Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

Computers & Industrial Engineering 60 (2011) 66–76

Contents lists available at ScienceDirect

Computers & Industrial Engineering


journal homepage: www.elsevier.com/locate/caie

Decision-making with distance measures and induced aggregation operators q


José M. Merigó ⇑, Montserrat Casanovas
Department of Business Administration, University of Barcelona, Av. Diagonal 690, 08034 Barcelona, Spain

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, we present a new decision-making approach that uses distance measures and induced
Received 17 December 2009 aggregation operators. We introduce the induced ordered weighted averaging distance (IOWAD) oper-
Received in revised form 27 July 2010 ator. IOWAD is a new aggregation operator that extends the OWA operator by using distance measures
Accepted 28 September 2010
and a reordering of arguments that depends on order-inducing variables. The main advantage of IOWAD
is that it provides a parameterized family of distance aggregation operators between the maximum and
the minimum distance based on a complex reordering process that reflects the complex attitudinal
Keywords:
character of the decision-maker. We studied some of IOWAD’s main properties and different particular
Decision-making
OWA operator
cases and further generalized IOWAD by using Choquet integrals. We developed an application in a
Distance measures multi-person decision-making problem regarding the selection of investments. We found that the main
Induced aggregation operators advantage of this approach is that it is able to provide a more complete picture of the decision-making
process, enabling the decision-maker to select the alternative that it is more in accordance with his
interests.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction maximum, the minimum and the average, among others (Ahn,
2009; Beliakov, Pradera, & Calvo, 2007; Chiclana, Herrera-Viedma,
There is a wide range of methods for decision-making in the lit- Herrera, & Alonso, 2007; Emrouznejad, 2008; Liu, 2008; Wang, Luo,
erature (Figueira, Greco, & Ehrgott, 2005; Gil-Aluja, 1999; Merigó, & Liu, 2007; Xu, 2005, 2009b; Yager, 1993, 2007, 2009a, 2009b,
2008, 2010; Merigó & Casanovas, 2009; Merigó, López-Jurado, 2009c). The use of the OWA operator in different types of distance
Gracia, & Casanovas, 2009; Wei, 2009; Xu, 2009a, 2009b, 2009c; measures have been studied by several authors (Karayiannis, 2000;
Yager, 2009a; Zarghami & Szidarovszky, 2009). A very useful tech- Merigó, 2008; Merigó & Gil-Lafuente, 2007, 2008a, 2008b, 2009b,
nique for decision-making is the Hamming distance (Hamming, 2010). Numerous authors have studied other developments con-
1950) and, more generally, all the distance measures (Karayiannis, cerning the OWA operator, refer, e.g., to Amin and Emrouznejad
2000; Kaufmann, 1975; Merigó, 2008; Merigó & Gil-Lafuente, (2006), Beliakov et al. (2007), Cheng, Wang, and Wu (2009),
2007, 2008a, 2008b, 2009b, 2010; Szmidt & Kacprzyk, 2000). The Kacprzyk and Zadrozny (2009), Liu, Cheng, Chen, and Chen
main advantage of using distance measures in decision-making is (2010), Merigó (2008, 2010), Merigó and Casanovas (2008, 2010a,
that we can compare the alternatives of the problem with some 2010b), Merigó, Casanovas, and Martı´nez (2010), Merigó and
ideal result (Gil-Aluja, 1999). Through this comparison, the alter- Gil-Lafuente (2008c, 2009a) and Yager and Kacprzyk (1997).
native with the closest result to the ideal is the optimal choice. An interesting extension of the OWA operator is the induced
Usually, when using distance measures in decision-making, we OWA (IOWA) operator (Yager & Filev, 1999). IOWA differs in that
normalize them by using the arithmetic mean or the weighted the reordering step is not developed with the values of the argu-
average (WA) obtaining the normalized Hamming distance ments but can be induced by another mechanism such that the
(NHD) and the weighted Hamming distance (WHD), respectively. ordered position of the arguments depends upon the values of
However, it is sometimes of interest to consider the possibility of their associated order-inducing variables. The IOWA operator has
parameterizing the results from the maximum distance to the received increasing attention in recent years (Chiclana et al.,
minimum distance. In this case, the ordered weighted averaging 2007; Herrera-Viedma, Chiclana, Herrera, & Alonso, 2007; Merigó,
(OWA) operator can be used (Yager, 1988). The OWA operator is 2008; Merigó & Casanovas, 2009; Merigó & Gil-Lafuente, 2009a;
a useful technique for aggregating information, providing a Wei, Zhao, & Lin, 2010; Wu, Li, Li, & Duan, 2009; Yager, 2003).
parameterized family of aggregation operators that includes the The aim of this paper is to present the use of the induced OWA
(IOWA) operator in decision-making with distance measures. We
q
This manuscript was processed by Area Editor Imed Kacem.
formulate a more general model by using order-inducing variables
⇑ Corresponding author. Tel.: +34 93 402 19 62; fax: +34 93 403 98 82. in the reordering process of the OWA aggregation. We thus
E-mail addresses: jmerigo@ub.edu (J.M. Merigó), mcasanovas@ub.edu (M. Casanovas). introduce a new aggregation operator: the induced ordered

0360-8352/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.cie.2010.09.017
J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76 67

weighted averaging distance (IOWAD) operator. The IOWAD oper- This paper is organized as follows. In Section 2, we briefly
ator is an aggregation operator that provides a parameterized fam- review basic concepts that are used throughout the paper. In Sec-
ily of distance aggregation operators that ranges from the tion 3, we present the IOWAD operator. Section 4 analyzes differ-
minimum to the maximum distance. The main advantage of the ent families of IOWAD operators. In Section 5, we develop an
IOWAD operator is that it is able to deal with complex attitudinal extension of IOWAD by using Choquet integrals. In Section 6, we
characters (or complex degrees of orness) in the decision process present a method for multi-person decision-making with the
by using order-inducing variables. In so doing, we are able to deal IOWAD operator in investment decisions and Section 7 develops
with more complex problems that are closer to real-world a numerical example of the new approach. Finally, we summarize
situations. the main conclusions of the paper in Section 8.
In order to see the usefulness of the IOWAD operator, let us
look into a real-world example in business decision-making. An 2. Preliminaries
important business decision, for example, is usually made by
the company’s board of directors. The decision involves the atti- In this section, we briefly describe the Hamming distance, the
tudinal character of a group of persons that must be coordinated OWA operator and the induced OWA operator.
into one simple decision according to the group’s interests. Obvi-
ously, the attitudinal character of this example is much more 2.1. The Hamming distance
complex than simply using the degree of optimism (degree of
orness) of the company. Note that in this example, we analyze The Hamming distance (Hamming, 1950) is a useful technique
the attitudinal character (degree of orness) in group decision- for calculating the differences between two parameters, such as
making problems, but the actual analysis would be much more problems with two elements or two sets. The Hamming distance
complex. Thus, a good method for analyzing this problem would can be useful in fuzzy set theory, for example, when calculating
be the use of order-inducing variables by using the IOWAD distances between fuzzy sets, interval-valued fuzzy sets and intui-
operator. tionistic fuzzy sets. To define the Hamming distance, we first
We study basic properties of the IOWAD operator and we define a distance measure. A distance measure must basically
consider a wide range of particular cases: the NHD; the WHD; accomplish the following properties:
the ordered weighted averaging distance (OWAD) operator; the
median-IOWAD; the olympic-IOWAD and the centered-IOWAD.  Non-negativity: D(A1, A2) P 0.
We see that each particular case is useful for a certain situation  Commutativity: D(A1, A2) = D(A2, A1).
according to the objectives of the decision-maker. Depending on  Reflexivity: D(A1, A1) = 0.
the particular type of operator used, the results may differ. Note  Triangle inequality: D(A1, A2) + D(A2, A3) P D(A1, A3).
that it is possible to generalize the aggregation operator by using
generalized and quasi-arithmetic means following the ideas of For two sets, A = (a1, . . . , an) and B = (b1, . . . , bn), we can define the
Merigó and Gil-Lafuente (2009a, 2009b). We also present a more Hamming distance as follows:
general formulation by using mixture and infinitary operators such
as the induced mixture distance (IMD) operator, the induced quasi- Definition 1. A normalized Hamming distance of dimension n is a
arithmetic mixture distance (Quasi-IMD) operator and the infini- mapping NHD: [0, 1]n  [0, 1]n ? [0, 1], such that:
tary IOWAD (1-IOWAD) operator. The main advantage of these !
generalizations is that they are able to provide a deeper represen- 1X n
NHDðA; BÞ ¼ jai  bi j ; ð1Þ
tation of the specific problem considered that includes the IOWAD n i¼1
operator as a particular case.
We also present an application of the new approach in a mul- where ai and bi are the ith arguments of the sets A and B
ti-person decision-making problem concerning the selection of respectively.
investments. The main advantage of this model is that it gives a Sometimes, when normalizing the Hamming distance, we
more complete view of the decision problem because it considers prefer to give different weights to each individual distance. In this
a wide range of distance aggregation operators according to the case, the distance is known as the weighted Hamming distance,
interests of the decision-maker. Moreover, by using several which can be defined as follows:
experts in the analysis, we obtain information that it is more ro-
bust because the opinion of several experts is always better than
Definition 2. A weighted Hamming distance of dimension n is a
the opinion of one. For doing so, we introduce a new aggregation
mapping WHD: [0, 1]n  [0, 1]n ? [0, 1] that has an associated
operator called the multi-person–IOWAD (MP–IOWAD) operator. P
weighting vector W of dimension n with W ¼ nj¼1 wj ¼ 1 and
We then study some of its main particular cases, such as the
wj 2 [0, 1], such that:
multi-person–OWAD (MP–OWAD) and the multi-person–WHD !
(MP–WHD). Note also that the IOWAD and the MP–IOWAD oper- X
n

