See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/373107691

Taguchi Method

Article in International Journal of Scientific Research in Science Engineering and Technology · August 2015

CITATIONS READS

0 695

3 authors, including:

Shankar Kadam
Bharati Vidyapeeths College of Engineering, Lavale
18 PUBLICATIONS 42 CITATIONS

SEE PROFILE

All content following this page was uploaded by Shankar Kadam on 14 August 2023.

The user has requested enhancement of the downloaded file.

 © 2016 IJSRSET | Volume 2 | Issue 2 | Print ISSN : 2395-1990 | Online ISSN : 2394-4099
Themed Section: Engineering and Technology

Taguchi Method
Shraddha Gaonkar, Nimisha Karanjavkar, Prof. S. N. Kadam
Department of Mechanical Engineering, Bharati Vidyapeeth’s college of Engineering Lavale, Pune, Maharashtra, India

ABSTRACT

The Taguchi method (Tm) is a powerful problem solving technique for improving process performance, yield and
productivity. It reduces scrap rates, rework costs and manufacturing costs due to excessive variability in processes.
It is based on ‘ORTHOGONAL ARRAY’ which gives much reduced ‘variance’ for the experiment with ‘optimum
setting’ of control parameters. By using Minitab17 software we can implement a Taguchi Design. Paper gives a
information about how to use a Taguchi by Minitab17.
Keywords: Taguchi Method, Orthogonal Array, Signal to noise ratio, Minitab17.

I. INTRODUCTION II. METHODS AND MATERIAL

Design of Experiments (DOE) is a powerful statistical Orthogonal Array

technique introduced by R. A. Fisher in England in the
1920's to study the effect of multiple variables Taguchi designs use orthogonal arrays, which estimate
simultaneously. He made effort to make this the effects of factors on the response mean and variation.
An orthogonal array means the design is balanced so
experimental technique more user-friendly and applied it
that factor levels are weighted equally. Because of this,
to improve the quality of manufactured products. each factor can be assessed independently of all the
other factors, so the effect of one factor does not affect
A Taguchi design is a designed experiment that chooses the estimation of a different factor. This can reduce the
a product or process so that functions more consistently time and cost associated with the experiment when
in the operating environment. Taguchi designs recognize fractionated designs are used.
that not all factors that cause variability can be
controlled. These uncontrollable factors are called noise
III. RESULTS AND DISCUSSION
factors. Taguchi designs try to identify controllable
factors (control factors) that minimize the effect of the 3. Experimental Procedure
noise factors. During experimentation, it can manipulate
noise factors to force variability to occur and then Before start using Minitab, need to choose control
determine optimal control factor settings that make the factors for the inner array and noise factors for the outer
process or product robust, or resistant to variation from array. Control factors are factors that can control to
the noise factors. A process designed with this goal will optimize the process. Noise factors are factors that can
produce more consistent output. A product designed affect the performance of a system but are not in control
with this goal will deliver more consistent performance during the intended use of the product.
regardless of the environment in which it is used. The A. Step-1
purpose of the current study was to determine, Taguchi
Design by Minitab17 for reducing variance. Go to Minitab17 then Stat > DOE > Taguchi >
Create Taguchi Design to generate a Taguchi design
(orthogonal array). Each column in the orthogonal array
represents a specific factor with two or more levels.
Each row represents a run; the cell values identify the

IJSRSET1622305 | Received : 14 April 2016 | Accepted : 20 April 2016 | March-April 2016 [(2)2: 843-845]
843
factor settings for the run. By default, Minitab's  Then, adjust the level of one or more factors that
orthogonal array designs use the integers 1, 2, 3, to substantially affect the mean (or slope) but not
represent factor levels. If you enter factor levels, the the signal-to-noise to put the response on target.
integers 1, 2, 3, will be the coded levels for the design.
3.2. Display the alias structure for a Taguchi
B. Step-2 design

After you create the design, you can Create an L8 Taguchi design with three 2-level factors.
use Stat > DOE > Modify Design to rename the factors, You enter the factors in C1, C2, and C3, and the
change the factor levels, add a signal factor to a static responses in C4 and C5. To display the alias structure:
design, ignore an existing signal factor (treat the design
as static), and add new levels to an existing signal factor. 1. Choose Stat > DOE > Factorial > Define
Custom Factorial Design.
C. Step-3 2. In Factors, enter C1 C2 C3.

After creating the design, we can 3. Click Low/High. Click OK in each dialog box.
use Stat > DOE > Display Design to change the units 4. Choose Stat > DOE > Factorial > Analyze
(coded or uncoded) in which Minitab expresses the Factorial Design.
factors in the worksheet. 5. In Responses, enter C4. Click OK.

