You are on page 1of 3

International Journal of Advanced Engineering Research and Studies


Review Article


*Hartaj Singh

Address for Correspondence

Mechanical Engineering Department, Mahant Bachittar Singh College of Engineering and Technology,
Jammu, India
The aim of this paper to review the Taguchi method is used to find the best process parameters and Improved quality results.
Taguchi technique investigates the variation in experiments, and generally approach of system, parameter and acceptance
aim have been significant in improving man-made quality worldwide. The highest possible performance is obtained by
determining the optimum combination of design factors. The present work focused by using L 9 (3)2 Orthogonal Array (OA)
on the processing steps to get the optimal values with the help of main effects graph, and Analysis of variance (ANOVA)
was employed to investigate the characteristics and experimental results are provided the effectiveness of this approach. This
technology has met the current needs of industry owing to its shorter design cycles and improved the design of quality.
KEYWORDS Taguchi method, Orthogonal array, Analysis of variance.

Taguchi methods are statistical approach developed
to enhance the quality of manufactured goods, and
more recently also applied to, engineering,
Professional statisticians have welcomed the goals
and improvements brought about by Taguchi
methods, particularly by Taguchi's development of
designs for studying variation, but have criticized the
incompetence of some of Taguchi's appliances [1-2].
A quick exploration in related journals, as well as the
World Wide Web, discloses that the method is being
successfully accomplished in assorted areas of
applications are wide which includes manufacturing,
military, automobile, aerospace and other industries.
The Taguchi method is aimed at the manufacturing
situations. The Taguchi Method has been extensively
elaborated and analyzed in published research works.
Box and Meyer [3] suggested a method to estimate
the variance of the response and identified factors
that affect it with small non-replicated designs.
Economic globalization and the rapid and continuous
appearing of new technologies mobilized to obtain
the maximum degree of competitiveness, high quality
products in short time in order to ensure their survival
and growth in the world wide market. In most quality
control situations, the goal is to produce output as
uniformly near a target value as possible and the
reduction of variation is now regarded. To understand
the workings of Taguchi methods is aimed at the
characteristics of interest. These three are: large
values are better, smaller values are better, target
values are better [4-5]. A quality characteristic is an
important dimension, property or attribute of a
manufactured output. The Taguchi approach can
empathize applications in interest with related pasture
design of VLSI; optimization of communication &
information networks, development of electronic
circuits, laser engraving of photo masks, cash-flow
optimization in banking, government policymaking,
runway utilization improvement in airports, and even
robust eco-design [6-8]. Manufacturing process
optimizations are powerful methods that provide
simulation scenarios that yield the desired outcome.
The optimization techniques could contain met
heuristic procedures and/ or classical optimization
methods [9] that involve setting a series of
IJAERS/Vol. I/ Issue III/April-June, 2012/39-41

parameters in order to obtain: (a) Maximum return on

budgets. (b) Most effective configuration of
machines. (c) Most effective allocation of raw
materials. (d) Optimal work force allocations to
minimize labor and total time. The objective of a
successful investigation is to obtain as much reliable
information as possible about the process and the
designs of experiment there are different methods are
used. In case when one observes many inputs and
interactions and there is also a time limitation the
Taguchi approach is used for Design of experiment
[10-12]. The Taguchi method is used whenever the
settings of interest parameters are necessary, not only
for manufacturing processes. Therefore, The Taguchi
approach is used in many domains such as:
environmental science [13-14], agricultural [15],
medicine [16], chemistry [17], physics [18], statistics
[19], management and business [20].
Orthogonal arrays are often employed in industrial
experiments to study the effect of numerous control
factors. An Orthogonal array is a type of experiment
wherever the columns for the independent variables
are orthogonal to one another. By orthogonal array
the analysis is simple and large saving in the
experiment effort. By using an orthogonal array of
standard procedure can be used for a number of
experimental situations. To describe an orthogonal
array [21], one must identify:
Number of factors to be studied.
Levels for each factor
The specific two-factor interactions to be
The special intricacy that would be encountered
in running the experiment.
The given parameter selected for the experiment was
percentage (%) of X as factor one, and Particle size
as factor two having the same material. The
experiment consists of two factors and three levels;
Table 1 indicates the factors and their levels.
Table 1. Process parameter with their values at
three levels.
Percentage (%) of X
Particle size, Y

Level 1

Level 2

Level 3

Although, Taguchi's approach was built on traditional

concepts of design of experiments (DOE) is a body of
statistical techniques for the effective and efficient set