ator are applicable to a wide range of situations such as fuzzy set WHDðA; BÞ ¼ wi jai  bi j ; ð2Þ
i¼1
theory, operational research, statistics, economics and engineer-
ing. This method is also applicable to different decision-making where ai and bi are the ith arguments of the sets A and B
problems such as in strategic decision-making, human resource respectively.
management, product management and financial management. Note that it is possible to generalize this definition to all real
The main advantage of using distance measures is that we can numbers by using Rn  Rn ? R. For the formulation used in fuzzy
compare the real-world information with ideal information set theory, see, for example, Gil-Aluja (1998), Kaufmann (1975),
and see which alternative better fits with the interests of the Merigó (2008) and Szmidt and Kacprzyk (2000).
decision-maker. For example, in human resource selection, we
can establish an ideal candidate that would perfectly fit the 2.2. The OWA operator
company and compare it with the real-world alternatives that
we have in the market and select the candidate with closest The OWA operator (Yager, 1988) provides a parameterized
results to the ideal one. family of aggregation operators that include the maximum, the
68 J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76

minimum and the average criteria as special cases. This operator order-inducing variables. For two sets X = {x1, x2, . . . , xn} and
can be defined as follows: Y = {y1, y2, . . . , yn}, IOWAD can be defined as follows:

Definition 3. An OWA operator of dimension n is a mapping OWA: Definition 5. An IOWAD operator of dimension n is a mapping
Rn ? R that has an associated weighting vector W of dimension n IOWAD: Rn  Rn  Rn ? R that has an associated weighting vector
P P
such that W ¼ nj¼1 wj ¼ 1 and wj 2 [0, 1], according to the follow- W such that wj 2 [0, 1] and W ¼ nj¼1 wj ¼ 1, according to the
ing formula: following formula:
X
n X
n
OWAða1 ; a2 ; . . . ; an Þ ¼ wj bj ; ð3Þ IOWADðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj ; ð6Þ
j¼1 j¼1

where bj is the jth largest ai. where bj is the jxi  yij value of the IOWAD triplet hui, xi, yii having
From a generalized perspective of the reordering step, it is the jth largest ui, ui is the order-inducing variable and jxi  yij is the
possible to distinguish between the descending OWA (DOWA) argument variable represented in the form of individual distances.
operator and the ascending OWA (AOWA) operator. The OWA In the following example, we present a simple numerical
operator is commutative, monotonic, bounded and idempotent. example showing how to use the IOWAD operator in an aggrega-
The OWA operator aggregates the information according to the tion process:
attitudinal character (or degree of orness) of the decision-maker
(Yager, 1988). The attitudinal character is represented according to
the following formula: Example 1. Assume the following arguments in an aggregation
  process: X = (7, 30, 10, 15), Y = (4, 20, 6, 9) with the following order
X
n
nj inducing variables U = (2, 6, 9, 7). Assume the following weighting
aðWÞ ¼ wj : ð4Þ
j¼1
n1 vector W = (0.2, 0.2, 0.3, 0.3). If we calculate the distance between
X and Y using the IOWAD operator, we get the following:
Note that a(W) 2 [0, 1]. The more weight W is located close to
IOWADðX; YÞ ¼ 0:2  j10  6j þ 0:2  j15  9j þ 0:3  j30  20j þ 0:3  j7  4j
the top, the closer a is to 1. In decision-making problems, the
¼ 5:9:
degree of orness is useful for representing the attitudinal character
of the decision-maker by using it as the degree of optimism or A fundamental aspect of the IOWAD operator is the reordering
pessimism. of the arguments based upon order-inducing variables. That is, the
Different families of OWA operators are found by using weights, rather than being associated with a specific argument as
different manifestations in the weighting vector, such as maxi- in the case with the usual Hamming distance, are associated with
mum, minimum and average criteria. For more information on the position given by the order-inducing variables. This reordering
other families, refer, for example, to Ahn (2009), Beliakov et al. introduces nonlinearity into an otherwise linear process.
(2007), Emrouznejad (2008), Liu (2008, 2010), Merigó (2008); Xu If D is a vector corresponding to ordered arguments bj, we call
(2005) and Yager (1993, 2007, 2009a, 2009b). this the ordered argument vector. If WT is the transpose of the
weighting vector, then the IOWAD operator can be presented as
2.3. The induced OWA operator follows:

IOWADðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ ¼ W T D: ð7Þ


The IOWA operator (Yager & Filev, 1999) is an extension of the
OWA operator. The main difference is that the reordering step is From a generalized perspective of the reordering step, it is possible
not carried out with the values of the arguments ai. In this case, to distinguish between descending (DIOWAD) and ascending
the reordering step is developed with order-inducing variables that (AIOWAD) orders. The weights of these operators are related by
reflect a more complex reordering process. The IOWA operator also wj ¼ wnjþ1 , where wj is the jth weight of the DIOWAD operator
includes as particular cases maximum, minimum and average cri- and wnjþ1 the jth weight of the AIOWAD operator.
teria. The IOWA operator can be defined as follows: Note that if the weighting vector is not normalized (Beliakov
P
et al., 2007), i.e., W ¼ nj¼1 wj – 1, then the IOWAD operator can
Definition 4. An IOWA operator of dimension n is a mapping be expressed as:
IOWA: Rn  Rn ? R that has an associated weighting vector W of 1 Xn
P
dimension n with W ¼ nj¼1 wj ¼ 1 and wj 2 [0, 1], such that: IOWADðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj : ð8Þ
W j¼1
X
n
IOWAðhu1 ; a1 i; hu2 ; a2 i; . . . ; hun ; an iÞ ¼ wj bj ; ð5Þ Note that IOWAD(hu1, x1, y1i, hu2, x2, y2i, . . . , hun, xn, yni) = 0 if and
j¼1 only if xi = yi for all i 2 [1, n]. Note also that IOWAD(hu1, x1,
where bj is the ai value of the IOWA pair hui, aii having the jth largest y1i, hu2, x2, y2i, . . . , hun, xn, yni) = IOWAD(hu1, y1, x1i, hu2, y2, x2i, . . . ,
ui, ui is the order-inducing variable and ai is the argument variable. hun, yn, xni).
Note that it is possible to distinguish between the descending An interesting issue arises when analyzing similarity measures,
IOWA (DIOWA) operator and the ascending IOWA (AIOWA) namely, the possibility of considering the dissimilarity measure.
operator (Merigó & Gil-Lafuente, 2009a). The IOWA operator is For the IOWAD operator, assuming that we are in the unit interval
also monotonic, bounded, idempotent and commutative (Yager & [0, 1], the dissimilarity measure is given by dissimilarity-
Filev, 1999). IOWAD = 1  IOWAD. As we can see, the dissimilarity is the dual
of the IOWAD operator. Note that the dissimilarity can be studied
in a similar way for all of the extensions mentioned in the rest of
3. The induced ordered weighted averaging distance operator this paper.
Other interesting generalizations can be developed following:
The IOWAD operator is a distance measure that uses the IOWA Mesiar and Pap (2008), Spirkova (2008) and Torra and Narukawa
operator in the normalization process of the Hamming distance. (2010). Following Spirkova (2008), we can develop the function in-
Thus, the reordering of the individual distances is developed with duced OWAD operator, which uses a generating function r for the
J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76 69

