implementation of taguchi methodology for …iaeme.com/masteradmin/uploadfolder/quality...

International Journal of Industrial Engineering Research and Development (IJIERD), ISSN 0976 –

6979(Print), ISSN 0976 – 6987(Online) Volume 2, Issue 1, May - October (2011), © IAEME

1

IMPLEMENTATION OF TAGUCHI METHODOLOGY FOR

DEFECT REDUCTION IN MANUFACTURING INDUSTRY “A

CASE STUDY”

Rajender Kumar

Student (M.Tech.)

Dept. of Mechanical Engineering

PEC University of technology

(Formerly Punjab Engineering College)

Sector-12, Chandigarh (India)-160012

E- mail: [email protected]

Dr. D. R. Prajapati

Assistant Professor in the department of Mechanical Engineering,

PEC University of Technology,

Chandigarh-160012 (India).

E-mail: [email protected]

Sukhraj Singh

Research Scholar in the department of Mechanical Engineering,

PEC University of Technology,

Chandigarh-160012 (India).

E-mail: [email protected]

ABSTRACT

Quality improvement technique, especially Taguchi methodology is used for designing

the experiments to investigate the effects of different parameters on the mean and

variance of process performance characteristics. In this paper, the Taguchi methodology

is used for the rejection control of flexible hose assembly in an automobile industry. An

attempt is made to correlate the effect of process parameters on the final product.

Experiments are carried out to validate the results, obtained by the implementation of

Taguchi methodology in automotive process industry and the most important parameters,

affecting the product life is found out. The optimal levels of the process parameters in

order to yield the optimum quality characteristics of the products are obtained. An

orthogonal array, the signal to noise (S/N) ratio, and analysis of variance (ANOVA) are

used to analyze the effect of selected process parameters and their levels. The results

International Journal of Industrial Engineering Research

and Development (IJIERD), ISSN 0976 – 6979(Print)

ISSN 0976 – 6987(Online) Volume 2

Issue 1, May – October (2011), pp. 01-14

© IAEME, http://www.iaeme.com/ijierd.html

IJIERD © I A E M E



2

indicate that the selected process parameters significantly affect the defects of flexible

hose assembly. The optimized process parameters i.e. lower crimping depth of 0.35 mm

and cap inner diameter (oversize) of 16 mm is obtained and which lead to minimize the

crimping leakage defects.

Keywords: Taguchi methodology, Cause and effect diagram, ANOVA, S/N Ratio,

Manufacturing and Assembling Process.

1.0 INTRODUCTION

Over the past few years, Taguchi Methodology has been espoused by many world-class

companies and has also a lot of successful cases. The main benefit of a Taguchi

methodology is the elimination of subjectivity in decision-making by creating a system

where everyone in the organization collects, analyzes and displays data in a consistent

manner. For manufacturing companies, the direct benefit of Taguchi methodology results

from the reduction in the number of defects due to improved manufacturing processes.

For these companies, high quality level is a measure of the process defect rate and thus

can be used to measure the quality of the manufacturing process. In the Taguchi method,

the results of the experiments are analyzed to achieve the objectives:

(i) To establish the best or optimal condition for the product or process,

(ii) To establish the contribution of individual factors, and

(iii) To estimate the response under optimal conditions.

Taguchi’s contributions to quality engineering include loss function associated with a

product or process, robust design and simplified statistical experiments using orthogonal

arrays. Taguchi methodology states that even the best available manufacturing

technology by itself is not an assurance that the final product will actually function in the

hands of its users as desired and so strongly advocated for the engineered products with

robust performance. Taguchi described entire concept in two basic ideas, namely, quality

should be measured by the deviation from a special target value rather than by

conformance to preset tolerance limits and quality cannot be ensured through the

inspection and rework, but must be built-in, through the appropriate design of the process

and product. The first concept underlines the basic difference between Taguchi methods

and statistical process control (SPC) methodology. The SPC emphasizes the attainment of

an attribute within tolerance range while Taguchi methods emphasize the attainment of

the specified target value and the elimination of variation.

