[9]
CHAPTER 2
LITERATURE REVIEW
2.1 LITERATURE REVIEW ON SINGLE OBJECTIVE OPTIMIZATION
THROUGH TAGUCHI’S METHOD
Taguchi’s parameter design offers a systematic approach for optimization of various
parameters with regard to performance, quality and cost (Phadke, 1989). The quality
of design can be improved by improving the quality and productivity in various
company-wide activities. Those activities concerned with quality include in quality of
product planning, product design and process design (Park 1996, Ranjit 2001).
Taguchi’s parameter design approach can reduce number of experiments to optimize
design for performance, quality and cost. Signal to Noise(S/N) ratio and orthogonal
array (OA) are two major tools used in robust design. S/N ratio measures quality with
emphasis on variation, and OA accommodates many design factors simultaneously
(Park 1996, Phadke 1998).
Taguchi method offers the quality of product is measured by quality characteristics
such as: nominal is the best, smaller is better and larger is better (Phadke, 1998 &
Ranjit, 2001).
Das et al. (1997a and 1997b), Choudhury and Apparao (1999) and Choudhury et al.
(1999) have developed different models for optimization of process parameters using
different responses such as surface roughness, tool wear, vibrations etc.
Antony J (2001) presents a step by step approach to the optimization of production
process of retaining a metal ring in a plastic body by a hot forming method through
the utilisation of Taguchi methods of experimental design and achieves the good
results.
[10]
Park and Yum (2003) developed a procedure with the help of Taguchi’s method for a
dynamic parameter design problem and explained it with the help of an illustrated
example.
Gopalsamy B.M. et al. ( 2009 ), applied Taguchi method to find out the optimum
machining parameters while hard machining of hard steel and uses L18 orthogonal
array, S/N ratio and ANOVA to study the performance characteristics of machining
parameters which are cutting speed, feed, depth of cut and width of cut while
considering surface finish and tool life as response. Results of the study obtained by
Taguchi method match closely with ANOVA and cutting speed is the most
influencing parameter.
Rajendrakumar (2011) focuses on a design of experiment based approach to obtain an
optimal setting of turning process parameters (cutting speed, feed rate and depth of
cut) that may yield optimal tool flank wear and subsequent optimal settings of the
parameters and it have been accomplished with using Taguchi’s parameter design
approach.
Ficici F. et al. (2011) uses the Taguchi method to study the wear behaviour of
boronized AISI 1040 steel. They use orthogonal array, S/N ratio and ANOVA to
investigate the optimum setting parameters. The control factors used here are
boronizing time, applied load, sliding distance and sliding speed with weight loss as
response variables. The study show that the boronizing time had the greatest effect on
the wear followed by sliding distance.
Feng and Wang (2002) develops an empirical model for the prediction of surface
roughness in finish turning while considering working parameters like material, feed
rate, cutting tool point angle, depth of cut, spindle speed and cutting time.
Gusri et al. (2008) applied Taguchi optimization methodology to optimize cutting
parameters in turning Ti-6Al-4v ELI with coated and uncoated cemented carbide
tools. They show that the cutting speed and type of tool have a very significant effect
on the tool life, and the feed rate and type of tool have also a very significant effect on
the surface roughness.
Fnides et al. (2008) conducted tests on X38CrMoV5-1 steel treated at 50 HRC,
machined by a mixed ceramic tool to study the influence of the following parameters:
[11]
feed rate, cutting speed, depth of cut and flank wear on cutting forces and on surface
roughness.
Shinde et al. (2011) focuses on the effect of different machining parameters on
surface finish during turning operation. They consider cutting speed, feed rate and
depth of cut as machining parameters.
Kaladhar et al. (2012) applied Taguchi method to determine the optimum process
parameters for turning of AISI 304 austenitic steel on CNC lathe. They conducted
tests at four levels of cutting speed, feed and depth of cut. The influence of these
parameters are investigated on the surface roughness and material removal rate
(MRR). The results revealed that cutting speed significantly affects the surface
roughness followed by noise radius while depth of cut affects the MRR most followed
by cutting speed.
Rodrigues et al. (2012) proposes a study for the effect of cutting speed, feed rate and
depth of cut on surface roughness and cutting force while turning mild steel using
high speed steel (HSS) cutting tool. Experiments were conducted on a precision
centre lathe and the influence of cutting parameters on surface roughness and cutting
force was studied with the help of analysis of variance (ANOVA) based on adjusted
approach and also used linear regression analysis.