P1
order-inducing variables such that r: I ? R, being that I  R is a appear in the aggregation process. Note that j¼1 wj ¼ 1. By using,
closed interval I = [a, b]. In this paper, we will use a more general the IOWAD operator we get the infinitary IOWAD (1-IOWAD)
representation by also using also a generating function for the operator as follows:
arguments such that s: Rm ? R. This generating function expresses X
1
the formation of the arguments when a previous analysis exists, 1-IOWADðhro ðu1 Þ; sp ðx1 Þ; sq ðy1 Þi; . . . ; hro ðun Þ; sp ðxn Þ; sq ðyn ÞiÞ ¼ wj bj ;
such as the use of a multi-person process where each argument j¼1
is constituted by the opinion of m persons. Moreover, we will use ð11Þ
a weighting function f for the weighting vector. Note that the use
of a weighting function fi in the weighting vector of the weighted However, note that the reordering process is much more complex,
average is known as the Losonczi mean (Losonczi, 1971). If the that is, we never know which argument is the largest argument be-
function is equal for all weights f, then we get the simple Losonczi cause we have an unlimited number of arguments. This problem
mean or the quasi-mixture operator (Spirkova, 2008). In this case, can be partially solved by using the order-inducing variables. For
we directly extend the approach by obtaining the function induced further reading about the usual OWA, see Mesiar and Pap (2008).
mixture distance (IMD) operator as follows (note that we could Note that a similar extension could be developed by using the
also refer to it as the Losonczi-IOWAD (Lo-IOWAD) operator). In IMD operator, thus obtaining the 1-IMD operator, and by using
this definition, we refer to the arguments as two sets X = {x1, generalized and quasi-arithmetic means, thus obtaining the
x2, . . . , xn} and Y = {y1, y2, . . . , yn}. 1-Quasi-IMD operator.
The IOWAD operator is commutative, monotonic, bounded,
Definition 6. An IMD operator of dimension n is a mapping IMD: idempotent, nonnegative and reflexive but it does not accomplish
[Rn  Rn  Rn ? R that has an associated vector of weighting always the triangle inequality. These properties can be proved with
functions f, r: I ? ]0, 1[, is a some positive continuous function, the following theorems:
s: Rm ? R, such that:
Theorem 1 (Commutativity – OWA aggregation). Assume f is the
IMDðhr o ðu1 Þ; sp ðx1 Þ; sq ðy1 Þi; . . . ; hr o ðun Þ; sp ðxn Þ; sq ðyn ÞiÞ IOWAD operator, then
Pn
j¼1 fj ðsy ðbj ÞÞsy ðbj Þ f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ f ðhu1 ; c1 ; d1 i; . . . ; hun ; cn ; dn iÞ; ð12Þ
¼ Pn ; ð9Þ
j¼1 fj ðsy ðbj ÞÞ
where (hu1, x1, y1i, . . ., hun, xn, yni) is any permutation of the arguments
where sy(bj) is the jsp(xi)  sq(yi)j value of the IMD triplet (hu1, c1, d1i, . . ., hun, cn, dni).
hro(ui), sp(xi), sq(yi)i having the jth largest ro(ui); ui is the order-
inducing variable; jsp(xi)  sq(yi)j is the argument variable Theorem 2 (Commutativity – distance measure). Assume f is the
represented in the form of individual distances; and o, p and q indi- IOWAD operator, then
cate that each order-inducing variable and each argument is formed
by using a different function. f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ f ðhu1 ; y1 ; x1 i; . . . ; hun ; yn ; xn iÞ: ð13Þ
Note that the IMD operator can be further generalized by using
generalized and quasi-arithmetic means (Merigó & Gil-Lafuente, Theorem 3 (Monotonicity). Assume f is the IOWAD operator;
2009a). The result is the induced generalized mixture distance if jxi  yij P jci  dij, for all ii, then
(IGMD) operator and the induced quasi-arithmetic mixture dis-
tance (Quasi-IMD) operator. The Quasi-IMD operator can be f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ P f ðhu1 ; c1 ; d1 i; . . . ; hun ; cn ; dn iÞ: ð14Þ
defined as follows:

Theorem 4 (Bounded). Assume f is the IOWAD operator, then


Definition 7. A Quasi-IMD operator of dimension n is a mapping
QIMD: Rn  Rn  Rn ? R that has an associated a vector of weight- minfjxi  yi jg 6 f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ
ing functions f, r: I ? ]0, 1[, is some positive continuous function,
s: Rm ? R, such that: 6 maxfjxi  yi jg: ð15Þ

QIMDðhr o ðu1 Þ; sp ðx1 Þ; sq ðy1 Þi; . . . ; hro ðun Þ; sp ðxn Þ; sq ðyn ÞiÞ ¼
Pn ! Theorem 5 (Idempotency). Assume f is the IOWAD operator;
j¼1 fj ðsy ðbj ÞÞgðsy ðbj ÞÞ if jxi  yij = a, for all i, then
¼ g 1 Pn ; ð10Þ
j¼1 fj ðsy ðbj ÞÞ
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ a: ð16Þ
where g is a strictly continuous monotonic function; sy(bj) is the
jsp(xi)  sq(yi)j value of the IMD triplet (hro(ui), sp(xi), sq(yi)i) having
the jth largest ro(ui); ui is the order-inducing variable; jsp(xi)  sq(yi)j Theorem 6 (Non-negativity). Assume f is the IOWAD operator, then
is the argument variable represented in the form of individual dis-
tances; and o, p and q indicates that each order-inducing variable f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ P 0: ð17Þ
and each argument is formed by using a different function.
Note that if g(b) = bk, we get the IGMD operator. Following Theorem 7 (Reflexivity). Assume f is the IOWAD operator, then
Merigó and Gil-Lafuente (2009a), we can obtain a wide range of
particular cases of the Quasi-IMD operator. For example, if g(b) = b, f ðhu1 ; x1 ; x1 i; . . . ; hun ; xn ; xn iÞ ¼ 0: ð18Þ
we obtain the IMD operator. If g(b) = b2, the induced quadratic
mixture distance (IQMD) operator. If g(b) ? b0, the induced Note that the IOWAD operator does not always accomplish the
geometric mixture distance operator (IGMD) and if g(b) = b1, the triangle inequality because we may find some special situations
induced harmonic mixture distance (IHMD) operator. where f (hu1, x1, y1i, . . . , hun, xn, yni) + f (hu1, y1, z1i, . . . , hun, yn, zni) < f
Another interesting extension uses infinitary aggregation oper- (hu1, x1, z1i, . . . , hun, xn, zni). In the following, we present a numerical
ators (Mesiar & Pap, 2008). Here, we can represent an aggregation example where we prove that the IOWAD operator does not
process where there are an unlimited number of arguments that always accomplish the triangle inequality.
70 J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76

