Taguchi Method for DOE

Introduction

• Taguchi methods of experimental design provide a

simple, efficient and systematic approach for the
optimization of critical parameters of any process.
• It is an effective method in investigating the effects of
multiple factors on performance as well as to study the
influence of individual factors to determine which factor
has more influence, which less.
• It minimizes to a great extent, the number of experiments
by means of incorporating readily available fractional
factorial matrices or orthogonal arrays.
• Taguchi experiment can be accomplished within a limited
time frame and at a reduced cost, with results comparable
to that of a full factorial experiment
 Taguchi’s Design Stages
• Taguchi recommends a three-stage process to achieve
desirable product quality by design :
– System Design
– Parameter Design
– Tolerance Design.
 Parameter Design
• The objective of the parameter design is to optimize
the settings of the process parameter values for
improving performance characteristics.
• Parameter design is the key step in the Taguchi
method to achieving high quality without increasing
cost.
• Orthogonal arrays, variance and signal to noise
analysis are the essential tools of parameter design.
 Orthogonal Arrays(OAs)
• For designing experiments, Taguchi utilized a special set of
tables, called orthogonal arrays (OAs), which represent the
smallest fractional factorial experiments
• The use of these arrays helps determine the least number of
experiments needed for a given set of factors thus reducing
the time and cost of experimentation.

Types of OA
• There are three types of OAs,
– OAs for two-level factors,
– OAs for three-level factors
– OAs for four-level factors
– OAs for mixed-level factors.
Taguchi Designs
 Insensitivity to Noise
• Noise = Factors which the engineer can not or
chooses not to control
– Unit-to-unit
• Manufacturing variations
– Aging
• Corrosion
• UV degradation
• wear
– Environmental
• human interface
• temperature
• humidity
 Example
Experimental Level 1 Level 2 Level 3
variable and
their levels
Spindle 250 355 500
speed(rpm)
Feed 0.18 0.24 0.32
rate(mm/rev)
Depth of 0.5 1 1.5
cut(mm)

Please use Taguchi’s L9 and L27 array for this

experimental table
 Signal to Noise (S/N) ratio

• To determine the effect of each variable has on

the output, the signal-to-noise ratio (or the SN
number) needs to be calculated for each
experiment conducted.
• In the Taguchi method, the S/N ratio is a
measure of quality characteristic deviation
from the desired value.
• The term signal represents the desirable value
(mean) and the noise represents the undesirable
value (standard deviation from the mean) for
the output characteristic.
 Types of S/N ratios
• There are three categories of the performance
characteristic in the analysis of the S/N ratio
– the smaller-the-better
– the larger-the-better
– the nominal- the-better.
Types of S/N ratios
 Types of S/N ratios
• S/NT if the objective is to reduce variability around a
specific target,
• S/NL if the system is optimized when the response is as
large as possible
• S/NS if the system is optimized when the response is as
small as possible.
• Regardless of the category of the performance
characteristic, the larger S/N ratio corresponds to the
better performance characteristic.
• The optimal level of the process parameters is the level
with the highest S/N ratio.
 • After, S/N ratio in the experiment calculated, which is then used
to create a response Table and plot a response diagram.
• The average response value of the S/N ratio of each factor is
calculated using Equation below

• The parameter with the strongest influence is determined by the

difference between the maximum and minimum value of the mean
of S/N ratios as Delta, and ranks based on Delta values.
• The term ‘rank’ denotes the effects of the parameters on the output.
Rank 1 to the highest Delta value, Rank 2 to the second highest,
and so on, to indicate the relative effect of each factor on the
response.
 Example
 In this study WH particle volume fraction (20%, 30%, 40%),
NaOH concentration (5%,10% & 15% w/v fraction ), treatment
time (6,12 & 24 hours ) and coupling agent dosage (5%,10% &
15%) are selected as the parameters.
Table Parameter and their levels
Levels
Parameter
designation
Parameter
1 2 3

WH Particle
A volume fraction 20 30 40
(%)
Concentration of
B 5 10 15
NaOH (%)
Treatment time
C 6 12 24
(Hour )
29 Coupling agent
D 5 10 15
dosage (%)
 Design of Experiments using Taguchi method
 The experiment was designed with Taguchi’s L9 (34) orthogonal array to reduce
the experiment numbers with the four-parameter or factor each have three
levels. Table Experimental design using L9 (34) Orthogonal Array
Levels of parameter settings

Experiment NaOH Coupling

Treatment
run WH Particle (%) concentration agent
time (Hrs)
(A) (%) dosage %
(C)
(B) (D)
1 20 5 6 5
2 20 10 12 10
3 20 15 24 15
4 30 5 12 15
5 30 10 24 5
6 30 15 6 10
7 40 5 24 10
8 40 10 6 15
9 40 15 12 5
30
 Table Experimental results for Tensile with calculated S/N Ratio
Experimental results for tensile strength (MPa)
Levels of parameter settings

Exp. Couplin
run Particle NaOH g agent Standard  
Treatme Trial Trial Trial
volume Concen dosage deviation S/N ratio
nt time 1 2 3
fraction tration volume error (Ƞ)
(C)
(A) (B) fraction
(D)

1 20 5 6 5 25.36 25.40 25.28 0.050 28.07837

2 20 10 12 10 25.95 25.83 25.85 0.052 28.25811
3 20 15 24 15 24.43 24.69 25.11 0.280 27.8675
4 30 5 12 15 28.92 29.46 28.11 0.555 29.19204
5 30 10 24 5 26.46 26.42 26.85 0.194 28.48932
6 30 15 6 10 28.15 28.42 27.12 0.560 28.90571
7 40 5 24 10 23.81 23.23 24.12 0.369 27.49912
8 40 10 6 15 25.69 25.65 25.62 0.029 28.18286
9 40 15 12 5 24.65 25.23 24.5 0.315 27.88462

31
The calculated S/N ratio is used to create a response table and mail
effect plot diagram
Average Response table of S/N ratio for Tensile strength

Particle Coupling
volume NaOH Treatment Agent
Level fraction (%) Concentration (%) Time (Hrs) dosage (%)
1 28.07 28.26 28.39 28.15
2 28.86 28.31 28.44 28.22
3 27.86 28.22 27.95 28.41
Delta 1.01 0.09 0.49 0.26
Rank 1 4 2 3

32
Example
Example 2