Petropoulos et al. (2005) develops a predictive model for cutting force components in
longitudinal turning of constructed steel with a coated carbide tool. Taguchi method is
used for the plan of experiments and the analysis is performed using response surface
methodology. Lastly a comparison was attempted to the result obtained with the help
of help of a well established semi-empirical and cutting resistance based Kienzle-
Victor model.
Singh and Kumar (2005 & 2006) obtain an optimal setting of turning process
parameters (cutting speed, feed rate and depth of cut) resulting in an optimal value of
the cutting force and feed force when machining EN24 steel with Tic-coated Tungsten
carbide inserts using Taguchi’s parameter design approach. They uses Taguchi’s L27
orthogonal array, signal to noise ratios (S/N) and analysis of variance (ANOVA) for
the study.
[12]
Kosaraju et al. (2012) investigate the effect of process parameters (cutting speed, feed
rate and depth of cut) on machinability performance characteristics and there by
optimization of turning of Titanium Grade 5 based on Taguchi’s L9 orthogonal array,
signal to noise ratios (S/N) and analysis of variance (ANOVA). The cutting speed was
identified as the most influential machining parameter on cutting force and
temperature.
Chorng et al., (2009) report that only the cutting speed affects significantly the
roundness of a cylindrical bar while the others are not. Rico et al., (2010) also
concludes the same result and also report that cutting speed-feed rate interaction and
cutting speed-depth of cut interaction significantly affects the roundness of cylindrical
bar.
Cicek et al., (2012) performed experimental trials using Taguchi orthogonal arrays to
obtain optimum surface roughness and roundness error values in the drilling of AISI
316 austentic stainless steel with untreated and treated drills and it was found that the
cutting speed had a significant effect on the surface roughness and that the cutting
speed and feed rate had a significant effects on the roundness error.
Sahoo and Sahoo (2011) develop a mathematical model for surface roughness while
turning tool steel using response surface methodology coupled with Taguchi design of
experiment. Taguchi’s S/N ratio and response surface methodology shows that the
feed rate is the most influencing parameter for surface roughness followed by depth of
cut and cutting speed has the less effect on the surface roughness.
Reddy and Valli (2011) shows the effect of process parameters on the machining of
EN-31 tool steel with copper as a tool in the rotary electrical discharge machining
process (EDM) with help of linear regression analysis and Taguchi’s method.
Material removal rate, tool wear ratio and surface roughness was used as response
variables. Results showed that MRR, TWR and SR was greatly influenced by peak
current. Experimental results also confirmed that simultaneous optimization of MRR,
TWR an SR was not possible for a given set of control factors.
Venkataramaiah (2011) used the Taguchi’s method with Grey relational analysis for
finding the optimum levels for the parameters which influence the production yield in
[13]
a foundry unit which manufactures cover plates. It is evident from the results that
there is consistency in the product quality with considerable yield.
Upadhye and Keswani (2012) used Taguchi’s method in sand casting process. The
parameters which were considered are moisture percentage, green compression
strength, mould hardness number and permeability. The expected improvement in
reduction in casting defects was found to be 40.82 percent.
Das et al. (2013) conducted a experimental study to investigate the effect of cutting
parameters on tool wear, surface roughness and material removal rate during the dry
turning of EN-31 steel. They also uses multiple regression analysis to develop a
relationship between cutting parameters and response variables which can be used to
estimate the values of response variables for any level of control parameters.
Modgil et al. (2012) presents a robust parameter design through Taguchi’s method
which has shown a breakthrough improvement in purity percentage of chemical X.
The means signal to noise ratio and standard deviation are predicted for optimal
setting and validated by producing 15 batches of inorganic chemical X with optimal
setting.
Kumar et al. (2013) uses Taguchi’s method and analysis of variance to study the
performance characteristics in turning operations. By using cutting speed of 150
m/min, depth of cut 15 mm and feed rate 0.15 mm/revolution, the optimum tool wear
was found as 0.142 which is close to the experimental value of 0.156.