Example 2. Assume X = (8, 10, 20), Y = (6, 6, 10) and Z = (4, 2, 10). case, it would mean replacing the tied arguments by their normal-
We assume the following weighting vector W = (0, 1, 0). We use the ized Hamming distance.
following order-inducing variables: U (X, Y) = (3, 7, 9), U(Y, Z) = In the analysis of the order-inducing variables of the IOWAD
(4, 6, 2) and U(X, Z) = (4, 6, 8). operator, we should note that the values used can be drawn from
any space, with having a linear ordering the only requirement.
IOWADðX; YÞ ¼ 0  j20  10j þ 1  j10  6j þ 0  j8  6j ¼ 4; Therefore, it is possible to use different kinds of attributes for the
IOWADðY; ZÞ ¼ 0  j6  2j þ 1  j6  4j þ 0  j10  10j ¼ 2; order-inducing variables that permit us to, for example, mix num-
IOWADðX; ZÞ ¼ 0  j20  10j þ 1  j10  2j þ 0  j8  4j ¼ 8: bers with words in the aggregations. Note also that in some situa-
tions it is possible to use the implicit lexicographic ordering
As we can see, 4 + 2 < 8; and therefore, IOWAD(X, Y) + IOWAD associated with words such as the ordering of words in dictionaries
(Y,Z) < IOWAD (X, Z). We have thus proved that the IOWAD operator (Yager & Filev, 1999).
does not always accomplish the triangle inequality.
Note that the order-inducing variables that we have used include 4. Families of IOWAD operators
the OWAD operator (Merigó & Gil-Lafuente, 2007, 2010) as a
particular case. Therefore, we have also proved with this example By using a different manifestation of the weighting vector, we
that the OWAD operator does not accomplish the triangle inequality. are able to obtain different types of IOWAD operators, such as
It is also possible to develop a similar formulation of the IOWAD the normalized Hamming distance (NHD), the weighted Hamming
operator by first reordering the arguments and then calculating the distance (WHD), the ordered weighted averaging distance (OWAD)
distances. However, this formulation does not accomplish the operator, the step-IOWAD, the window-IOWAD, the median-
usual properties of distance measures, such as commutativity. This IOWAD, the olympic-IOWAD and the centered-IOWAD.
measure can be formulated in the following way:
X
n Remark 1. For example, the maximum distance, the minimum
gðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ ¼ wk ck ; ð19Þ distance, the step-IOWAD, the NHD, the WHD and the OWAD are
k¼1 obtained as follows:
where ck is the jxk  ykj value of the triplet hui x, i, yii; xk and yk are  The maximum distance is found if wp = 1 and wj = 0, for all j – p,
the ith argument variables of the sets X = {x1, . . . , xn} and Y = and up = max{ai}.
{y1, . . . , yn}, respectively, having the kth largest ui; and ui is the  The minimum distance if wp = 1 and wj = 0, for all j – p, and
order-inducing variable. up = min{ai}.
Note that if the reordering k is equal to the reordering j, this  More generally, if wk = 1 and wj = 0 for all j – k, we get the step-
measure becomes the IOWAD operator described in Eq. (6), which IOWAD operator.
can be proven as follows:  The NHD is formed when wj = 1/n for all i.
 The WHD is obtained when the ordered position of ui is the
same as ai.
Theorem 8. Assume f is the IOWAD operator and g is the measure  The OWAD is found if the ordered position of ui is the same as
explained in Eq. (19). If k = j, then the ordered position of the values of jxi  yij, for all i.
gðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ
¼ f ðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ: ð20Þ Remark 2. Another particular case is the olympic-IOWAD opera-
tor. This operator is obtained when w1 = wn = 0, and for all others,
Another issue to consider is the different measures used in the
wj ¼ 1=ðn  2Þ. Note that if n = 3 or n = 4, the olympic-IOWAD
OWA literature for characterizing the weighting vector. For exam-
becomes the median-IOWAD.
ple, we could consider the entropy of dispersion, the balance oper-
ator, the divergence of W and the degree of orness (Merigó, 2008;
Yager, 1988). The entropy of dispersion is defined as follows: Remark 3. Note that it is possible to present a general form of the
X
n olympic-IOWAD operator, considering that wj = 0 for j = 1, 2, . . . ,
HðWÞ ¼  wj lnðwj Þ: ð21Þ k, n, n  1, . . . , n  k + 1; and for all others, wj ¼ 1=ðn  2kÞ, where
j¼1 k < n/2. Note that if k = 1, then this general form becomes the usual
olympic-IOWAD. If k = (n  1)/2, then it becomes the median-
For the balance operator, we get
IOWAD operator.
Xn  
n þ 1  2j
BALðWÞ ¼ wj : ð22Þ
n1
j¼1 Remark 4. Additionally, it is also possible to present the contrary
case of the general olympic-IOWAD operator. In this case, wj = (1/
For the divergence of W, we get
2k) for j = 1, 2, . . . , k, n, n  1, . . . , n  k + 1; and wj = 0, for all others,
X
n  2
nj where k < n/2. Note that if k = 1, then we get the contrary case of
DIVðWÞ ¼ wj  aðWÞ : ð23Þ
n1 the median-IOWAD.
j¼1

The degree of orness can be defined in the following way:


Remark 5. Another interesting family is the S-IOWAD operator
X
n  
nj based on the S-OWA operator (Yager, 1993). These operators can
aðWÞ ¼ wj : ð24Þ
n1 be subdivided in three classes: the ‘‘orlike”, the ‘‘andlike” and the
j¼1
generalized S-IOWAD operators. The generalized S-IOWAD opera-
where wj is the wj weight of the IOWAD aggregation ordered tor is obtained when wp = (1/n)(1  (a + b)) + a and up = max{ai},
according to the values of arguments jxi  yij. wq = (1/n)(1  (a + b)) + b and uq = min{ai} and wj = (1/n)(1 
A further issue is the problem of ties in the reordering process of (a + b)) for j – p, q; where a, b 2 [0, 1] and a + b 6 1. Note that if
the order-inducing variables. To solve this problem, we recom- a = 0, the generalized S-IOWAD operator becomes the ‘‘andlike”
mend the policy explained by Yager and Filev (1999), namely, S-IOWAD, and if b = 0, it becomes the ‘‘orlike” S-IOWAD. Note also
replacing the tied arguments by their average. Note that in this that if a + b = 1, we get the induced Hurwicz distance criteria.
J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76 71

Remark 6. A further family that could be used is the centered- X


n
ICDIAðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ ¼ bj ½mðAðiÞ Þ  mðAði1Þ Þ;
IOWAD operator, which is based on Yager (2007). An IOWAD oper- j¼1
ator can be defined as a centered aggregation operator if it is
ð26Þ
symmetric, strongly decaying and inclusive.
where bj is the jxi  yij value of the ICDIA triplet hui, xi, yii having the
 It is symmetric if wj = wj+n1. jth largest ui; ui is the order-inducing variable; the jxi  yij is the
 It is strongly decaying when: i < j 6 (n + 1)/2, then wi < wj, argument variable represented in the form of individual distances;
and when i > j P (n + 1)/2, then wi < wj. A(i) = {x(1), . . . ,x(i)} i P 1; and A(0) = ;.
 It is inclusive if wj > 0. This approach can be generalized by using both generalized and
quasi-arithmetic means (Merigó & Gil-Lafuente, 2009a). For exam-
Note that it is possible to consider a softening of the second ple, by using quasi-arithmetic means, we get the induced quasi-
condition by using wi 6 wj instead of wi < wj (softly decaying arithmetic Choquet distance integral aggregation (Quasi-ICDIA)
centered-IOWAD operator). Another particular situation of the operator, which can be defined as follows:
centered-IOWAD appears if we remove the third condition (non-
inclusive centered-IOWAD).
Definition 11. Let m be a fuzzy measure on X. An induced quasi-
arithmetic Choquet distance integral aggregation (Quasi-ICDIA)
Remark 7. Using a similar methodology, we could develop numer- operator of dimension n is a function QICDIA: Rn  Rn  Rn ? R,
ous other families of IOWAD operators. For more information, refer such that:
to Ahn (2009), Beliakov et al. (2007), Chiclana et al. (2007), !
Emrouznejad (2008), Liu (2008), Merigó (2008), Merigó and X
n
1
QICDIAðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ g gðbj Þ½mðAðiÞ Þ  mðAði1Þ Þ ;
Gil-Lafuente (2008a, 2008b, 2009a), Xu (2005), Yager (1993, j¼1
2009a, 2009b, 2009c).
ð27Þ
5. Choquet integrals with distance measures and induced
aggregation operators where g is a strictly continuous monotonic function; bj is the jxi  yij
value of the ICDIA triplet hui, xi, yii having the jth largest ui, ui is the
Following Bolton, Gader, and Wilson (2008), Choquet (1953), order-inducing variable; jxi  yij is the argument variable repre-
Mesiar (1995), Tan and Chen (2010), and Yager (2004), it is possi- sented in the form of individual distances; A(i) = {x(1), . . . , x(i)} i P 1;
ble to develop an extension of the IOWAD operator by using the and A(0) = ;.
discrete Choquet integral. This method results in the induced Cho- Note that this new approach can also be formulated in a more
quet distance integral aggregation (ICDIA) operator. Before pre- complete way by using a function for the formation of the order-
senting this new result, let us define the concept of fuzzy inducing variables and the arguments as explained in Definition 6,
measure and the Choquet integral. The fuzzy measure (non- that is, by using a Quasi-IMD triplet (hro(ui), sp(xi), sq(yi)i).
additive measure) was introduced by Sugeno (1974) and it can A key feature in this new aggregation operator is that it includes
be defined as follows: a wide range of aggregation operators. For example:

