Decision Making: Applications in Management and Engineering

Vol. 4, Issue 2, 2021, pp. 76-105.

ISSN: 2560-6018
eISSN: 2620-0104
DOI: https://doi.org/10.31181/dmame210402076i

SPECIFIC CHARACTER OF OBJECTIVE METHODS FOR

DETERMINING WEIGHTS OF CRITERIA IN MCDM
PROBLEMS: Entropy, CRITIC, SD
Irik Z. Mukhametzyanov 1,2*

1 Ufa State Petroleum Technological University, Ufa, Russia

2 Center for Strategic and Interdisciplinary Research, Ufa Federal Research Centre of
the Russian Academy of Sciences, Ufa, Russia

Received: 22 March 2021;

Accepted: 18 May 2021;
Available online: 11 June 2021.

Original scientific paper

Abstract. The comparative analysis of objective methods for determining
the weights of criteria in the problems of multi-criteria decision-making is
carried out. It is shown that the use of methods for determining the
weights of criteria, based on formal processing of the decision matrix
(Entropy, CRITIC, Standard deviation) for MCDM problems in some cases
is not correct. It is demonstrated that the Entropy weighting method
(EWM) is highly sensitive to evaluation of probabilities of states based on
the decision matrix. For the Entropy method two modifications of
estimation of probabilities of states are proposed that partially eliminate
the contradictions of the basic EWM method. The first modification
(EWM.df) is based on a statistical approach and it estimates the
probabilities of states based on attribute distribution function. The second
modification (EWM.dsp) estimates the probabilities of states based on the
relative dispositions of attributes. Two options both have their supporting
rationale. The analysis of integrated weighing methods is carried out and
various options for aggregation of weights are given. An integrated EWM-
Corr-method is proposed which allows to re-allocate the weights obtained
by the Entropy method among correlated criteria.
Key words: Multi-criteria decision making, weights estimation, Entropy
weighting method, CRITIC method, Standard Deviation method, integrated
weighting methods.

* Corresponding author.
E-mail addresses: IZmukhametzyanov@gmail.com (I. Z. Mukhametzyanov).
 Specific character of objective methods for determining weights of criteria…