Ultrasound based sonication process was used for deriving the nano-crystals of
sirolimus in a narrow range. Seven critical process were selected with three levels and
optimized with Taguchi’s L18 orthogonal array design. Detailed statistical analysis
like t-test, regression analysis and descriptive statistics of the results have been carried
out (Gabdhi et al., 2012).
The Taguchi method was reported to alter the surface properties of commercial
Degussa P25 TiO2, which could used as visible light driven photocatalyst and was
investigated to determine the material characteristics with the use of Taguchi’s L9
orthogonal array (Su et al., 2011).
[14]
Gustavo C-A(1998) used Taguchi method to optimize DNA amplification finger
printing (DAF) . Quadratic loss function penalize deviation from predicted values and
L9 and L18 orthogonal array revealed the effects and interactions of amplification
reaction components and thermal cycling parameters. Here, Taguchi’s method holds
potential for as an optimization tool in molecular biology.
Demirci et al. (2011), conduct a study to achieve the fatigue life parameters of
GFR/epoxy filament wound composites pipe according to ASTM D 2992. Taguchi’s
L9 orthogonal array and S/N ratios was used and a stress level was determined to be
the most important parameters on fatigue life among the filament angle, surface
crack-depth ratio and stress levels.
Aghakhani (2011) explains proper selection of input welding parameters is necessary
to obtain a good quality weld and subsequently increase the productivity of process.
In this method, with the help of Taguchi’s method and regression analysis a
mathematical model was developed using parameters such as wire feed rate, welding
voltage, nozzle to plate distance welding speed and gas flow rate on weld dilution.
Taguchi’s method was also applied in optimization of abrasive wear behaviour of
FeCrC coating composite (Yildiz T. and Gur A.K., 2011). The effect of parameters
levels on mean lowest wear value were analysed by ANOVA and the optimum wear
resistance value was obtained with help of Taguchi’s method.
Taguchi’s method with L9 orthogonal array was used to optimize the fabrication of
bovine serum albumin (BSA) nanoparticle. Agitation speed , initial BSA
concentration, pH and temperature were considered as control parameters and
according to the Taguchi analysis temperature and agitation speed were the most
influencing parameters on the particle size (Jahanshahi et al., 2008).
Mehravar et al. (2011) optimized the fabrication of Lactablumin nanoparticle by
applying the Taguchi’s method with pH, temperature and agitation speed as process
variables. The nanoparticle size at the determined condition was less than 220 nm at
the optimal condition of pH 2.5, temperature 500
C and agitation speed 750 rpm.
Esme (2009) shows the application of Taguchi’s method and ANOVA in the
optimization of resistance spot welding process with electrode force, welding current,
electrode diameter and welding times as process parameters. The level of importance
[15]
of welding parameters on tensile shear strength was determined by using analysis of
variance (ANOVA).
Kim and Lee (2009) presents a systematic approach to determine the optimal process
parameters associated with hybrid welding (combination of laser beam and gas metal
arc welding) of aluminium alloy (AA5052-H32) using Taguchi’s method. Welding
direction, laser power, laser focus, voltage, wire feed rate, root opening balance and
travelling speed were considered as process parameters.
Dobrzanski et al. (2007) find the optimum parameters to produce Twintex (glass and
polypropylene) tubes by filament winding with fibres temperature, winding speed,
number of layers and roving as control parameters. Results show that the fibres
temperature is very significant parameter both in tensile strength and shear test.
Thakur et al. (2010) presents an experimental investigation for optimization of tensile
shear strength of resistance spot welding for galvanized steel using Taguchi’s method.
They used L27 orthogonal array, ANOVA and F test for determining the most
significant affecting the spot welding performance with welding current, welding
time, electrode diameter and electrode force were used as process parameters.
Taguchi’s method was also used to find out the optimal process parameters for an
injection moulding machine that was used to produce a consumer product (plastic
tray) from polypropylene plastic material. Orthogonal array, signal to noise ratio and
analysis of variance were employed to study the bending characteristics of tray under
constant load (Kamaruddin et al., 2004).
Rama Rao and Padmanabhan (2012) presents an experimental investigation of
electrochemical machining process of Al/5%SiC composites with voltage, feed rate
and electrolyte concentration as control factors and material removal rate as response
variable. Taguchi’s orthogonal array, signal to noise ratio (S/N ratio), ANOVA and
regression analysis were used to find out the optimum parameter levels.