Definition 8. Let X be a universal set X = {x1, x2, . . . , xn} and P(X) the
power set of X. A fuzzy measure on X is a set function on m:  The quasi-arithmetic Choquet distance integral aggregation
P(X) ? [0, 1] that satisfies the following conditions: (Quasi-CDIA): When the ordered position of the order-inducing
variables ui is the same than the ordered position of bj such that
bj is the jth largest of the jxi  yij.
(1) m(;) = 0, m(X) = 1 (boundary conditions) and
 The induced generalized Choquet distance integral aggregation
(2) if A, B 2 P(X) and A # B, then m(A) 6 m(B) (monotonicity).
(IGCDIA): When g(b) = bk.
 The generalized Choquet distance integral aggregation (GCDIA):
The Choquet integral (Choquet, 1953) can be defined as follows
When g(b) = bk and the ordered position of the order-inducing
in its discrete form:
variables ui is the same than the ordered position of bj such that
bj is the jth largest of the jxi  yij.
Definition 9. Let f be a positive real-valued function f: X ? R+ and
m be a fuzzy measure on X. The (discrete) Choquet integral of f with We can also consider a wide range of families of all the previous
respect to m is: cases following the methodology explained in Section 4 and in
X
n Merigó and Gil-Lafuente (2009a). For example, we could analyze
C m ðf1 ; f2 ; . . . ; fn Þ ¼ fðiÞ ½mðAðiÞ Þ  mðAði1Þ Þ; ð25Þ the following cases:
i¼1

where () indicates a permutation on X such that  The induced Choquet distance integral aggregation (ICDIA):
f(1) P f(2) P    P f(n). In other words, f(i) is the ith largest value in when g(b) = b.
the set {f1, f2, . . . , fn}; A(i) = {x(1), . . . , x(i)} i P 1; and A(0) = ;.  The induced quadratic Choquet distance integral aggregation
In the following, we present the ICDIA aggregation as an (IQCDIA): when g(b) = b2.
extension of the Choquet integral that uses order-inducing vari-  The induced harmonic Choquet integral aggregation: when
ables and distance measures. Note that by using distance measures g(b) = b1.
we are able to compare two sets of variables. Also by using order-  The Choquet distance integral aggregation (CDIA) (Bolton et al.,
inducing variables we can deal with complex reordering processes 2008): when g(b) = b, and the ordered position of the order-
in the analysis. This operator can be defined as follows: inducing variables ui is the same as the ordered position of bj
such that bj is the jth largest of the jxi  yij.
Definition 10. Let m be a fuzzy measure on X. An induced Choquet  The quadratic Choquet distance integral aggregation (QCDIA):
distance integral aggregation (ICDIA) operator of dimension n is a when g(b) = b2, and the ordered position of the order-inducing
function ICDIA: Rn  Rn  Rn ? R, such that: variables ui is the same as the ordered position of bj.
72 J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76

 The harmonic Choquet distance integral aggregation: when the distance between the ideal investment and the different
g(b) = b1, and the ordered position of the order-inducing vari- alternatives considered. Note that it is possible to consider a
ables ui is the same as the ordered position of bj. wide range of IOWAD operators, such as those described in
Sections 3 and 4.
We could also consider more complex situations by using inter- Step 5: Use the weighted average (WA) to aggregate the infor-
val numbers, fuzzy numbers, linguistic variables and more com- mation of the decision-makers E by using the weighting vector
plex techniques. V. The result is the collective  payoff matrix (jxhi – yhij)mn. Thus,
P
Note that the Quasi-ICDIA operator includes several other cases, jxhi  yhi j ¼ pk¼1 v k xkhi  ykhi .
such as the step-Quasi-ICDIA, the Olympic-Quasi-ICDIA, the Step 6: Calculate the aggregated results by using the IOWAD
median-Quasi-ICDIA, the centered-Quasi-ICDIA, the S-Quasi-ICDIA, operator explained in Eq. (6). Consider different particular man-
and many more. However, we believe that those presented here ifestations of the IOWAD operator by using different expres-
are some of the most relevant. sions in the weighting vector, as explained in Section 4.
Step 7: Adopt decisions according to the results found in the
6. Multiperson decision-making with the IOWAD operator previous steps. Select the alternative/s that provides the best
result/s. Moreover, establish an ordering or a ranking of the
The IOWAD operator is applicable in a wide range of situations, alternatives from the most to the least preferred alternative to
such as decision-making, statistics, engineering and economics. In enable consideration of more than one selection.
summary, all of the studies that use the Hamming distance can be
revised and extended by using this new approach. Note that this aggregation process can be summarized using the
In this paper, we consider a decision-making application in the following aggregation operator, we call the multi-person–IOWAD
selection of investments by using a multi-person analysis. The (MP–IOWAD) operator.
main motivation for using the IOWAD operator in the selection
of investments is that the decision-maker wants to decide accord- Definition 12. An MP–IOWAD operator is an aggregation operator
P
ing to a complex attitudinal character and needs to use the opin- that has a weighting vector V of dimension p with pk¼1 v k ¼ 1 and
ions of several persons (experts) to correctly assess the problem. vk 2 [0, 1], and a weighting vector W of dimension n with
P n
This can be useful in numerous situations, such as when the board j¼1 wj ¼ 1 and wj 2 [0, 1], such that:
of directors of a company wants to make a decision. Obviously, the       
MP—IOWAD hu1 ; x11 ; . . . ; xp1 ; y11 ; . . . ; yp1 i; . . . ; hun ; x1n ; . . . ; xpn ;
attitudinal character of the board of directors is highly complex be-
cause it involves the decisions of different persons, and their inter-  1  X n
yn ; . . . ; ypn i ¼ wj bj ; ð28Þ
ests may be different. j¼1
The process to follow in the selection of investments with the IO-
WAD operator in multi-person decision-making can be summarized where bj is the jxi -yij value of the MP–IOWAD triplet hui, xi, yii hav-
as follows. Note that in the literature, we may find several other ing the jth largest  k ui, k ui isk thek  order-inducing variable; and
Pp
  , x  y  is the argument variable pro-
group decision making models, e.g., Alonso, Cabrerizo, Chiclana, jxi  yi j ¼ k¼1 v k xi  yi i i
Herrera, and Herrera-Viedma (2009), Cabrerizo, Alonso, and vided by each person (or expert) represented in the form of individ-
Herrera-Viedma (2009), Wei (2009) and Xu (2009a, 2009b, 2009c). ual distances.
The MP–IOWAD operator has properties similar to those
Step 1: Let A = {A1, A2, . . . , An} be a set of finite alternatives, and explained in Section 3, such as the distinction between descending
C = {C1, C2, . . . , Cn}, a set of finite characteristics (or attributes), and ascending orders, quasi-mixtures operators, and so on.
forming the matrix (xhi)mn. Let E = {E1, E2, . . ., Ep} be a finite set The MP–IOWAD operator includes a wide range of particular
of decision makers. Let V = (v1, v2, . . ., vp) be the weighting vector cases following the methodology explained in Section 4. Thus, we
Pp
k¼1 v k ¼ 1 and vk 2 [0, 1].
of the decision-makers such that can find as special cases:
Each decision-maker provides their own payoff matrix
 
ðkÞ
xhi .  The multi-person–normalized Hamming distance (MP–NHD)
mn operator.
Step 2: Fixation of the ideal levels of each characteristic to form
 The multi-person–weighted Hamming distance (MP–WHD)
the ideal investment (see Table 1) where P is the ideal invest-
operator.
ment expressed by a fuzzy subset, Ci is the ith characteristic
 The multi-person–OWAD (MP–OWAD) operator.
to consider and yi 2 [0, 1]; i = 1, 2, . . ., n, is a number between 0
 The multi-person–OWA (MP–OWA) operator.
and 1 for the ith characteristic. Each decision-maker provides
ðkÞ  The multi-person–IOWA (MP–IOWA) operator.
their own ideal investment yi .
 The multi-person–WA (MP–WA) operator.
Step 3: Calculate the order-inducing variables (uhi)mn to be
used in the payoff matrix for each alternative h and character-
Note that it is also possible to consider more complex situations
istic i. Calculate also the weighting vector W to be used in the
by using different types of aggregation operators in the aggregation
IOWAD aggregation. Note that W = (w1, w2, . . ., wn) such that
Pn of the experts’ opinions because in Definition 12 we assumed that
j¼1 wj ¼ 1 and wj 2 [0, 1]. the experts’ opinions were aggregated by using the WA operator.
Step 4: Compare the ideal investment with the different alterna-
However, it is also possible to use the OWA operator and the IOWA
tives considered using the IOWAD operator for each expert
operator, among others. Moreover, it is possible to develop a
(person). In this step, the objective is to express numerically
similar model by using Choquet integrals obtaining the multi-
person–ICDIA (MP–ICDIA) operator.

