You are on page 1of 17

ROBUST DESIGN -TAGUCHI PHILOSOPHY

ANIS ATIKAH AHMAD


▪ Taguchi contribution
▪ Taguchi philosophy
▪ Approach to engineering design process
1

Taguchi loss function, used to measure financial loss to society


resulting from poor quality

2
The philosophy of off-line quality control, designing products
and processes so that they are insensitive ("robust") to
parameters outside the design engineers control

3
Innovations in the statistical design of experiments, notably the
use of an outer array for factors that are uncontrollable in real
life, but are systematically varied in the experiment.
Genichi Taguchi,
Quality Engineering Thinker
Products and processes should be designed so that they are
robust to external sources of variability (insensitive to
uncontrollable factors that will influence the product or process).

Experimental design methods are an engineering tool to help


accomplish this objective.

Operation on-target is more important than conformance to


specifications.
1. System Design
• This may involve the development of a
prototype design and will determine the
materials, parts, and assembly system to be
used

2. Parameter Design
• identifying easily controllable factors and
settings that minimise performance
variation.

3. Tolerance Design
• to determine the best tolerances for the
parameters (specification limits).
• A way to fine-tune result of the parameter
design by tightening/loosening the tolerance of
factors.
Consider the development of an elastometric connector that would deliver the
required pull-off force when assembled with a nylon tube. There are four
controllable factors, each at three levels (s), and three noise or uncontrollable
factors, each at two levels (E conditioning time, F conditioning temperature, and
G conditioning relative humidity).
▪ Taguchi crossed array design

Uncontrollable factors

Controllable factors

23 design for the


outer array
it is a 34-
2 design

for inner
array
factors A and C
have larger effects
than do B and D.
The Effects of Controllable Factors on the
Signal to Noise Ratio

In terms of maximizing SNL we would


select AMedium, CMedium, BMedium, and DLow.

In terms of maximizing the average


pull-off force y , we would choose
AMedium, CDeep, BMedium and DLow.
Notice that there is almost no
difference between CMedium and CDeep.
The implication is that this choice of
levels will maximize the mean pull-off
force and reduce variability in the
pull-off force.

The Effects of Controllable Factors on Each Response


▪ Taguchi advocates claim that the use of the SN ratio generally eliminates the
need for examining specific interactions between the controllable and noise
factors, although sometimes looking at these interactions improves process
understanding.
However, the author in this study found
that the AG and DE interactions were
large.

Analysis of these interactions, shown in


Figure 4, suggests that AMedium is best. (It
gives the highest pull-off force and a
slope close to zero, indicating that if
we choose AMedium the effect of relative
humidity is minimized.)
The analysis also suggests that DLow
gives the highest pull-off force
regardless of the conditioning time.
▪ When cost and other factors were taken into account, the experimenters in this
example finally decided to use AMedium, BThin, CMedium, and Dlow. (BThin was much less
expensive than BMedium, and CMedium give slightly less variability than CDeep.)
Array No of No of
factor level
L4 (2) 3 2
L8 (2) 7 2
L12 (2) 11 2
L16 (2) 15 2
L32 (2) 31 2
L9 (2) 4 3
L18 (2) 1 2
L18(3) 7 2
L27 (3) 13 3
L16 (4) 5 4
L32 (2) 1 2
L32 (4) 9 4
L64 (4) 21 4
▪ Taguchi use signal to noise ratios response variables
◼ use SNT if the objective
is to reduce variability
around a specific target,


Which
SN if the system is
L
optimized when the
equation
response is as largeto
possible, and
as


use?
SN if the system is
S
optimized when the
response is as small as
possible.

◼ Factor levels that


maximize the
appropriate SN ratio are
optimal.
▪ In summary, we should support Taguchi's philosophy of quality
engineering.
▪ However, we must rely on simpler, more efficient methods that are easier
to learn and apply to carry this philosophy into practice. The response
surface modeling framework is an ideal approach to process
optimization, which is fully adaptable to the robust parameter design
problem.

You might also like