International Journal of Advanced Engineering Research and Studies

of data for a number of purposes. Two significant
ones are the investigation of research hypotheses and
the accurate resolve of the relative effects of the
many different factors that influence the quality of a
product or process. DOE can be employed both in
product design phase and production phase [22]. In
present investigation design of experiment was
performed as L9 orthogonal array was chosen, which
has 9 rows and 2 columns as shown in Table 2. And
therefore, using Taguchi software 'Qualitek-4 (Demoversion)' by Newtek.
Table 2.
Experimental layout using L9 (32)
orthogonal array.
L9 (32)Experimental Trail

Level 1

Level 2


In case of multi-objectives, reduce observations [23]

into results and perform analysis to:
Determine factor influence (Main Effect).
Identify significant factors (ANOVA).
Determine optimum condition and estimate
Calculate confidence interval of optimum
Finally, the process parameters can be verified
by confirmation test.
To choose among several mixture design approaches,
such as optimal, simplex centroid, simplex lattice,
and extreme vertices and ABCD designs and specify
number of replicates for mixture designs. The quality
characteristics investigated in this study that the
combination of parameters and their levels make the
best combination say optimal quality characteristic to
be achieved [24]. In this type bigger is better present
the main effect graph. The average effects of
percentage of the material at level 1, 2, and 3
respective values as average at the points are X1, X2,
and X3. And the average values of levels at; L1, L2,
L3 of given particle size: Y1, Y2, and Y3 as shown in
Figure 1.

Figure1. Main effects graph, average value verses

In analysis of variance many quantities as:
(a) Degree of freedom
(b) Sums of squares
(c) Variance
IJAERS/Vol. I/ Issue III/April-June, 2012/39-41


(d) F-ratio
(e) Pure sum
(f) Percentage contribution
ANOVA can be computed and organized in a
standard tabular format. Utilize this step to review a
number of examples analyses to build the confidence
in interpreting the experimental results [25]. The
purpose of the analysis of variance (ANOVA) was to
investigate which parameters significantly affected
the quality characteristics.
(a) Degree of freedom: Total degree of freedom is
equal to the degree of freedom is associated with
number of levels plus degree of freedom is associated
with error. Therefore, the total degree of freedom is
equal to N-1. It can be calculated by this formula:
VT = VA + VE
VT = Total degree of freedom.
VA = Degree of freedom for a controlled factor or Number of
VE = Degree of freedom for error.

VT = N-1
VA = Number of levels -1
VE = VT - VA
(b) Sums of squares: The sum of squares is a
measure of the deviation of the experimental data
from the mean value of the data. Summing each
squared deviation emphasizes the total deviation. The
sum of squares (SS)A analysis for the factors and total
variation is
(SS)A = A12 + A22 + A32 T2
nA1 nA2 nA3 N
This is the value for the percentage (%) participation
of factor one, X variations can be observed in terms
of sum of squares. Also the participation of Particle
size, Y can be
calculated by:
(SS)B = B12 + B22 + B32 T2
nB1 nB2 nB3 N
(c) Variance: Variance measures the distribution of
the mean of data. Since the data is representative of
only a part of all possible data, DOF rather than the
number of observations is used in the calculation.
Variance = Sum of Squares
Degrees of freedom
VA = (SS)A
Similarly, VB = (SS)B
(d) F- Ratio or Variance ratio is used to measure the
significance of the factor under investigation with
respect to the variance of all the factors included in
the error term. When the computed F value is less
than the value determined from the F tables at the
selected level of significance, the factor does not
contribute to the sum of the squares within the
confidence level. The F values are calculated by:
FRA = Vm = Mean variance
Error variance = (SS)A / f
FRB = (SS)B / f
(e) Pure sum of squares are determined by using
Pure sum of squares, (S')A = (SS)A DOF VE
(S')B = (SS)B DOF VE
(f) The percentage contribution for any factor is
obtained by dividing the pure sum of squares for that