Table 1 7. Illustrative example


Ideal investment.

C1 C2 ... Ci ... Cn In the following, we develop a brief illustrative example of the


new approach in a multi-person decision-making problem con-
P= y1 y2 ... yi ... yn
cerning investment selection.
J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76 73

Assume a decision-maker wants to invest money in a company. Table 3


After analyzing the market, he considers five possible alternatives: Characteristics of the investments – expert 2.

C1 C2 C3 C4 C5 C6
(1) Invest in a chemical company called A1. A1 0.6 0.8 0.5 0.6 0.4 0.8
(2) Invest in a food company called A2. A2 0.7 0.6 0.8 0.6 0.7 0.7
(3) Invest in a computer company called A3. A3 0.7 0.6 0.8 0.7 0.8 0.8
(4) Invest in a car company called A4. A4 0.6 0.7 0.5 0.6 0.8 0.7
A5 0.7 0.8 0.7 0.7 0.6 0.8
(5) Invest in a furniture company called A5. A6 0.6 0.4 0.8 0.7 0.6 0.7
(6) Invest in a pharmaceutical company called A6.

After careful review of the information, the group of experts


Table 4
establishes the following general information about the invest-
Characteristics of the investments – expert 3.
ments. They summarize the information of the investments in six
general characteristics C = {C1, C2, C3, C4, C5, C6}: C1 C2 C3 C4 C5 C6
A1 0.7 0.6 0.6 0.6 0.4 0.7
 C1: Benefits in the short term. A2 0.7 0.6 0.7 0.6 0.6 0.7
A3 0.6 0.5 0.8 0.5 0.8 0.8
 C2: Benefits in the mid term.
A4 0.6 0.7 0.7 0.5 0.8 0.6
 C3: Benefits in the long term. A5 0.7 0.8 0.6 0.7 0.6 0.8
 C4: Risk of the investment. A6 0.4 0.5 0.9 0.7 0.6 0.6
 C5: Difficulty of the investment.
 C6: Other factors.
Table 5
The group of company experts is constituted by three persons, Ideal investment.
each offering their own opinions regarding the results obtained
C1 C2 C3 C4 C5 C6
with each investment. The results are shown in Tables 2–4. Note
that the results are valuations (numbers) between 0 and 1. E1 0.9 0.9 1 0.9 0.8 0.9
E2 0.8 0.9 1 1 0.9 0.8
According to the objectives of the decision-maker, each expert
E3 0.8 0.9 1 1 0.9 0.8
establishes his own ideal investment. The results are shown in
Table 5.
To analyze the attitudinal character of the group of experts, we
consider that they use order-inducing variables shown in Table 6, Table 6
Order-inducing variables.
which represents the complex attitudinal character in the decision
process. Note that in this example, the decision-maker assumes a C1 C2 C3 C4 C5 C6
different attitudinal character for each alternative because the re- A1 12 6 24 17 30 14
sults given by each alternative are not equal. The main advantage A2 17 13 9 12 7 4
of using order inducing variables is that we can represent complex A3 16 14 12 10 8 6
A4 14 17 20 12 9 6
decision processes that include psychological factors such as time A5 22 25 27 14 16 18
pressure, personal affects to each alternative and other related A6 15 13 11 17 19 21
aspects.
With this information, we can make an aggregation to make a
decision. First, we aggregate the information of the three experts
to obtain a unified payoff matrix represented in the form of indi- seems to be in closest accordance with his interests. Note that all
vidual distances between the available and ideal alternatives. We of this analysis is done in uncertainty so we do not know the cor-
use the WA to obtain this matrix while assuming that rect answer until the future becomes the present.
V = (0.3, 0.3, 0.4). The results are shown in Table 7. In this example, we consider the maximum distance, the mini-
It is now possible to develop different methods based on the mum distance, the NHD, the WHD, the step-IOWAD (k = 2), the in-
IOWAD operator for the selection of an investment. We are able duced Hurwicz distance criteria (a = 0.4), the OWAD, the AOWAD,
to provide a more complete picture to the decision-maker because the IOWAD, the AIOWAD, the median and the olympic-IOWAD
we are able to consider different future scenarios. Due to our operators. We assume the following weighting vector
uncertainty, we do not know which scenario is the correct sce- W = (0.1, 0.1, 0.2, 0.2, 0.2, 0.2). The results are shown in Tables 8
nario. Therefore, the representation of different particular cases and 9.
that could happen (from the minimum to the maximum) seems As we can see, for most of the cases the best alternative is A3 be-
to be useful for gaining a complete picture of the different future cause it seems to be the one with the lowest distance to the ideal
situations. Thus, the decision-maker knows the results that can investment. However, for some particular situations, we may find
be obtained with each alternative and thus, select the one that another optimal choice. Therefore, it is of interest to establish an

Table 2 Table 7
Characteristics of the investments – expert 1. Collective results in the form of individual distances.

C1 C2 C3 C4 C5 C6 C1 C2 C3 C4 C5 C6
A1 0.7 0.8 0.6 0.7 0.5 0.9 A1 0.16 0.18 0.43 0.34 0.44 0.04
A2 0.8 0.6 0.9 0.7 0.6 0.7 A2 0.1 0.3 0.21 0.34 0.24 0.13
A3 0.5 0.4 0.8 0.3 0.8 0.8 A3 0.23 0.4 0.2 0.47 0.07 0.03
A4 0.6 0.7 0.6 0.7 0.8 0.6 A4 0.23 0.2 0.39 0.38 0.07 0.2
A5 0.9 0.8 0.4 0.7 0.7 0.8 A5 0.07 0.1 0.43 0.27 0.24 0.03
A6 0.8 0.3 0.7 0.7 0.6 0.7 A6 0.25 0.49 0.19 0.27 0.27 0.17
74 J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76

Table 8 operators. We have also extended the IOWAD by using Choquet


Aggregated results 1. integrals, thus obtaining the ICDIA and Quasi-ICDIA operators.
Maximum Minimum NHD WHD Step (k = 2) Hurwicz We have analyzed an application of the new approach in a mul-
A1 0.44 0.04 0.265 0.284 0.43 0.284 ti-person decision-making problem regarding the selection of
A2 0.34 0.1 0.22 0.224 0.3 0.118 investments. To do so, we have introduced the MP–IOWAD opera-
A3 0.47 0.03 0.233 0.217 0.4 0.11 tor. We have seen that this approach provides better information
A4 0.39 0.07 0.245 0.251 0.2 0.276 for decision-making because it is able to consider a wide range of
A5 0.43 0.03 0.19 0.211 0.1 0.334
A6 0.49 0.17 0.273 0.254 0.27 0.182
scenarios depending on the interests of the decision-maker. More-
over, by using order-inducing variables, it is possible to consider
different scenarios according to complex attitudinal characters.
We have also seen that, depending on the particular type of aggre-
gation operator used, the results may lead to different decisions.
Table 9 In future research, we expect to develop further extensions of
Aggregated results 2. this approach by using other characteristics in the decision process,
OWAD AOWAD IOWAD AIOWAD Median Olympic such as uncertain information (e.g., interval numbers, fuzzy num-
bers, and linguistic variables), weighted and generalized aggrega-
A1 0.231 0.298 0.231 0.284 0.19 0.2425
A2 0.2 0.241 0.224 0.227 0.275 0.2725 tion operators, and more complex structures. We will also consider
A3 0.193 0.27 0.217 0.27 0.335 0.285 other decision-making problems and other applications.
A4 0.217 0.267 0.235 0.267 0.305 0.22
A5 0.158 0.218 0.175 0.177 0.05 0.11 Acknowledgements
A6 0.252 0.292 0.284 0.26 0.26 0.32

We would like to thank the anonymous reviewers for


valuable comments that have improved the quality of the paper.
Support from projects JC2009-00189 and MC238206 is gratefully
Table 10 acknowledged.
Ordering of the investments.