Ballal et al. (2012) describes the use of Taguchi’s method to find a specific range and
combination of turning parameters which are cutting speed, feed rate and depth of cut
and surface finish, tool wear and material removal rate as response variables in
turning of brake drum of FG 260 gray cast iron material.
[16]
Kamaruddin et al. (2010) attempts the Taguchi’s method to improve the shrinkage of
plastic tray made from plastic blend of 75% polypropylene and 25% polyethylene by
optimizing the injection moulding parameters. The analysis of the results shows that
the optimal combination for low shrinkage are low melting temperature, high
injection pressure, low holding pressure, low holding time and low cooling time.
Taguchi robust design method with dynamic signal to noise ratio (S/N) and two step
optimization process can be used to optimize the head valve system for developing
more efficient and low noise compressor. Also this method can be applied to rotary,
scroll and other type of compressors (Park, 1996).
Hachicha et al. (2008) apply Taguchi method to the route selection problem as an
optimization technique to get back to the simple cell formation problem (CF) which
can be solved by other numerous procedures. In addition the main effect of the each
part and ANOVA were introduced as sensitivity analysis.
Thomas and Antony (2005) applies both the Taguchi and Shainin design of
experiment techniques to optimize the design of honeycomb composite joints. They
compared the operational effectiveness of Taguchi and Shainin DOE techniques
through their application to improve the joint strength of an aerospace structural
problem.
Vlachogiannis and Roy (2005) implemented the Taguchi’s method of robust design in
scientific field of automation control. The control factors affecting the performance of
PID (proportional integral derivative) controllers under noise conditions are regulated.
An example of the proposed method was demonstrates that given performance
criteria, Taguchi method can provide the optimal values for the fine PID tuning in the
presence of model parameter uncertainities.
Alagic (2008) present the application of design of experiment and Taguchi’s method
in the industrial condition while testing of gear oil pump. The experiment was
conducted on the basis of Taguchi’s L16 orthogonal array, on testing engine board
specially designed for gear pump performance measuring. Working pressure,
revolution per minute (rpm), working oil temperature, normal pressure angle and type
of oil were selected as control parameters and gear pump flow capacity as response
characteristics.
[17]
Mohamed et al. (2008) used the Taguchi’s method to identify the several factors that
may affect the performance of destination sequence distance vector (DSDV) routing
protocol and it was found that the traffic load has a stronger influence on pocket
delivery ratio followed by the pause time. And lastly optimal setting for the best
performance was determined.
Ponappa et al. (2010) conducted experiments using Taguchi methodology and
ascertained the effects of electrical discharge machining (EDM) process parameters
and studied the effects of EDM parameters on drilled-hole quality such as taper and
surface finish. Microwave–sintered magnesium nano alumina composites were used
as work materials.
Mohd Amri Lajis et al. (2009) applied Taguchi’s robust design approach to determine
the main effects, significant factors and optimum machining condition during the
performance of EDM while machining Tungsten carbide ceramics with a graphite
electrode.
Zhang et al. (2007) applied Taguchi’s method of optimization in order to find out the
optimal cutting conditions for end milling by varying cutting parameters. Results
show that Taguchi’s method is an effective method for optimizing the surface
roughness in end milling operation.
Muzammil et al. (2003) uses Taguchi’s method of robust design for optimization of
gear blank casting process and demonstrated that casting process involve a large
number of parameters affecting the various quality characteristics of the product. With
the help of signal to noise ratios optimum level of process parameters were obtained.
Shahin applied the Taguchi’s robust design technique to develop the weight loss
model of aluminium composites with 10wt. %SiC particles by molten metal mixing in
terms of grain size with the help of Taguchi’s orthogonal array and signal to noise
ratios (S/N).
Chen et al. (1996) presents the use of Taguchi’s method of experimental design in
optimizing the process parameters for micro-engraving of iron oxide coated glass
using a Q-switched Nd:YAG laser with beam expansion ratio, focal length, average
laser power, pulse repetition rate and engraving speed as control parameters and
[18]
engraving line width as response variable with the help of L16 orthogonal array,
signal to noise ratio and ANOVA.
Shukla and Shah (2010) investigated the effect of friction stir welding parameters
such as rotational speed, welding speed and axial force on tensile strength of 6061 T6
aluminium alloy. They use Taguchi’s L9 orthogonal array, signal to noise ratios and
analysis of variance (ANOVA) for this. The result shows that the rotational speed is
the dominant factor followed by welding speed and axial force shows the minimal
effect on the tensile strength.