The characteristics can be controlled, such that a lower or a higher value in a particular

quality-influencing factor produces the preferred result. Thus, the levels of influencing

factors, to produce the best results, can be predicted. There are two different

methodologies in carrying out the complete Orthogonal Analysis. A common approach is

to analyze the average result of repetitive runs or a single run through ANOVA analysis,

as discussed above. The other approach, which is a better method for multiple runs, is to

use signal to noise (S/N) ratios for the same steps in the analysis. The objective of S/N

analysis is to determine the most optimum set of the operating conditions from variations

of the influencing factors within the results. The signals, in this case, will be those

factors, which are invariant.



3

2.0 LITERATURE REVIEW

Taguchi (1986) has introduced several new statistical tools and concepts of quality

improvement that depend heavily on the statistical theory of experimental design. Some

applications of Taguchi’s methods in the foundry industry have shown that the variation

in casting quality caused by uncontrollable process variables could be minimized. Otto

and Antonsson (1991) extended the Taguchi’s method, which involved variables, each of

which having a range of values which might be used for analysis. Tuning parameters, as a

part of the design process, were also demonstrated within Taguchi’s method. They also

extended this method to solve design problems with constraints, involving the methods of

constrained optimization. They compared their design with other methods of searching

the design space and their definitions of an optimal solution. Ghani et al. (2004) used the

Taguchi method in the optimization of end milling process parameters. They applied it to

optimize the cutting parameters in end milling when machining hardened steel with a tin-

coated carbide insert tool under the semi-finishing and finishing conditions of high-speed

cutting.

Shaji and Radhakrishnan (2003) performed an analysis of the process parameters in

surface grinding with graphite as the lubricant based on the Taguchi method. They dealt

with the analysis of the process parameters such as speed, feed, in feed and mode of

dressing as influential factors on the force components and surface finish developed, and

based on Taguchi’s experimental design methods. Vlachogiannis and Roy (2005) worked

on fine-tuning of proportional integral derivative (PID) controllers, under model

parameter uncertainties (noise), using the Taguchi method. The Taguchi method provided

sub-optimal values for fine PID tuning in the presence of model parameter uncertainties

(noise). They also calculated the contribution of each factor to the variation of the mean

and the variability of error. The expected cost savings for PID under optimum condition

were calculated. They also performed confirmation experiments on a real PID controller.

Esme (2009) investigated the effect of welding parameters on the tensile shear strength in

the resistance spot welding (RSW) process. The experimental studies were conducted

under varying electrode forces, welding currents, electrode diameters, and welding times.

The level of importance of the welding parameters on the tensile shear strength was also

determined by using analysis of variance (ANOVA). The optimum welding parameter

combination was obtained by using the analysis of signal-to-noise (S/N) ratio. Their

confirmation tests indicated that it is possible to increase tensile shear strength

significantly by using the Taguchi method. The experimental results confirmed the

validity of the used Taguchi method for enhancing the welding performance and

optimizing the welding parameters in the resistance spot welding process. Mahapatra and

Chaturvedi (2009) work on the modelling and analysis of behaviour of composites

reinforced with short fibre drawn from agricultural resources. The optimum size of short

fibre, just capable of transferring the load and flexibility during preparation had been

studied through a systematic modelling approach and empirical model. The effect of

various test parameters and their interactions had been studied to find out optimal



4

parameter setting for minimum wear (weight loss). The recommended fibre length of 7-8

mm for minimum wears of the composites.

Badkar et al. (2010) presented the application of Taguchi method and its utility concept

for optimizing the laser process parameters in laser transformation hardening of

commercially pure titanium using a continuous-wave 2-kW, Nd:YAG laser. They used

Taguchi tools, such as analysis of variance (ANOVA), signal-to-noise ratio, and additive

model had been used to analyze and evaluate the optimum combination levels of laser

transformation hardening process parameters. The optimization results revealed that a

combination of higher levels of scanning speed (SS) and FP focused position (FP), i.e.,

increase in defocused beam with negative focal length along with laser power (LP) in the

lower level, is an essential laser hardening parameter to simultaneously minimize the HD

and maximize the HBW.

