International Journal of Advanced Engineering Technology

E-ISSN 0976-3945

Research Paper

TAGUCHI LOSS FUNCTION AS OPTIMISED MODEL FOR

SUPPLIER SELECTION AND EVALUATION
Rajesh Kumar1 Rituraj Chandrakar2 Anil Kumar3 Haldhar ram chandrakar4
1

Address for Correspondence

Associate Professor and Head of Department in Mechatronics Engineering at CSIT, DURG

2, 3, 4
Lecturer in the Department of Mechanical Engineering at CSIT, DURG
ABSTRACT
A survey of the literature reveals the fact that there have been continuous efforts to evaluate suppliers by devising numerous
methodologies. Some of the literature suggests techniques to combine both cardinal and ordinal preferences during the
selection of suppliers. Many of the cases, wherein the fuzzified techniques and ANN methods are used, need to be tested
before applying those in real-world supplier selection decisions. Sometimes, use of a cumbersome technique may frustrate
the DM as well as managers to adopt the methodology wherein the decision does not seek such complex techniques. So,
Taguchi Loss Function has been proposed as a model for supplier selection.
KEY WORDS: ANN methods, Taguchi Loss Function, MCDM.

INTRODUCTION
Supply chain management is a process of planning,
implementing, and controlling the operations of the
supply-chain network catering to the requirements of
customers (purchasers) as efficiently as possible. One
of the primary activities of a value chain model is to
provide service to the customers thereby adding value
to the value-chain network. Further, the goal of any
organization is to maximize the value creation while
minimizing the costs. Thus, selection of a supplier
plays a crucial role in a value chain, or present days
supply-chain network of any organization as it
demands trading off among cardinal and ordinal
preferences of the decision makers (DM) in an
optimal way.
Effective supplier selection calls for robust analytical
methods and decision support tools that are able to
trade off multiple subjective and objective criteria.
Dickson identifies a set of 23 criteria considered by
purchasing managers under different supplier
selection scenario.
A supplier selection decision is inherently a multicriteria problem and a decision of strategic
importance to companies. Supplier selection
decisions within a supply-chain network are
complicated as potential options for such selection
decisions are evaluated on more than one criterion.
Supplier selection is a multi-criteria decision-making
problem which includes both qualitative and
quantitative factors.
LITERATURE REVIEW
Taguchi Methods have been discussed extensively in
different platforms, such as panel discussions, books,
articles, etc., especially since the early 1980s when
applications to different industries began in the
Western Hemisphere.
Taguchis two most important contributions to
quality engineering are the use of Gausss quadratic
loss function to quantify quality and the development
of robust designs (Parameter 12 and Tolerance
design). Taguchis robust designs have widespread
applications upstream in manufacturing to fine tune a
process in such a manner that the output is insensitive
to noise factors. Nearly half of this article deals with
Taguchis parameter and tolerance designs. Several
papers about Taguchi methods originated from the
Centre for Quality and Productivity Improvement at
the University of Wisconsin (CQPI). A number of
reports evaluated Taguchi methods from a statistical
IJAET/Vol.III/ Issue I/January-March, 2012/268-270