[19]
2.2 LITERATURE REVIEW ON MULTI-OBJECTIVE OPTIMIZATION
THROUGH TAGUCHI’S METHOD
Taguchi method of robust design has significantly improved the quality and at the
same time reduces the cost. Most of the researchers have focused on optimizing a
single objective optimization while studying applications on Taguchi method. A
single setting of process parameters may be optimal for one response but the same
setting may yield detrimental results for other responses. In Such cases, a need arises
to obtain an optimal setting of the process parameters so that the product can be
produced with optimum or near optimum responses (Singh & Kumar, 2006).
Tong L. et al. (1997), proposed a procedure in this study to achieve the optimization
of multi-response problems in the Taguchi method which includes four phases, i.e.
computation of quality loss, determination of the multi-response S/N ratio,
determination of optimal factor/level condition and performing the confirmation
experiment.
Singh & Kumar (2006) discusses a case study on En24 steel turned parts using
titanium carbide coated tungsten carbide inserts. The multi-machining characteristics
have been optimized simultaneously using Taguchi’s parameter design approach and
the utility concept. The paper used a single performance index, utility value, as a
combined response indicator of several responses.
Al-Refaie et al. (2010) proposes a simple and very effective approach for solving the
multi-response problem in the Taguchi method. Each quality response is transformed
into signal-to-noise (S/N) ratio. The average S/N ratio is calculated for each factor
level, and then weighted with respect to the level of the largest average S/N ratio for
this factor. The average weight of each factor level is obtained from all responses. The
factor level with the largest level weight is selected as the optimal level foe that
factor.
Agastra et al. (2012) illustrated a novel multi-objective algorithm based on Taguchi’s
technique and its performance assessed. Results indicate a generally better behaviour
of the proposed algorithm in terms of convergence and spreading over the Pareto front
with respect to the GA (genetic algorithm) benchmark. This multi-objective technique
[20]
has proven to be a valid alternative to the popular genetic algorithm (GA) based
approach.
Kazancoglu et al. (2011) investigated the multi-response optimization of the turning
process for an optimal parametric combination to yield the minimum cutting forces
and surface roughness with the maximum material-removal rate (MRR) using a
combination of a Grey relational analysis (GRA) and the Taguchi method. Nine
experimental runs based on an orthogonal array of the Taguchi method were
performed to derive objective functions to be optimized within the experimental
domain. The Taguchi approach was followed by the Grey relational analysis to solve
multi-response optimization problem.
Surace et al. (2009) develop a method for the analysis of the effects of the foaming
parameters on the quality of foam parts and to determine their optimal combination.
The effects of the foaming parameters are studied by the Taguchi method, applied to
design an orthogonal array. A multi-objective optimization approach is then proposed
by simultaneously minimizing the relative density and maximizing the absorbed
energy efficiency. The corresponding Pareto set has been found and it can be used for
choosing a qualified solution for manufacturing problems.
Jeyapaul et al. (2006) used genetic algorithm (GA) approach along with the Taguchi’s
method of optimization to investigate the effect of different machining parameters on
multiple performance characteristics during a gear hobbing operation. The main
objective of the paper was to determine the optimum levels of selected machining
parameters which optimize the profile errors and helix errors in gear hobbing
operation.
Su and Tong (1997) also propose a methodology on the basis of combination of PCA
(Principle component analysis) and Taguchi’s method of optimization to optimize the
case of multi quality characteristics problems and achieve the good results. This
method consist of series of steps which are capable of decreasing the uncertainty in
the decision making.
Antony (2000) also presents a case study for optimizing multi-objective problems in
industrial experiments by using the combination of Taguchi’s loss function and
Principle component analysis (PCA).
[21]
Waghmare et al. (2011) presents the use of utility concept along with Taguchi
methodology to optimize the RSW machine setting for multiple quality
characteristics. Taguchi’s modified L16orthogonal array is used for experimentation
along with using a logarithmic scale for getting preference value and weightage is
provided to each quality characteristics as per customer requirement, to determine
overall utility.