Elangovan and Narayanan (2010) found that the formability of perforated sheet depend

on chemistry of the material, the forming parameters, dimensions of the perforations and

ligament width. The forming limit diagram for perforated Al 8011 sheet of 1mm

thickness had been evaluated and the influences of the diameters of perforations and

ligament widths on the forming limit strains had been studied. Wazed et al. (2010)

measured the individual and combined impacts of common parts and machines in

manufacturing, under bottleneck and uncertain conditions. They used machine

breakdown and quality variation to create uncertainty. The authors examined the delivery

performances such as (i) throughput of the finished products, (ii) average production time

and (iii) work-in-progress in the system for different experimental scenarios. They

developed the simulation models, based on a live case from a Malaysian company. They

concluded that batch size of 12 in bottleneck, 2 common parts and 4 common machines

ensure the best outcomes of the system under the storm of uncertainties. The main

contribution of their research is to find out the best batch size in bottleneck point under

uncertainties, commonalities and capacity constraint.

3.0 INTRODUCTION OF INDUSTRY AND PRODUCT

It is the pioneer industry in the manufacturing of automobile tubes in India and was

established in 1969. It supplies its products to New Holland, Suzuki, DCM Toyota,

Nissan, and Mahindra. It also manufactures automotive hoses and tubes. It signed up a

joint venture with TRI (Tokai Rubber Industries of Japan) in 2005. It is ISO-9000 and

ISO-14001certified since 2003. It supplies Pipe Assemblies, Hose Assemblies, Tubing’s,

Multi Layer Hoses, Stainer Assemblies, Molded Components, and Pipes etc. for all kinds

of worldwide automotives products. Flexible hose assembly pipes are used in auto motive

vehicles for fuel and air supply.

3.1 In-House Rejection

The main problem is associated with the crimping joint failure in the flexible hose

assembly. The in-house rejection data from April 2009 to March 2010 is considered for

this industry. A study is carried out on this internal rejection data and during the analysis.



5

It is observed that the main causes for the rejection are Crimping Leakage, Banjo Face

Leakage, Pipe Leakage, Banjo Fault, Brazing Leakage, Crimping Fault, Crack (Nut,

Nipple and Pipe), Hose Puncture, imperfect Fitment etc. Table 1 shows the various

defects, contributing the rejection of Flexible Hose Assembly, as given below.

Table 1: Defect wise in-house rejection of Flexible Hose Assembly

Causes Rejection Quantity

(in Nos.)

Rejection

(in percentage)

Crimping Leakage 1598 59.16

Banjo Face Leakage 587 21.73

Pipe Leakage 147 05.44

Banjo Fault 134 04.96

Brazing Leakage 125 04.62

Crimping Fault 64 02.36

Crack (Nut, Nipple & Pipe) 21 00.77

Hose Puncture 13 00.48

Improper Fitment 12 00.46

TOTAL 2701 100

Table 1 shows that the leakage in the Crimping Joint is the major defect which,

contributing about 59% in the rejection of Flexible Hose Assembly. The cause & effect

diagram for the crimping leakage defect is prepared for finding the possible causes for

leakage from crimping joint, as shown in Figure 1.

Figure 1: Cause and effect diagram of crimping joint failure

Unskilled Worker Cap ID Oversize

Absenteeism Improper Material

Less Crimping Pressure Crimping die not checked

Less Crimping Force

Crimping die worn out

Variation in Insertion Length Crimping joint fouling

of pipe with sharp corner of the Tank

Crimping Joint

Failure

Causes of

Failure

Personnel, Staffing, etc Materials, Policies, Regulations, etc

Procedures, Methods, Specs, etc. Plant, Machines and Equipments,

etc.



6

3.2 Monthly rejection details

Further it is analyzed that one of the major defect for internal rejection is the leakage

from the Crimping Joint. Table 2 shows the rejection quantity in parts per million (PPM)

and monthly percentage rejection as given below.

Table 2: Monthly crimping leakage rejection

Month Rejection Quantity Total Production Parts Per Million

(PPM)

Percentage

Rejection

APR’09 126 126700 994 0.0994

48 MAY’09 128 104999 1219 0.1219

06 JUN’09 136 121409 1120 0.1120

18 JUL’09 142 120545 1178 0.1177

98 AUG’09 155 141523 1095 0.1095

23 SEP’09 138 131710 1048 0.1047

76 OCT’09 135 144045 937 0.0937

21 NOV’09 136 150705 902 0.0902

43 DEC’09 118 115959 1018 0.1017

6 JAN’10 124 144493 858 0.0858

17 FEB’10 132 130767 1009 0.1009

43 MAR’10 128 141878 902 0.0902

18 TOTAL 1598 157473

3 12282

4.0 ANALYSIS

The Probable causes of the Crimping leakage problem are found from cause and effect

diagram and listed below.