D. Step-4
Minitab displays the alias structure at the bottom of the
Conduct the experiment and collect the response data. Session window. You can disregard the other output
The experiment is done by running the complete set of TABLE I
noise factor settings at each combination of control STANDARD ORTHOGONAL ARRAY
factor settings (at each run). The response data from
each run of the noise factors in the outer array are Run A B C D
usually aligned in a row, beside the factor settings for 1 1 1 1 1
that run of the control factors in the inner array. 2 1 2 2 2
3 1 3 3 3
E. Step-5 4 2 1 2 3
5 2 2 3 1
6 2 3 1 2
Use Stat > DOE > Taguchi > Analyze Taguchi
Design to analyse the experimental data. It must analyse 7 3 1 3 2
8 3 2 1 3
each response variable separately with Taguchi designs.
Although Taguchi analysis accepts multiple response 9 3 3 2 1
columns, these responses should be the same variable
measured under different noise factor conditions. 4. Measurement System
F. Step-6  To trust your experimental results, you need to
verify that your measurement system is accurate. It
Use Stat > DOE > Taguchi > Predict Taguchi necessary to verify the measurement systems that
Results to predict signal to noise ratios and response are used both to measure the response and to set the
characteristics for selected new factor settings. factor levels.
 If the experiments are part of a larger improvement
3.1. Two-step Optimization for Taguchi project, such as a six sigma project, the
Designs measurement system for the response should have
been validated previously. Make sure that you verify
The goal of a robust parameter design is usually to the measurement system for the factors as well. For
determine factor settings that will minimize the example, if you set an oven at 350 degrees F, you

 First, set all factors that have a substantial effect

variability of the response about some ideal target value. need to be certain that the temperature is actually
350 degrees F.
on the signal-to-noise ratio at the level where the
signal-to-noise is maximized.

International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)

844
 Design combination: After you create your design, output as well as the variation. These measures

 The signal to noise ratio provides a measure of

review the actual combinations of factor settings for are called signal to noise ratios.
each experimental run to make sure they are feasible
the impact of noise factors on performance. The
and safe to run

larger the S/N, the more robust the product is
Trial and Runs: Performing trial runs before you run against noise.
an experiment is useful, if time and budget permits.
Trial runs allow you to: IV. CONCLUSION

i. Assess the consistency of materials in the The Taguchi approach to quality engineering places a
experiment. great deal of emphasis on minimizing variation as the
ii. Check the measurement systems for the main means of improving quality. The idea is to design
experiment. products and processes whose performance is not
iii. Test the experimental procedures and ensure that affected by outside conditions and to build this in
operators perform them correctly. during the development and design stage through the
iv. Check that the different combinations of factor use of experimental design. The method includes a set of
levels can be run safely. tables that enable main variables and interactions to be
v. Obtain preliminary estimates of variation. investigated in a minimum number of trials. Taguchi
Method uses the idea of Fundamental Functionality,
Control Chart example which will facilitate people to identify the common
goal because it will not change from case to case and
The control chart is a graph used to study how a process can provide a robust standard for widely and frequently
changes over time. Data are plotted in time order. A changing situations. It is also pointed out that the
control chart always has a central line for the average, Taguchi Method is also very compatible with the human
an upper line for the uppercontrol limit and a lower line focused quality evaluation approaches that are coming
for the lower control limit. These lines are determined up.
from historical data.
V. ACKNOWLEDGEMENT

I take this opportunity to express my true sense of

gratitude towards my seminar guide Prof. S. N. Kadam
who at every discrete step in the study of this seminar,
contributed with his valuable guidance and provided
with perfect solution for every problem that arose.

VI. REFERENCES
Figure 1. Signal to Noise Ratio [1] Taguchi, G., Wu, Y.Introduction to off-line Quality Control
Nagaya, Japan: Central Japan Quality Control Organization,
1979’
[2] Taguchi, G. Introduction to Quality Engineering Tokyo, Japan:
Asian Productivity Organization, 1986
[3] Khan, R. M. (2013). Problem solving and data analysis using
minitab: A clear and easy guide to six sigma methodology (1st
ed.). West Sussex, United Kingdom: Wiley.
[4] Phadke M.S., 1989. Quality engineering using robust design.
Englewood Cliffs, NJ: Prentice-Hall.
[5] Montgomery D.C., 1991. Design and analysis of experiments.
John Wiley and Sons, NewYork.
[6] Singh H & Kumar P, Quality Optimization of Turned Parts
Figure 2: Example of SN ratio graph (EN24) by Taguchi Method, Productivity, Vol. 44 (1),(2003)
pp. 43-49.
[7] Singh H & Kumar P,Effect on Power Consumption forTurned
 Taguchi's emphasis on minimizing deviation from
parts using Taguchi Technique, Productivity, Vol.45 (2), July-
Sept,(2004).
target led him to develop measures of the process [8] Singh H & Kumar P, Optimizing cutting force for turnedparts
output that incorporate both the location of the by taguchi parameter design approach, Indian Journalof
Engineering and Material Science,Vol-12, April (2005)
Minitab’s Help Section.

International Journal of Scientific Research in Science, Engineering and Technology (ijsrset.com)

845

View publication stats