Ordering Ordering
Appendix A. Proofs of the theorems
Maximum A2 A4 A5 A1 A3 A6 OWAD A5 A3 A2 A4 A1 A6
Minimum A3 = A5 A1 A4 A2 A6 AOWAD A5 A2 A4 A3 A6 A1
NHD IOWAD Proof of Theorem 1. Let
A5 A2 A3 A4 A1 A6 A5 A3 A2 A1 A4 A6
WHD AIOWAD X
n
A5 A3 A2 A4 A6 A1 A5 A2 A6 A4 A3 A1
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj ; ð29Þ
Step- A5 A4 A6 A2 A3 A1 Median- A5 A1 A6 A2 A4 A3 j¼1
IOWAD IOWAD
X
n
Hurwicz A3 A2 A6 A4 A1 A5 Olympic- A5 A4 A1 A2 A3 A6 f ðhu1 ; c1 ; d1 i; . . . ; hun ; cn ; dn iÞ ¼ wj ej : ð30Þ
IOWAD
j¼1

Because (hu1, x1, y1i, . . . , hun, xn, yni) is a permutation of


(hu1, c1, d1i, . . . , hun, cn, dni), and we have jxi  yij = jci  dij, for all i,
ordering of the investments for each particular case. Note that the and then
best choice is the one with the lowest distance. The results are
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ f ðhu1 ; c1 ; d1 i; . . . ; hun ; cn ; dn iÞ: 
shown in Table 10.
As we can see, depending on the particular type of distance Note that the commutativity of the IOWAD can also be studied
aggregation operator used, the results may be different, thus lead- from the context of a distance measure, which can be proved with
ing to different decisions. Note that the main advantage of using the following theorem:
the IOWAD operator is that we can use order-inducing variables
that represent complex reordering processes in the aggregation
Proof of Theorem 2. Let
in order to consider more complex information in the decision-
making problem. X
n
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj ; ð31Þ
j¼1
X
n
8. Conclusions f ðhu1 ; y1 ; x1 i; . . . ; hun ; yn ; xn iÞ ¼ wj ej : ð32Þ
j¼1
We have presented a decision-making approach that uses dis-
tance measures and induced aggregation operators. This approach Because jxi  yij = jyi  xij, for all i, then
is based on the use of the IOWAD operator. IOWAD is an extension
of the OWA operator that uses the Hamming distance and order- f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ f ðhu1 ; y1 ; x1 i; . . . ; hun ; yn ; xn iÞ: 
inducing variables in the reordering process. The main advantage
of this operator is that it is able to consider complex attitudinal
Proof of Theorem 3. Let
characters in the decision process. This is a key feature in deci-
sion-making because there are usually numerous factors that affect
X
n
decisions, such as when the decision-maker is actually a group of f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj ; ð33Þ
persons like a company’s board of directors. j¼1
To provide a more complete formulation, we have further gen- X
n
eralized the IOWAD operator by using mixture operators, general- f ðhu1 ; c1 ; d1 i; . . . ; hun ; cn ; dn iÞ ¼ wj ej : ð34Þ
ized and quasi-arithmetic means and infinitary aggregation j¼1
J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76 75

Because jxi  yij P jci  dij, for all i, then Alonso, S., Cabrerizo, F. J., Chiclana, F., Herrera, F., & Herrera-Viedma, E. (2009).
Group decision making with incomplete fuzzy linguistic preference relations.
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ P f ðhu1 ; c1 ; d1 i; . . . ; hun ; cn ; dn iÞ:  International Journal of Intelligent Systems, 24(2), 201–222.
Amin, G. R., & Emrouznejad, A. (2006). An extended minimax disparity to determine
the OWA operator weights. Computers & Industrial Engineering, 50(3), 312–316.
Beliakov, G., Pradera, A., & Calvo, T. (2007). Aggregation functions: A guide for
Proof of Theorem 4. Let max{jxi  yij} = c and min{jxi  yij} = d; practitioners. Berlin: Springer-Verlag.
then Bolton, J., Gader, P., & Wilson, J. N. (2008). Choquet integral as a distance measure.
IEEE Transactions on Fuzzy Systems, 16(4), 1107–1110.
Cabrerizo, F. J., Alonso, S., & Herrera-Viedma, E. (2009). A consensus model for group
X
n X
n X
n
decision making problems with unbalanced fuzzy linguistic information.
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj 6 wj c ¼ c wj ; ð35Þ International Journal of Information Technology and Decision Making, 8(1),
j¼1 j¼1 j¼1 109–131.
X
n X
n X
n Cheng, C. H., Wang, J. W., & Wu, M. C. (2009). OWA-weighted based clustering
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj P wj d ¼ d wj : ð36Þ method for classification problem. Expert Systems with Applications, 36(3),
j¼1 j¼1 j¼1
4988–4995.
Chiclana, F., Herrera-Viedma, E., Herrera, F., & Alonso, S. (2007). Some induced
Pn ordered weighted averaging operators and their use for solving group decision-
Because j¼1 wj ¼ 1, we get
making problems based on fuzzy preference relations. European Journal of
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ 6 c; ð37Þ Operational Research, 182(1), 383–399.
Choquet, G. (1953). Theory of capacities. Annals Institute Fourier, 5, 131–295.
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ P d: ð38Þ Emrouznejad, A. (2008). MP–OWA: The most preferred OWA operator. Knowledge-
Based Systems, 21(8), 847–851.
Therefore, Figueira, J., Greco, S., & Ehrgott, M. (2005). Multiple criteria decision analysis: State of
the art surveys. Boston: Springer.
minfjxi  yi jg 6 f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ 6 maxfjxi  yi jg:  Gil-Aluja, J. (1998). The interactive management of human resources under
uncertainty. Dordrecht: Kluwer Academic Publishers.
Gil-Aluja, J. (1999). Elements for a theory of decision in uncertainty. Dordrecht:
Kluwer Academic Publishers.
Proof of Theorem 5. Because jxi  yij = a, for all i, we have Hamming, R. W. (1950). Error-detecting and error-correcting codes. Bell Systems
Technical Journal, 29(1), 147–160.
X
n X
n X
n Herrera-Viedma, E., Chiclana, F., Herrera, F., & Alonso, S. (2007). Group decision-
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj ¼ wj a ¼ a wj : ð39Þ making model with incomplete fuzzy preference relations based on additive
consistency. IEEE Transactions on Systems, Man and Cybernetics B, 37(1),
j¼1 j¼1 j¼1
176–189.
Pn Kacprzyk, J., & Zadrozny, S. (2009). Towards a generalized and unified
Because j¼1 wj ¼ 1, we get
characterization of individual and collective choice functions under fuzzy and
nonfuzzy preferences and majority via ordered weighted average operators.
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ a: 
International Journal of Intelligent Systems, 24(1), 4–26.
Karayiannis, N. (2000). Soft learning vector quantization and clustering algorithms
based on ordered weighted aggregation operators. IEEE Transactions on Neural
Proof of Theorem 6. Let Networks, 11(5), 1093–1105.
Kaufmann, A. (1975). Introduction to the theory of fuzzy subsets. New York: Academic
X
n Press.
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj ; ð40Þ Liu, J. W., Cheng, C. H., Chen, Y. H., & Chen, T. L. (2010). OWA rough set model for
j¼1
forecasting the revenues growth rate of the electronic industry. Expert Systems
with Applications, 37(1), 610–617.
Because jxi  yij P 0, for all i, we obtain Liu, X. (2008). A general model of parameterized OWA aggregation with given
orness level. International Journal of Approximate Reasoning, 48(2), 598–627.
f ðhu1 ; x1 ; y1 i; . . . ; hun ; xn ; yn iÞ P 0:  Losonczi, L. (1971). Uber eine neue klasse von mittelwerte. Acta Scientifica
Mathematica Univ. Szeged, 32(1), 71–78.
Merigó, J. M. (2008). New extensions to the OWA operators and their application in
decision making. PhD thesis, Department of Business Administration, University
Proof of Theorem 7. Let of Barcelona, Spain (In Spanish).
Merigó, J. M. (2010). Fuzzy decision making with immediate probabilities.
X
n Computers & Industrial Engineering, 58(4), 651–657.
f ðhu1 ; x1 ; x1 i; . . . ; hun ; xn ; xn iÞ ¼ wj bj ; ð41Þ Merigó, J. M., & Casanovas, M. (2008). Using fuzzy numbers in heavy aggregation
j¼1 operators. International Journal of Information and Communication Engineering,
4(7), 487–492.
Because xi = xi, jxi  xij = 0, for all i, therefore, we get Merigó, J. M., & Casanovas, M. (2009). Induced aggregation operators in decision
making with the Dempster–Shafer belief structure. International Journal of
f ðhu1 ; x1 ; x1 i; . . . ; hun ; xn ; xn iÞ ¼ 0:  Intelligent Systems, 24(8), 934–954.
Merigó, J. M., & Casanovas, M. (2010a). Fuzzy generalized hybrid aggregation
operators and its application in decision making. International Journal of Fuzzy
Systems, 12(1), 15–24.
Proof of Theorem 8. Let
Merigó, J. M., & Casanovas, M. (2010b). The fuzzy generalized OWA operator and its
application in strategic decision making. Cybernetics & Systems, 41(5), 359–370.
X
n Merigó, J. M., Casanovas, M., & Martı́nez, L. (2010). Linguistic aggregation operators
gðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ ¼ wk ck ; ð42Þ for linguistic decision making based on the Dempster–Shafer theory of
k¼1 evidence. International Journal of Uncertainty, Fuzziness and Knowledge-Based
Xn Systems, 18(3), 287–304.
f ðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ ¼ wj bj ; ð43Þ Merigó, J. M., & Gil-Lafuente, A. M. (2007). The ordered weighted averaging distance
operator. Lectures on Modelling and Simulation, 8(1), 1–11.
j¼1
Merigó, J. M., & Gil-Lafuente, A. M. (2008a). Using the OWA operator in the
Minkowski distance. International Journal of Electrical and Computer Engineering,
Because k = j, ck = bj, and thus, jxk  ykj = jxi  yij, for all i, we obtain 3(3), 149–157.
Merigó, J. M., & Gil-Lafuente, A. M. (2008b). On the use of the OWA operator in the
gðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ Euclidean distance. International Journal of Electrical and Systems Engineering,
¼ f ðhu1 ; x1 ; y1 i; hu2 ; x2 ; y2 i; . . . ; hun ; xn ; yn iÞ  2(8), 451–457.
Merigó, J. M., & Gil-Lafuente, A. M. (2008c). The generalized adequacy coefficient
and its application in strategic decision making. Fuzzy Economic Review, 13(2),
17–36.
References Merigó, J. M., & Gil-Lafuente, A. M. (2009a). The induced generalized OWA operator.
Information Sciences, 179(6), 729–741.
Ahn, B. S. (2009). Some remarks on the LSOWA approach for obtaining OWA Merigó, J. M., & Gil-Lafuente, A. M. (2009b). OWA operators in generalized distances.
operator weights. International Journal of Intelligent Systems, 24(12), 1265–1279. International Journal of Mathematical and Computer Sciences, 5(1), 11–18.
76 J.M. Merigó, M. Casanovas / Computers & Industrial Engineering 60 (2011) 66–76