1. Less crimping pressure

2. Less Insertion length of pipe hose/ Hose in cap

3. Cap Inner Diameter oversize

4. Uneven crimping depth

5. Improper Materials

6. Crimping operation not performed.

4.1 Application of Taguchi Approach

Taguchi methodology is used to find the optimal solution regarding causes for rejection.

Three main defects: Less Crimping Depth, Hose Length Variation and Cap Inner

Diameter (Oversize) are considered for the Crimping Leakage of Hose Pipe Assembly.

Table 3 shows the process parameter with their ranges, as shown below.



7

Table 3: Process parameters with their ranges and values at three

levels Parameter designation Process parameter Range

A Less Crimping Depth (mm) 0.35-0.4

B Hose Length Variation (mm) 18-22

C Cap I.D. Oversize (mm) 15.5-16

4.2 Selection of Orthogonal Array

Before selecting the particular orthogonal array, two major factors are considered:

1. The number of parameters and interaction of interest,

2. The number of levels for the parameters of interest.

In this case study, the L8 orthogonal array is considered, as shown in Table 4.

Table 4: L8 orthogonal array Trial

No. A B A*B C A*C B*C A*B*C

1 1 1 1 1 1 1 1

2 1 1 1 2 2 2 2

3 1 2 2 1 1 2 2

4 1 2 2 2 2 1 1

5 2 1 2 1 2 1 2

6 2 1 2 2 1 2 1

7 2 2 1 1 2 2 1

8 2 2 1 2 1 1 2

5.0 EXPERIMENTATION

Once the parameters and interaction of parameter are assigned to a particular column of

the selected orthogonal array, the factors at different levels are assigned for each trial.

The experiments are conducted twice for the same set of parameters, using a single

repetition randomization technique. The crimping leakage defects occurring in each trial

conditions are measured. The flexible hose assembly is made against the trial conditions,

as given in Table 5.

Table 5: Experimental L8 Array Trial No. A B C

Less Crimping

Depth (mm)

Hose Length Variation

(mm)

Cap I.D. Oversize

(mm)

1 0.35 18 15.5

2 0.35 18 16.0

3 0.35 22 15.5

4 0.35 22 16.0

5 0.4 18 15.5



8

6 0.4 18 16.0

7 0.4 22 15.5

8 0.4 22 16.0

The average of the crimping leakage defects is computed for each trial condition and it is

shown in Table 6.

Table 6: Percentage defects in Experiments Trail

No.

Percentage Defects in Experiments

Level 1 Level 2

1 16.5 17.3

2 18.4 19.2

3 16.8 17.5

4 17.3 19.1

5 18.4 21.2

6 22.4 23.2

7 18.2 21.3

8 19.3 22.2

5.1 Sample Calculations for Signal to Noise Ratios

The crimping leakage defects are “Lower the Better” type of quality characteristics.

Lower the better S/N ratios are computed for each of the 8 trials and sample calculations

are also given as under.

Lower is better: S/NLB ratio = -10 log [(Σy2i)/n]

For the case of minimizing the performance characteristic, the following values of the

S/N ratio are calculated as:

For the case of maximizing the performance characteristic, the S/N ratio is calculated as:

Where, n is the number of observation and yi are the different experimental values for

various trials.

Sample calculations

For Example: Trial No., Table Number 7;

Percentage sum of defects = percentage defect at level 1 (16.5) + percentage defect in

level 2 (17.3) = 33.8.

Average percentage defect = percentage sum of defects/ 2 and

Average S/N ratio= -10log (16.52+17.3

2)/2 = -24.56



9

Table 7 shows the Crimping defect values and S/N ratios as shown below.

Table 7: Crimping defect values and S/N ratio against trial number

Percentage Defects in Experiment

Trial No. 1 2 Sum Average S/N Ratio

1 16.5 17.3 33.8 16.9 -24.56

2 18.4 19.2 37.6 18.8 -25.48

3 16.8 17.5 34.3 17.15 -24.68

4 17.3 19.1 36.4 18.2 -25.21

5 18.4 21.2 39.6 19.8 -25.95

6 22.4 23.2 45.6 22.8 -27.16

7 18.2 21.3 39.5 19.75 -25.93

8 19.3 22.2 41.5 20.75 -26.36

Total 147.3 161 -205.33

5.2 Other sample Calculations

1. Correction Factor (C.F.)

Correction Factor (C.F.) = (Sum Total) 2/ No.

of Observations

= (16.5+17.3+…………..+22.2) 2 / 16

C.F. = 5940.56

2. The total sum of squares

SStot = Y2

- C.F.