standpoint. The primary ones are by Box and Fung

(1986), Box, Bisgaard and Fung (1988), Box and
Jones (1992), Bisgaard (1990), Czitrom (1990),
Bisgaard and Diamond (1990), Bisgaard (1991),
Bisgaard (1992), Bisgaard and Ankenman (1993),
and Steinberg and Burnsztyn (1993). In these reports,
the parameter design received the most attention.
These authors confirm that Dr. Taguchi made
important contributions to quality engineering;
however, it may not be easy to apply his techniques
to real life problems without some statistical
knowledge. Specifically, the use of signal-to-noise
ratios in identifying the nearly best factor levels in
order to minimize quality losses may not be efficient.
Three important discussions on Taguchi methods are
published in Techno metrics by Leon et al. (1987),
Box (1988), and Nair (1992). Some other
performance measures are given and discussed as
alternative to signal-to-noise ratios by Leon et
al.(1987) and Box (1988). Taguchis parameter
design is discussed extensively by a group of
scientists in a discussion panel chaired by Nair
(1992). Their major point is that Taguchi methods do
not have a statistical basis and signal-to-noise ratios
pose some computational problems.
Shoemaker and Tsui (1991) studied Taguchis
parameter design from the standpoint of cost. They
claimed that putting controllable and uncontrollable
factors in two separate arrays, inner and outer, will
result in more experimental runs. Montgomery (1997,
pp. 622-641) highlights the same difficulty in a
Taguchi parameter design. We tend to agree with
these authors that more cost may be involved in a
Taguchi crossed-array design than a combined
single- array classical design as long as the output is
either an STB or LTB type QCH. When the output is
of magnitude type (i.e., QI requires either decreasing
the signal or increasing the signal), we illustrated
above that unless the CV is larger than 17%, the
traditional classical DOE will identify factors that
significantly impact the mean of the output and this
will in turn pave the way to improve quality.
However, when the output is of nominal dimension,
it is best to invest the extra capital to identify the
controls (these are the factors that control process
variation) and signals (these are factors that impact
the mean but have negligible effect on variability)
and go through the Taguchi two-step procedure of
first reducing variability followed by getting the

International Journal of Advanced Engineering Technology

mean on target. Thus our recommendation to any
engineer is to use DOE by all means as an upstream
QI tool. If the engineer does not have sufficient
recourses and the QCH is either STB or LTB and the
CV < 20%, then use a single-array classical FFD
maximizing design resolution. If the QCH is of
nominal-the-best type, then by all means use
Taguchis parameter design even if more
experimentation is required. Further, if the noise
factors are environmental variables, it is generally
best to place such variables in an outer array and treat
them as uncontrollable. Box and Jones (1992) discuss
an alternative to a Taguchi crossed-array design when
the noise factors are environmental. Tsui (1996)
reviews and gives probable problems of Taguchi
methods. According to Kim and Cho (2000), it is
expensive to arrive at a process having on target
mean and minimum variance with Taguchi methods.
They suggest an alternative model based on an
asymmetric quality loss to obtain the most
economical process mean. Robinson et al. (2004) in a
recent article gather previous arguments and
alternative approaches to Taguchi methods.
Alternative performance measures are discussed and
are compared with signal-to noise ratios. Also
Taguchis parameter design is reviewed from
different perspectives. It is nearly impossible to
discuss all the works related to Taguchi methods. We
have tried to mention the main articles that discuss
the pros and cons of Taguchis contributions..
METHODOLOGY
An integrated model of Taguchi loss function is
proposed to solve the supplier selection problems.
The Taguchi loss function is applied to calculate the
loss of each selection criteria.
Supplier evaluation and selection problem has been
studied extensively. In contemporary supply chain
management, the performance of potential suppliers
is evaluated against multiple criteria rather than
considering a single factor-cost. The contemporary
supply management is to maintain long term
partnership with suppliers, and use fewer but reliable
suppliers. Extensive multi-criteria decision making
approach have been earlier proposed for supplier
selection, such as the analytic hierarchy process
(AHP), analytic network process (ANP), case-based
reasoning (CBR), data envelopment analysis (DEA),
fuzzy set theory, genetic algorithm (GA),
mathematical programming, simple multi-attribute
rating technique (SMART), and their hybrids.
In view of the survey of the literature addressing the
supplier selection problem using different
tools/methods, it has been found that there is a
constant need to enhance the effectiveness of the
purchasing decision.
Supplier selection is MCDM problem, because in
Supplier selection problem,
There are M alternatives rated on N
determining conditions called criteria.
The alternatives are denoted as Si ( for i =
1,2,3,M)
The criteria as Cj ( j = 1,2,3,..N)
The main objective of this work is to devise a
methodology for the selection of best supplier which
is able to obtain competitive advantage and achieve
supply chain objectives. To achieve this business
IJAET/Vol.III/ Issue I/January-March, 2012/268-270