Besseris (2008) proposes a simple methodology in solving multi-response
optimization problems by employing Taguchi methods and a non-parametric
technique. Here the concept of Super Rank (SR) was used. This paper provides a new
approach to the multi-objective optimization by supporting Taguchi’s method through
a super ranking concept that leads to non-parametric resolution.
Tong et al. (1997) applied Taguchi method towards a multi-response production
process. The proposed optimization procedure includes four phases which are capable
of decreasing the uncertainty in engineering judgement when the Taguchi method is
applied. The four phases are computation of quality loss, determination of multi-
response signal to noise ratio (MRSN), determination of optimal factor level
combination and lastly performing the confirmation experiment. The results of the
case studies indicates that the proposed procedure can achieve the optimization of
multi-response problems in the Taguchi’s method.
Liao and Chen (2002) propose a data envelopment analysis ranking (DEAR) approach
as an effective means of optimizing the multi-response problem. Includes a series of
steps from the proposed approach which are capable of decreasing uncertainty caused
by engineering judgment in the Taguchi method and overcoming the short comings of
PCA.
Antony et al. (2006) proposed a four step procedure to resolve the parameter design
problem involving multiple responses. This approach employs the advantage of both
artificial intelligence tool (neuro-fuzzy model) and Taguchi method of experimental
design to tackle problems involving multiple responses optimization.
Lan (2009) uses L9 orthogonal array of Taguchi experiment is selected for optimizing
the multi-objective machining for surface roughness, tool wear and material removal
rate (MRR) while turning through CNC. By using Technique for Order Preference by
[22]
Similarity to Ideal Solution (TOPSIS), the multiple objectives can additionally be
integrated and introduced as the signal to noise ratio which were analyzed by
MINITAB.
Wu (2008) presents a simple approach to optimizing multi-quality characteristics
based on quality loss function. This can be demonstrated by a numerical example of
poly-silicon deposition process for minimizing surface defects and achieve the target
thickness in a very scale integrated circuit. The proposed method does not require any
complicated calculations.
Kumar et al. (2013) presents a utility concept coupled with Taguchi’s method for
multi-response optimization during turning of unidirectional glass fiber reinforced
plastics composite using polycrystalline diamond cutting tool. Taguchi’s L18
orthogonal array was employed to carry out experimental work. The adequacy of
models was tested on output responses with the help of analysis of variance. Here an
attempt has also been made to optimize cutting parameters using multi-objective
characteristics for the developed prediction models using utility concept.
Wang et al. (2007) focus on optimizing the fused deposition modelling (FDM) of
rapid prototyping (RP) systems. The used main steps of the procedure were setting the
weight of each quality characteristics of the previous Taguchi method, obtaining the
multiple quality characteristics by integrating the Gray theory and finally obtaining a
set of optimal control parameters. This research demonstrates the how optimizing
FDM process can improve the low product strength, bad surface strength and high
dimensional errors.
Routara et al. (2010) coupled Taguchi’s method with utility concept in a multi-
objective optimization problem during a case study of CNC end milling of UNS C
34000 medium leaded brass. The study aimed at evaluating the best process
environment which could which could simultaneously satisfy multiple requirements
of surface quality. The case indicates the application feasibility of methodology
proposed for multi-response optimization and off-line control of multiple quality
characteristics in CNC end milling.
Kansal et al. (2006) describes an investigation into the optimization of the EDM
process when silicon powder was suspended into the dielectric fluid of EDM.
[23]
Taguchi’s method with multiple performance characteristics has been adopted to
obtain an overall utility value that represents the overall performance of powder
mixed EDM (PMEDM). The obtained experimental results indicate that the peak
current and the concentration of silicon powder suspended into dielectric fluid are
most significant parameters. Moreover the performance of PMEDM has improved
over EDM.
Kaladhar et al. (2011) uses a multi characteristics response optimization model based
on Taguchi and utility concept to optimize process parameters speed, feed, depth of
cut and noise radius on multiple performance characteristics of surface roughness
(Ra) and material removal rate (MRR) during turning of AISI 202 Austenitic stainless
steel using a CVD coated cemented carbide tool. Taguchi’s L8 orthogonal array was
used for experimental planning and ANOVA and F-test were used to analyze the
result.
Badkar et al. (2011) presents the application of Taguchi’s method and 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. In this
study a set of optimal laser hardening parameters such as laser power (LP), scanning
speed (SS) and focused position (FP) were evaluated through the Taguchi’s method of
orthogonal array and utility concept.