International Journal of Advanced Engineering Research and Studies

factor by ST and multiplying the result by 100. The
percentage contribution is denoted by P and can be
calculated using the following equations.
PA = (S')A 100
Also, PB = (S')B 100
The optimum condition is determined based on the
quality characteristic selected for the analysis. It is a
common practice to only include the significant
factor in calculating:
Total contribution from, all factors
Current grand average of performance
Expected result, at optimum condition
And the optimum condition and performance can be
obtained by these results and which can be participate
and in concert a significant task as working on
various tests [26].
The confirmation tests were performed by selecting
the set of parameters and the confidence level is
taken as numeric value 90 to 95%, achieving best
value of the response. It is an expected result at
optimum condition. From the analysis, error
associated with all the experimental results were its
value also may obtain. Error showed due to the more
parameters also be considered while machining or
testing or multiphase microstructure of the material
property [27].
The Taguchi method is especially suitable for
industrial use, but it can also be used for scientific
research purposes and it emphasizes a mean
performance characteristic value close to the target
value rather than a value within certain specified
limits, thus improving the product quality. In present
study, design of experiment was performed by L9
orthogonal array was chosen by considering two
factors and three levels were employed to analyze the
influence of process parameters by using main effects
and analysis of variance to get the optimal conditions
and performances that means the best parameters
within the experimental results. Finally, the optimum
combinations of parameters are achieved by
confirmation tests were conducted to verify the
results are experimentally obtained.
1. Rosa JL, Robin A, Silva Baldan M.B., Journal of Materials
Processing Technology, 209 (2009) 1181-1188.
2. Selden PH., "Sales Process Engineering, A Personal
Workshop", Milwaukee, Wisconsin, ASQ Quality Press.
3. Box G, Meyer RD., "Dispersion effects from fractional
designs",Technometrics 28(1) (1986) 19-27.
4. Benardos PG, Vosniakos GC.,"Prediction of surface
roughness in CNC face milling using neutral networks and
Taugchi's design of experiments", Robotics and computer
Integrated Manufacturing 18 (2002) 343-354.
5. Montgomery DC., "Design of analysis of experiments. 4 ed.
USA: John Wiley&Sons".(1997).
6. Phadke MS., "Quality engineering using robust design",
Prentice Hall, Englewood Cliffs, New Jersey, (1989).
7. Ross, P.J., "Taguchi Techniques for quality Engineering,
McGraw-Hill", New York, NY, (1998).
8. Taguchi G., "Quality engineering (Taguchi methods) for the
development of electronic circuit technology," IEEE
Transactions on Reliability, 44 (1995) 225-229.

IJAERS/Vol. I/ Issue III/April-June, 2012/39-41


9. Czyzak P, Jaszkiewiez A., Journal of Multi-Criteria Decision

Analysis, 7(1) (1998)
10. Roy RK., "Design of experiments using Taguchi approach",
John Willey & Sons, Inc., New York, (2001).
11. Casalino G, Currcio F, Memola F, Capece M., (2005),
Journal of materials processing technology., 167 (2005) 422428.
12. Ozcelik B, Erzurumlu T., Journal of materials processing
technology. 171 (2006)
13. Daneshvar N, Khataee AR, Rasoulifard MH., Pourhassan M.,
Journals of Hazardous Materials., 143(1-2) (2007) 214-219
14. Du Plessis BJ, De Villiers GH., "The application of Taguchi
method in the evaluation of mechanical flotation in waste
activated sludge thickening Resources, Conservation of
recycling", 50(2) (2007) 202-210.
15. Tasirin SM, Kamarudin SK, Ghani JA, Lee KF., Journal of
Food Engineering"., 80(2) (2007) 695-700.
16. Ng EY, Ng WK.,(2006), Journal of Medical and Biological
Engineering and Computing., 44(1-2) (2006) 131-139.
17. Houng JY, Liao JH, Wu JY., Shen SC., Hsu HF.,
"Enhancement of asymmetric bioreduction of ethyl 4-choloro
acetoacetate by the design of composition of culture medium
and reaction conditions", Process bio-chemistry.,137(1)
(2007) 1-7.
18. Wu CH, Chen WS., "Injection molding and injection
compression molding of three beam grating of DVD pickup
lens", Sensors and Actuators A., Physical., 125(2) (2006)
19. Romero-Villafranca R, Zunica L, Romero-Zunica R., Journal
of Statistical Planning and Inference, 137(4) (2007) 14881495.
20. Elshennawy AK.,"Quality in a new age and the body of
knowledge for quality engineers, TQM and Business
Excellence., 15(5-6) (2004) 603-614.
21. Fisher RA.,"Statistical Methods for researcher workers",
Oliver and Boyd: London, UK, (1925).
22. Ranjit KR.,"Design of experiments using the Taguchi
approach", 16 steps to product and process improvement;
John Wiley & Sons: Hoboken, NJ, (2001).
23. Fisher RA., Journal of Min. Agr. Engl., 33 (1926) 503-513.
24. Taguchi G., "Designing quality into products and processes",
Unipub/ Kraus, NY, (1986).
25. Barrado,E, Vega M,
Grande P,
Del Valle DL.,
"Optimization of a purification method for metal-containing
waste water by using of a Taguchi experimental design",
Water Res. 30 (1996) 2309-2314.
26. Taugchi G, Konishi S., "Taguchi Methods: Orthogonal array
and Linear graphs", American Supplier Institute, USA,
27. Roy, R.K., "Competitive Manufacturing Series, Van
Nostrand Reinhold, New York, (1990).