E-ISSN 0976-3945

goal, The DMs should apply the best method and

apply accurate criteria to analyze and solve supplier
selection problems.
Based on the literature review (Chin-Nung Liao,
2010), five leading criteria are used for supplier
evaluation and selection, which include product
quality, offering price, delivery lead time, service,
and warranty degree. In addition, the experience and
financial stability of suppliers are considered
important attributes which are included in the study.
1. Product quality
2. Price
3. Delivery time
4. Service satisfaction
5. Warranty degree
Some specific criteria of supplier
Experience time
Financial stability
TAGUCHI LOSS FUNCTIONS
Taguchis loss function is an effective method for
quality engineering. the quality losses occur when the
product deviates beyond the specification limit,
thereby
becoming
unacceptable(Pi&
Low,
2005).Taguchi defines quality as the loss imparted
by any product to society after being shipped to a
customer, other than any loss caused by its intrinsic
function
(Antony
&
Kaye,2000;
Ross,
1996).Taguchi proposed a narrower view of
characteristic acceptability to indicate that any
deviation from a characteristics target value results
in a loss and a higher quality measurement is one that
will result in minimal variation from the target value.
Taguchis loss function is classified into three types
of functions: nominal-is-best characteristics, smalleris-better
characteristics
and
large-is-better
characteristics. The proper function depends on the
magnitude of variation and the variation is allowed in
both directions from the target value. This target can
be the center within two-sided specification limits,
called the two-sided equal or nominal-is-best loss
function (see Fig. 1) and it can be formulated in Eq.
(1).
L(y) =k(y-m) 2
(4.1)
Where L(y) is the loss associated with a particular
value of equality character y; m is the nominal value
of the specification; k is the average loss coefficient,
and its value is a constant depending on the cost at
the specification limits and the width (e.g., m ) of
the specification; where is the customers
tolerance.

Figure 1 Nominal-is-best loss function.

In addition, the other two loss functions include the
one-sided minimum and the one-sided maximum
specification limit functions, called the smaller-isbetter and the larger-is-better loss functions (see Figs.
2 and 3) which are formulated in Eqs. (2) and (3)
respectively.

International Journal of Advanced Engineering Technology

Figure 2 Smaller-is-better loss function

Figure 3 Higher-is-better loss function

L(y) =k. (y) 2 , k= A/2 (4.2)
L(y) =k/ (y) 2, k= A.2. (4.3)
where A is average quality loss; and all other variable
are the same as those in the nominal-is-best loss
function.
SCOPE
Identification of the criteria's for evaluating
suppliers.
Identification of candidate-supplier
Formulation of the model
Solving the formulated model using ILOG
CPLEX
Selection of the best supplier
RESULT
It can be concluded that Taguchi Loss Function plays
an important role in the simplification and the
verification of the supplier selection, which has been
observed in the explanation in the paper.
REFERENCES
1.

Chin-Nung Liao, (2010). Supplier selection project

using an integrated Delphi, AHP and Taguchi loss
function, Prob Stat Forum (03) 118-134.
2. Dickson, G. W. (1966). An analysis of supplier
selection system and decision. Journal of Purchasing,
2(1), 517.
3. Kethley, R. B., & Waller, T. A. (2002). Improving
customer service in the real estate industry: A property
selection model using Taguchi loss functions. Total
Quality Management, 13(6), 739748.
4. Pi, W. N., & Low, C. (2005). Supplier evaluation and
selection using Taguchi loss functions. The
International Journal of Advanced Manufacturing
Technology, 26,155160.
5. Talluri, S., Baker, R.C., (2002). A multi-phase
mathematical programming approach for effective
supply chain design. European Journal of Operational
Research 141 (3),544558.
6. Ellram, L. (1990). The supplier selection decision in
strategic partnerships. Journal of Purchasing and
Materials Management, 26(1), 814.
7. Li, H. C. (2003). Quality loss functions for the
measurement of service quality. The International
Journal of Advanced Manufacturing Technology, 21,
2937.
8. Liao, C. N. (2009). Formulating the multi-segment goal
programming. Computers &Industrial Engineering, 56,
138141.
9. Ross, P. J. (1996). Taguchi techniques for quality
engineering. New York: McGraw-Hill.
10. Saaty, T. L. (1980). The analytic hierarchy process.
New York: McGraw-Hill.

