiment.
The inner array is used to study the effects of the
design parameters we wish to study.
An outer array is used to model the noise factors
that may impact the performance of the product in
the field.
Two of the Taguchis simpler Orthogonal arrays
are:
L
4
(2
3
) and the L
8
(2
7
):
L
4
(2
3
)
Factors
run 1 2 3
1 1 1 1
2 1 2 2
3 2 1 2
4 2 2 1
1 2
3
The L
8
(2
7
) Orthogonal Array and its Linear Graphs
Factors
Run 1 2 3 4 5 6 7
1 1 1 1 1 1 1 1
2 1 1 1 2 2 2 2
3 1 2 2 1 1 2 2
4 1 2 2 2 2 1 1
5 2 1 2 1 2 1 2
6 2 1 2 2 1 2 1
7 2 2 1 1 2 2 1
8 2 2 1 2 1 1 2
3
4
7
6
1
5
2
1
6
3
7
4
5
2
Example
In 1987 Taguchi published a paper in quality progress
giving an example of his approach. The objective was to
maximize the pull-off force of a connector to a nylon
tube for an automotive application so SN
L
. The factors
studied and there levels are tabled below along with the
results:
Outer Array (L
8
) E
F
G
1
1
1
1
1
2
1
2
1
1
2
2
2
1
1
2
1
2
2
2
1
2
2
2
Inner Array (L
9
) Responses
Run A B C D Ave SN
l
1 1 1 1 1 15.6 9.5 16.9 19.9 19.6 19.6 20.0 19.1 17.5 24.0
2 1 2 2 2 15.0 16.2 19.4 19.2 19.7 19.8 24.2 21.9 19.4 25.5
3 1 3 3 3 16.3 16.7 19.1 15.6 22.6 18.2 23.3 20.4 19.0 25.3
4 2 1 2 3 18.3 17.4 18.9 18.6 21.0 18.9 23.2 24.7 20.1 25.9
5 2 2 3 1 19.7 18.6 19.4 25.1 25.6 21.4 27.5 25.3 22.8 26.9
6 2 3 1 2 16.2 16.3 20.0 19.8 14.7 19.6 22.5 24.7 19.2 25.3
7 3 1 3 2 16.4 19.1 18.4 23.6 16.8 18.6 24.3 21.6 19.8 25.7
8 3 2 1 3 14.2 15.6 15.1 16.8 17.8 19.6 23.2 24.2 18.3 24.8
9 3 3 2 1 16.1 19.9 19.3 17.3 23.1 22.7 22.6 28.6 21.2 26.2
Design Factors Levels
A Interference Low Medium High
B Connector wall thickness Thin Medium Thick
C Insertion depth Shallow Medium Deep
D Percent adhesive Low Medium High
Noise Factors Levels
E Conditioning Time 24h 120h
F Conditioning Temp 72 150
G Conditioning Humidity 25% 75%
Taguchi used the L
8
design to model the noise factors
and the L
9
(3
4
) series of orthogonal arrays to model
the design factors. The L
9
design is as follows:
Factors
Run 1 2 3 4
1 1 1 1 1
2 1 2 2 2
3 1 3 3 3
4 2 1 2 3
5 2 2 3 1
6 2 3 1 2
7 3 1 3 2
8 3 2 1 3
9 3 3 2 1
1
3,4
2