= (16.5) 2

+ (17.3) 2

+………(22.2) 2

– 5940.56

SStot = 69.0344

3. Average value for casting defects and S/N ratio at different levels

The mean response, referring to the average value of the performance characteristics (the

crimping leakage defects and S/N ratios) for each parameter at different levels is given in

Table 8. The Figure 2 shows the graphical representation of the crimping leakage defects

at different levels.

Table 8: Average value for crimping defects and S/N ratios at different levels

Factors

Average value of casting defects and S/N ratio at different levels

Level 1 Level 2

Crimping Defect S/N ratio Crimping Defect S/N ratio

A 17.76 -24.98 20.78 -26.35

B 19.58 -25.79 18.96 -25.55

C 18.4 -25.28 20.14 -26.05



10

Figure 2: Average value of Crimping Defect on different levels

17.76

20.78

19.58

18.96

18.4

20.14

17

17.5

18

18.5

19

19.5

20

20.5

21

A1 A2 B1 B2 C1 C2

AVERAGE VALUE OF CRIMPING DEFECT ON

DIFFERENT LEVEL'S

D

E

F

E

C

T

% Series1



11

Figure 3 shows the graphical representation of the average values of S/N ratio at

different levels.

4. Calculation of Sum of Squares for different factors

SSA = (Sum total in level 1)2 + (Sum total in level 2)

2 - C.F.

No. of Observations

= 36.3

Similarly, the sums of squares for other factors are calculated, as shown in Table 9.

Table 9: Sum of Squares for various factors Sr. No. Source Sum of Squares

1 A 36.3

2 B 1.5

3 A*B 0.77

4 C 12.08

5 A*C 0.28

6 B*C 2.03

7 A*B*C 0.33

SStot 53.29

Calculation of the error sum of squares

SSe = SStot – ( SSA+ SSB+…………………………+SSD)

= 69.034 – 53.294

= 15.74

-24.98

-26.35

-25.79

-25.55

-25.28

-26.05

-26.5

-26

-25.5

-25

-24.5

-24

A1 A2 B1 B2 C1 C2

Average value of S/N ratio at different level's

Sig

nal to

No

ise R

ati

o (

S/N

)



12

5.3 Analysis of Experimental Results

Table 10 shows the ANOVA table for crimping leakage defect in flexible hose assembly.

Table 10: ANOVA TABLE for crimping defects in flexible hose

assembly ANOVA for crimping defects

Source Sum of

Square

(SS)

Degree of

Freedom

Variance F ratio Results Percentage

Contribution

A 36.3 1 36.3 18.4498 Significant 52.58

B 1.5 1 1.5 0.76239 0.021

A*B 0.77 1 0.77 0.39136 0.011

C 12.08 1 12.08 6.13977 Significant 17.49

A*C 0.28 1 0.28 0.14231 0.004

B*C 2.03 1 2.03 1.03177 0.029

A*B*C 0.33 1 0.33 0.16773 0.004

Error (e) 15.74 8 1.9675

Total 69.03 15

6.0 RESULTS

It is clear from the ANOVA table that the Less Crimping Depth (mm) and Cap Inner

Diameter (Oversize) have significant effect on the Flexible Hose Assembly. The

optimum levels for these factors can be obtained by examining the level averages of the

factors, as shown in Table 11.

Table 11: Affecting Parameters value at the different levels average Factors Level Average

1 2

Less Crimping Depth (mm) 17.76 20.78

Oversized Cap I.D. (mm) 18.4 20.14

Various experiments are conducted on flexible hose assembly product for this case study.

ANOVA, along with interpretation method is used to obtain the percentage contribution

of each parameter and optimum levels of each parameter are shown in Table 12.