IJAET/Vol.III/ Issue I/January-March, 2012/268-270

E-ISSN 0976-3945

11. Thompson, K. N. (1990). Vendor profile analysis.

Journal of Purchasing and Materials Management,
26(4), 1118.
12. Willis, T. H., & Houston, C. R. (1990). Vendor
requirements and evaluation in a just-in-time
environment. International Journal of Operations and
Production Management, 10(4), 4150.
13. Wu, C. C., Lin, P. C., & Chou, C. Y. (2006).
Determination of price and warranty length for a
normal lifetime distributed product. International
Journal of Production Economics, 102, 95107.
14. Yoon, K., & Hwang, C. L. (1995). Multiple attribute
decision making: An introduction. Thousand Oaks, CA:
Sage.
15. Talluri, S., Narasimhan, R., (2003). Vendor evaluation
with performance variability: A maxmin approach.
European Journal of Operational Research 146 (3),
543552.
16. Wang, G., Huang, S.H., Dismukes, J.P., (2004).
Product-driven supply chain selection using integrated
multi-criteria
decision-making
methodology.
International Journal of Production Economics 91 (1),
115.
17. Hong, G.H., Park, S.C., Jang, D.S., Rho, H.M., (2005).
An effective supplier selection method for constructing
a competitive supply-relationship. Expert Systems with
Applications 28 (4), 629639.
18. Liu, F.H.F., Hai, H.L., (2005). The voting analytic
hierarchy process method for selecting supplier.
International Journal of Production Economics 97 (3),
308 317.
19. Chen, C.T., Lin, C.T., Huang, S.F., (2006). A fuzzy
approach for supplier evaluation and selection in supply
chain management. International Journal of Production
Economics 102 (2), 289301.
20. Huan-JyhShyur, Hsu-Shih Shih, (2006). A hybrid
MCDM model for strategic vendor selection,
Mathematical and Computer Modelling (44) 749761
21. Wei-Ning Pi Chinyao Low, (2006).
Supplier
evaluation and selection via Taguchi loss functions and
an AHP, Int J Adv Manuf Technol (27) 625630.
22. Talluri, S., Narasimhan, R., Nair, A., 2006. Vendor
performance with supply risk: A chance-constrained
DEA approach. International Journal of Production
Economics 100 (2), 212222.
23. Chan, F.T.S., Kumar, N., (2007). Global supplier
development considering risk factors using fuzzy
extended
AHP-based
approach.
OMEGA

International Journal of Management Science 35 (4),

417431.
24. Florez-Lopez, R., (2007). Strategic supplier selection in
the added-value perspective: A CI approach.
Information Sciences 177 (5), 11691179.
25. Liao, Z., Rittscher, J., (2007). A multi-objective
supplier selection model under stochastic demand
conditions. International Journal of Production
Economics 105 (1), 150159.
26. Chou, S.Y., Chang, Y.H., (2008). A decision support
system for supplier selection based on a strategyaligned fuzzy SMART approach. Expert Systems with
Applications 34 (4), 22412253.
27. Ho, W., (2008). Integrated analytic hierarchy process
and its applications A literature review. European
Journal of Operational Research 186 (1), 211228.
28. Kull, T.J., Talluri, S., (2008). A supply-risk reduction
model using integrated multi-criteria decision making.
IEEE Transactions on Engineering Management 55 (3),
409419.