Taguchi DOE

Download as pdf or txt
Download as pdf or txt
You are on page 1of 24

Taguchi DOE

MEIE5224
History
• Since 1960 Taguchi method used improving
Japanese product.
• 1980s  becomes importance.

• Old method Inspection Vs. Design


Improvement
Robust Design
• “engineering methodology for improving
productivity during research and development so
that high-quality products can be produced
quickly and at low cost”(Phadke, 1989).

• Minimize the effect of variation without


eliminating the causes.

• robust design, a design that has minimum


sensitivity to variations in uncontrollable factors.
Taguchi Quality Loss Function
• Famous Story ( Sony US Vs. Sony Japan)
• Color Density ( “m: Target”±5)

US focus on meeting the tolerance while Japan focus on meeting the target
Loss Function

 Ao Outside m   o
L
0 Otherwise

L  k  y  m
2
Average Loss Function
The average loss function when N units produced will be:


L  k S     m
2 2

S 2 : variance
Taguchi suggests a three-
: mean stage process:
monitor the process variables during • System design
production and adjust the process to reduce • Parameter design
manufacturing imperfections so that response • Tolerance design.
parameters fall within the specified tolerances
But this add cost to manufacturing
System Design
• System design stage is where new ideas,
concepts and knowledge in the areas of
science and technology are utilized by the
design team to determine the right
combination of materials, parts, processes an
design factors that will satisfy functional and
economical specifications.
Tolerance Design
• Most Engineers tend to make tight tolerance
which means higher cost.

• tightening the tolerances increases the cost of


the product or process because it requires
better materials, components, or machinery
to achieve the tighter tolerances as we
discussed in earlier chapters.
Parameter Design
• Parameter design is related to finding the
appropriate design factor levels

• make the system less sensitive to variations in


uncontrollable noise factors, i.e., to make the
system robust.
How to use Taguchi in Parameter design

• Control factors are those factors which can


easily be controlled ( material selection, cycle
time, mold temperature)

• Noise factors are factors that are difficult or


impossible or too expensive to control.
– Inner noise
– Outer noise
– In between production noise
Example of noise
Orthogonal Array
Employed by Taguchi Method: Systematically vary levels
L4 , L9 , L12 Common Ones
DOE-Taguchi
• #experiments is obtained by combining the L9
array of control factors (the outer array) with
the L8 array of noise factors (the inner array).
How to get SN ration

n
Yi
Y  output     n : # replication
i 1 n

Y  Y 
2
n
S 
i

i 1 n 1
Example:

Factor A Factor B Factor C run1 run2 Average Standard


deviation
1 1 1 10.5 11.05 10.775 0.3889
1 2 1 12.05 12.10
12.075 0.035355
2 1 2 13.36 10.36
11.86 2.12132
DOE-Taguchi
Then what SN
• After calculating SN based on your context:
1. SN ratios and average responses are plotted
for each factor against each of its levels.

2. Pick the factor level which (1) best maximize


SN and (2) bring the mean on target (or
maximize or minimize the mean, as the case
may be).
Example Manually (hand calculation)
• case study “The Taguchi Approach to Parameter Design,”
by D. M. Byrne and S. Taguchi, Quality Progress, Dec. 1987,
pp. 19 26.

• “The experiment that is being conducted seeks to


determine a method to assemble an elastometric
connector to a nylon tube while delivering the requisite
pull-off performance suitable for an automotive
engineering application. The primary design objective is to
maximize the pull-off force while secondary
considerations are made to minimize assembly effort and
reduce the cost of the connector and assembly.”
Experiment illustration
Factors considered

Interaction is importation for the noise factors


Experiment results
These can’t be controlled outside the lab ! (keep in your mind)
Compute the average at
Each factor level for both
-Mean
-S/N ratio
Analysis of Experiments
Summary including cost and easy to
assemble

Meaning of ‘—” insignificant effect


Using Minitab for 8 factors example

Look at the rank for significance

You might also like