Table 12: Process parameter optimum level with optimal value Parameter

designation

Process parameter Optimum

level

Optimum

value

Percent

Contribution (%)

A Less Crimping Depth (mm) 1 0.35 52.58

C Cap I.D. Oversize (mm) 2 16 17.49



13

7.0 CONCLUSIONS

The contribution of individual quality influencing factors is the deciding key of the

control to be enforced on the product design. A commonly applied statistical treatment -

The Analysis of Variance (ANOVA) is used to analyze the results of the Orthogonal

Array (OA) experiment in product design and to determine how much variation each

quality-influencing factor has contributed. By studying the main effects of each one of

the factors, the general trends of the influencing factors towards the product or process

can be characterized.

Table 12 clearly shows that two most significant parameters affect both the mean and

variation of the crimping leakage defects and also shows the optimal settings of each

parameter to reduce the crimping leakage defects and hence improves the quality of

flexible hose assembly at lowest cost. Before the application of Taguchi’s method, the

parameters of the crimping leakage process were more arbitrary and difficult to control

and, of course, the product quality has instability problems. Taguchi’s method yielded

optimized control factors, resulting in superior product quality and stability. From the

analysis, it is found that the improvement in the quality at the lowest possible cost can be

achieved by Taguchi’s method of parameter design. It is also possible to identify the

optimum levels of signal factors at which the noise factor’s effect on the response

parameter is minimized. The optimized process parameters i.e. lower crimping depth of

0.35 mm and cap inner diameter (oversize) of 16 mm are obtained by Taguchi

methodology, which lead to minimize the crimping leakage defects.

The confirmation experiments also indicate a clear picture of every factor’s contribution

to the variation in the flexible hose assembly manufacturing process and improvement in

the quality without any additional investment.

REFERENCES

1. Badkar, D.S.; Pandey, K. S. and Buvanashekaran, G. (2010),“Parameter

optimization of laser transformation hardening by using Taguchi method and

utility concept,” International Journal of Advanced Manufacturing Technology,

Volume 1, Number 1, pp. 123-135.

2. Elangovan, K. and Narayanan, C.S. (2010),“Application of Taguchi approach on

investigation of formability for perforated Al 8011 sheets,” International Journal

of Engineering, Science and Technology, Volume 2, Number 5, pp. 300-309.

3. Esme, U. (2009),“Application of Taguchi Method for the optimization of

resistance spot welding process,” The Arabian Journal for Science and

Engineering, Volume 34, Number 2B, pp. 519-528.

4. Ghani, J.A.; Choudhury, I.A. and Hassan, H.H. (2004),“Application of Taguchi

method in the optimization of end milling parameters,” International Journal of

Materials Processing Technology, Volume 145, Issue 1, pp. 84-92.

5. Krishnamoorthi, K.S. (2006),“A First Course in Quality Engineering Integrating

Statistical and Management Methods of Quality”, New Jersey: Prentice Hall,

Pearson, Upper Saddle River.



14

6. Mahapatra, S.S. and Chaturvedi, V. (2009),“Modelling and analysis of abrasive

wear performance of composites using Taguchi approach,” International Journal

of Engineering, Science and Technology, Volume 1, Number 1, pp. 123-135.

7. Otto, K. N. and Antonsson, E. K. (1991),“Application of Taguchi method to

analyze the impacts of commonalities in multistage production under bottleneck

and uncertainty,” International Journal of the Physical Sciences, Volume 5,

Number 10, pp. 1576-1591.

8. Shaji, S. and Radhakrishnan, V. (2003),“Analysis of process parameters in surface

grinding with graphite as lubricant based on the Taguchi method,” Journal of

Materials Processing Technology, Volume 141, Issue 1, pp. 51-59.

9. Taguchi, G. (1986) “Introduction to Quality Engineering,” Proceeding on Asian

Productivity Organizations, Malaysia, pp. 132-147.

10. Vlachogiannis, J. G. and Roy, R. K. (2005),“Application of Taguchi method to

analyze the impacts of commonalities in multistage production under bottleneck

and uncertainty,” The TQM Magazine, Volume 17, Number 5, pp. 456-466.

11. Wazed, M. A.; Ahmed, S. and Nukman, Y. (2010),“Application of Taguchi

method to analyze the impacts of commonalities in multistage production under

bottleneck and uncertainty,” International Journal of the Physical Sciences,

Volume 5, Number 10, pp. 1576-1591.

implementation of taguchi methodology for …iaeme.com/masteradmin/uploadfolder/quality...

Documents