[24]
2.3 LITERATURE REVIEW ON APPLICATION OF TAGUCHI’S METHOD
IN SERVICE SECTOR AND OTHER INDUSTRIES
The application of Taguchi’s method of robust design in service industry was first
made by M.C. Holcomb in 1994. Here, Taguchi’s method was applied in the
designing of logistic services with the help of Taguchi’s inner and outer orthogonal
arrays and signal to noise ratios (S/N).
Macfarlane and Eager (1995) successfully applied Taguchi’s robust design of
experiment technique in a case of hospital emergency room. They use Taguchi’s L12
orthogonal array design and reduces the length of stay of patients in a hospital
emergency room.
Raajpoot et al. (2008) makes an important methodological and operational
contributions to the retail service design literature. They uses a Taguchi design
comprising of inner L8(27) and outer 2
2 arrays. Signal-to-noise (S/N) ratio was used to
test design robustness because it considers both mean and variation in measuring
robustness and S/N ratios can also help in selecting the design that simultaneously
maximized the choice probabilities and minimized the performance variations. This
paper also highlights the impact of less studied factors way finding and customer
incompatibility.
Taner and Sezen (2007) show how to successfully apply Taguchi’s method in health
care industry to improve the quality of medical images. This paper shows that the
performance of any imaging equipment can be measured by signal to noise ratio
(S/N). Data were collected from a database of 82 diagnostic thoracic computed
tomography (CT) scans and signal to noise ratios were calculated. With the help of
Taguchi’s method, robust and reliable medical systems were designed to avoid the
observer bias.
Taner and Sezen (2009) propose a new, objective and consistent method for the
calculation of diagnostic efficiency in medical field with the help of hybrid method of
Taguchi and data envelopment analysis (DEA) approach. This proposed method
reflects the diversity of inputs and outputs by incorporating the stepwise application
[25]
of sensitivity, specificity, levelling threshold and efficiency score. A hypothetical case
study was given which involves the eight readers of X-ray films in clinical radiology
for the clear understanding of the method.
Odoobadi (2009) provide a tool for decision makers to consider both tangible and
intangible factors while making decision regarding the investments in advanced
manufacturing technologies (AMT). This study shows the use of Taguchi loss
function to quantify the intangible benefits. The propose procedure helps companies
to rank the technology alternatives and lastly to identify the best technology to be
adopted.
Odoobadi (2009) again provide a tool for decision makers to make more informed
decisions regarding their outsourcing decision regarding the selection of suppliers.
This method uses the Taguchi’s loss function for the inclusion of intangibles in
evaluation and selection of suppliers. To achieve a single measure during the study,
different weighted loss scores are combined to determine single weighted score for
each of the supplier. The supplier who receives the minimum loss score will be finally
selected.
Taner and Antony (2005) establishes the critical score and screening accuracy of the
CAGE questionnaire in the three treatment settings, primary health care, walk-in
clinic and the emergency room. Taguchi’s robust design techniques were applied to
the three screens of the CAGE questionnaire with sensitivity and specificity analysis.
This paper suggests that, to reduce the misclassification rates of alcohol abuse,
screening system should concentrate first on developing ways to standardize
protocols.
Taner and Antony (2006) show how Taguchi method can be applied to the health
care. In this paper, Taguchi loss function integrated with the performance and
parameters of the design of medical applications. This paper shows that when the
patient requirements are consistently met then lower losses can provide an impetus to
improve patient satisfaction. This paper outlines the areas in the health care industry
where Taguchi’s method can be easily applied.
Morsi et al. (2004) applied Taguchi’s method for the simultaneous assessment of the
effects of multiple variables in the tumour microenvironment. The aim of this study
[26]
was to evaluate the ability of Taguchi methods to investigate the effect of several
factors simultaneously on the death or survival of the malignant cells. A major finding
of the study was that the Taguchi method predicted the combination of factors that
results in the lowest survival of malignant cells.
Bayumi et al. (2013) uses L16 orthogonal array for five control parameters which are
font size, font style, line spacing, text/background colour combination and viewing
distance with four levels of each and the other variables like table height, inclination
angle and illumination were fixed. Based on experimental data, analysis of variance
was undertaken which determines the process parameters most contributing to the
optimal level.