Merigó, J. M., & Gil-Lafuente, A. M. (2010). New decision making techniques and Xu, Z. S. (2005). An overview of methods for determining OWA weights.
their application in the selection of financial products. Information Sciences, International Journal of Intelligent Systems, 20(8), 843–865.
180(11), 2085–2094. Xu, Z. S. (2009a). An automatic approach to reaching consensus in multiple
Merigó, J. M., López-Jurado, P., Gracia, M. C., & Casanovas, M. (2009). A method attribute group decision making. Computers & Industrial Engineering, 56(4),
under uncertain information for the selection of students in interdisciplinary 1369–1374.
studies. International Journal of Humanities and Social Sciences, 4(4), 260–267. Xu, Z. S. (2009b). Fuzzy harmonic mean operators. International Journal of Intelligent
Mesiar, R. (1995). Choquet-like integrals. Journal of Mathematical Analysis and Systems, 24(2), 152–172.
Applications, 194(2), 477–488. Xu, Z. S. (2009c). An interactive approach to multiple attribute group decision
Mesiar, R., & Pap, E. (2008). Aggregation of infinite sequences. Information Sciences, making with multigranular uncertain linguistic information. Group Decision and
178(18), 3557–3564. Negotiation, 18(2), 119–145.
Spirkova, J. (2008). Generalization of IGOWA operators. In Proceedings of the IPMU Yager, R. R. (1988). On ordered weighted averaging aggregation operators in multi-
2008 International Conference, Torremolinos, Spain, 2008 (pp. 220–226). criteria decision making. IEEE Transactions on Systems, Man and Cybernetics B,
Sugeno, M. (1974). Theory of fuzzy integrals and its applications. PhD thesis, Tokyo 18(1), 183–190.
Institute of Technology, Japan. Yager, R. R. (1993). Families of OWA operators. Fuzzy Sets and Systems, 59(2),
Szmidt, E., & Kacprzyk, J. (2000). Distances between intuitionistic fuzzy sets. Fuzzy 125–148.
Sets and Systems, 114(3), 505–518. Yager, R. R. (2003). Induced aggregation operators. Fuzzy Sets and Systems, 137(1),
Tan, C., & Chen, X. (2010). Induced Choquet ordered averaging operator and its 59–69.
application in group decision making. International Journal of Intelligent Systems, Yager, R. R. (2004). Choquet aggregation using order inducing variables.
25(1), 59–82. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems,
Torra, V., & Narukawa, Y. (2010). Some relations between Losonczi’s based OWA 12(1), 69–88.
generalizations and the Choquet–Stieltjes integral. Soft Computing, 14(5), Yager, R. R. (2007). Centered OWA operators. Soft Computing, 11(7), 631–639.
465–472. Yager, R. R. (2009a). Weighted maximum entropy OWA aggregation with
Wang, Y. M., Luo, Y., & Liu, X. (2007). Two new models for determining OWA applications to decision making under risk. IEEE Transactions on Systems, Man
operator weights. Computers & Industrial Engineering, 52(2), 203–209. and Cybernetics A, 39(3), 555–564.
Wei, G. W. (2009). Uncertain linguistic hybrid geometric mean operator and its Yager, R. R. (2009b). Prioritized OWA aggregation. Fuzzy Optimization and Decision
application to group decision making under uncertain linguistic environment. Making, 8(3), 245–262.
International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, Yager, R. R. (2009c). On the dispersion measure of OWA operators. Information
17(2), 251–267. Sciences, 179(22), 3908–3919.
Wei, G. W., Zhao, X., & Lin, R. (2010). Some induced aggregating operators with Yager, R. R., & Filev, D. P. (1999). Induced ordered weighted averaging operators.
fuzzy number intuitionistic fuzzy information and their applications to group IEEE Transactions on Systems, Man and Cybernetics B, 29(1), 141–150.
decision making. International Journal of Computational Intelligence Systems, 3(1), Yager, R. R., & Kacprzyk, J. (1997). The ordered weighted averaging operators: Theory
84–95. and applications. Norwell, MA: Kluwer Academic Publishers.
Wu, J., Li, J. C., Li, H., & Duan, W. Q. (2009). The induced continuous ordered Zarghami, M., & Szidarovszky, F. (2009). Revising the OWA operator for fuzzy
weighted geometric operators and their application in group decision making. stochastic multi criteria decision making. European Journal of Operational
Computers & Industrial Engineering, 56(4), 1545–1552. Research, 198(1), 259–265.

You might also like