In: Statistics and Probability
Comparison of ANOVA with Taguchi Approach: as a report
Solving the same problem with ANOVA and Taguchi
The Taguchi method involves reducing the variation in a process
through robust
design of experiments. The overall objective of the method is to
produce high quality
product at low cost to the manufacturer. The Taguchi method was
developed by Dr. Genichi Taguchi, a method for designing
experiments to investigate how different
parameters affect the mean and variance of process performance
characteristics that defines how well the process is functioning.
The Taguchi method gives the S/N ratio asthe performance index to
evaluate the characteristics of the product or process. It can
beeasily defined as the ratio of the mean (signal) to the standard
deviation (noise) by S/N ratio. The S/N ratios may be depended on
the particular type of performance
characteristics, including smaller-is-better (ZS) or
larger-is-better (ZL).
Analysis of variance (ANOVA) is one of the statistical models
used to study the
difference among group means plus their connected procedures like
differences between groups. In ANOVA, the variance observed in a
prescribed parameter is divided into parts attributable to various
sources of deviation. ANOVA provides a statistical test of
means
for several groups are equal or not and accordingly generalizes the
t-test for more than two groups. By many t-tests of two-samples
will increase chance of a type I error.
Because of which, ANOVAs are useful in comparing three or more
means for statistical significance.
In this study, Orthogonal array L9(3
3
) [50] experimental design method was
chosen to determine the experimental plan. In this study the
control parameters like
porosity, heat input, and thermal conductivity of material were set
as a level as shown in
Table 9.5. In order to observe the effect of noise to source ratio
on the heat transfer
coefficient each experiment was repeated three times under the same
condition as per
L9(3
3
) table. Values were determined by comparing the standard method
and analysis of
variance (ANOVA) which is based on the Taguchi method. The
objective was to obtain
performance characteristics (maximum heat transfer coefficient)
hence, larger the better
was chosen.
As explained above the
parameters chosen are: Heat input, Porosity and Fin
material and the heat transfer coefficient was the measure of the
outcomes of varying
these parameters. In this study, from experimental readings, the
average heat transfer coefficient
(ha) was used to calculate the S/N ratio.
Both the values of ha and S/N
ratio, are presented in aboveTable. After calculating
the S/N ratio for each experiment, the average S/N value is
calculated for each factor and
level. For example, the mean S/N ratio for the heat input level II
can be calculated by
averaging the S/N ratios for experiment no. 1, 4, 7, and for level
II experiment no. 2, 5, 8
and for level III experiment no 3, 6, 9. The mean S/N ratio for
each level of the other
parameters can be computed in similar manners that are presented in
the response Table(below). The main effect of each parameter is
nothing but difference of highest and lowest value among the
levels.
Last table indicates that porosity ratio having 50.23 %
contribution and more
significant, material (thermal conductivity) having 29.38%
contribution and heat input
having 20.39%contribution and less significant influence upon the
maximum heat
transfer coefficient in our study.