1. Introduction
A significant number of multi-criteria decision-making methods (MCDM) constructs
the performance indicator of alternatives, taking into account the weights of the
criteria (Hwang & Yoon, 1981; Triantaphyllou, 2000; Tzeng & Huang, 2011; Pamucar
& Savin, 2020). The weights of the criteria quantify their importance and can
significantly affect the outcome of the decision-making process. Many authors note
the high sensitivity of the solution to variations in the weights of the criteria (Barron
& Schmidt, 1988; Mareschal, 1988; Wolters & Mareschal, 1995; Li et al., 2013;
Karande et al. 2016; Mukhametzyanov & Pamučar, 2018; Zolfani et al., 2020). Given
the importance of the weighing procedure, the balanced choice requires
comprehensive analysis.
As you know, the weights of the components of a complex system can be obtained in
different ways. One of the generally accepted classifications of methods for assessing
criteria weights divides methods into three categories based on subjective, objective
and integrated or combined approach to weighing (Goodwin & Wright, 1998, Bobko
et al., 2007; Ginevicius & Podvezko, 2005; Jahan et al., 2012, Odu, 2019; Ali et al.,
2021).
The determination of the subjective weight is based on the opinion of experts or
expert groups representing the views of various stakeholders. These are such
methods as the direct ranking method (DR) (Goodwin & Wright, 1998; Roberts &
Goodwin, 2002; Von Winterfeldt & Edwards, 1986), the point allocation (PA) method
(Doyle et al., 1997; Roberts & Goodwin, 2002), the ranking method (Ahn & Park,
2008; Barron, 1992, 1996; Roberts & Goodwin, 2002; Solymosi & Dombi, 1986;
Milosevic et al., 2021), methods of programming (Pekelman & Sen, 1974; Shirland et
al., 2003; Deng et al., 2004), Delphi method (Hwang & Yoon, 1981), pair-wise
comparison (AHP) (Saaty, 1980; Takeda et al., 1987), step-wise weight assessment
ratio analysis (SWARA) (Kersuliene et al., 2010), full consistency method (FUCOM)
(Pamučar et al., 2018; Pamucar & Ecer, 2020), Level Based Weight Assessment
(LBWA) (Žižović & Pamucar, 2019). One of the important problems of subjective
methods is an assessment of the consistency of expert opinions. For example, the
AHP-method defines a consistency index, which improves the reliability of the
weight estimates. Other procedures for assessment of the consistency of expert
judgments are based on statistical methods and correlation.
The category of objective assessment methods is based on the use of information
about the criteria and their interactions contained in the decision-making matrix.
These are such methods as entropy weighting method (EWM) (Lotfi & Fallahnejad,
2010; Wu et al., 2011; He et al., 2016), CRiteria Importance Through Inter-criteria
Correlation (CRITIC) (Diakoulaki et al., 1995), standard deviation (SD) and their
modifications (Jahan et al., 2012; Žižović et al., 2020; Alosta et al., 2021). For these
methods, there is no answer to the question of how fully and objectively the limited
sample of attributes of alternatives describes the value of the criteria. Obviously, the
result is completely determined by the decision matrix. In particular, this article
provides examples of how strongly the weights obtained for this category of
methods for specially constructed decision-making matrices can differ. This paper
analyzes the impact of normalization of the decision matrix and inversion of cost
attributes on the weights estimation results. The hypersensitivity of the Entropy
method to the estimation of state probabilities based on the decision matrix is
77
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
shown. For the Entropy method, a statistical approach to assessing the probability of
states is proposed, which partially eliminates the contradictions of the formal
method.
The essence of the integrated methods consists in combining the weighing results
obtained by different methods with the subsequent use of the normalization of
values by the Sum-method (Ma et al., 1999; Xu, 2004; Ustinovičius, 2001; Jahan et al.,
2012; Vinogradova et al., 2018; Đalić et al., 2020; Singh et al., 2020). It is believed
that integration can overcome the disadvantages of subjective and objective
weighing methods. The integration of weights obtained by different methods takes
different forms. In some cases, the combination procedure strengthens (weakens)
the weights for “strong” (“weak”) criteria, when the criterion had a greater (lower)
weight in both weighing methods, in other cases, the weight of the criteria is
smoothed out when the criterion had a larger one in one of the methods and less
weight in a different weighing method.
The paper presents various options for aggregating weights. Numerical examples
show the importance of preliminary analysis of the results when choosing the final
version of the aggregation procedure, taking into account the peculiarities of the
decision-making problem. An integrated EWM-Corr method is proposed, which
allows redistributing the weights obtained in the Entropy method between
correlated criteria.
The idea that the assessment of weights is purely technical in nature and therefore
objective is erroneous. It is rather difficult to make them objective, since there can be
different problems with different data structures, varying degrees of uncertainty,
etc. The existence of many methods and their modifications leading to different
results indicates that the estimation of weights requires serious analysis. Since the
design and determination of weights can always be interpreted in terms of value
judgments, the procedure should also include the subjective opinions of individual
experts. The differencies in assessments of the surrounding world, as noted by the
famous subjective idealist George Berkeley, are due to the fact that people's
perceptions are far from identical. Their correct agreement limits the decision
obtained by majority rule.
The purpose of this work is to show the characteristic features, shortcomings and
possible contradictions of some methods for evaluating the weights of criteria for
complex and critical analysis when solving problems of making multi-criteria
decisions.

2. Preliminaries. Objective weighting methods: Entropy, CRITIC, SD

The category of objective assessment methods is based on the use of information

about the criteria and their interaction contained in the decision-making matrix. The
standard deviation or an “entropy” measure of importance have been proposed for
quantifying contrast intensity and thus deriving objective weights of criteria (Zeleny,
1982). For objective weight estimation methods, it is believed that the greater the
scatter of the values of the attribute of alternatives (the greater the difference
between the values of the elements in the column, or the greater the variance, or the
less entropy), the greater valuable information the criterion (indicator) contains and
the higher the criterion weight. The criterion in which all alternatives have the same
performance does not offer any additional information. The general structure of
78
 Specific character of objective methods for determining weights of criteria…

objective methods for assessing weights is shown in the scheme in Figure 1. Data
transformation (or normalization) is intended to co-ordinate the measurement
scales of individual attributes. In order to maintain the proportions between natural
and normalized values, linear transformations are used.

Linear data Calculation Calculation

transformation of the key of the weight as
and inversion of the indicator intensity of q
cost attributes
aij rij qj wj=qj/∑ qj

Figure 1. General structure of objective methods for evaluating the weight of

criteria.
Entropy — EWM-method, multiplication of standard deviation and correlation —
CRITIC-method and standard deviation — SD-method are used as key indicators.
The weights of the criteria are determined as the intensities of the key indicators
using the Sum-normalization.
Below are step-by-step weight estimation algorithms using the three most
commonly used methods for solving MCDM problems — Entropy, CRITIC and SD.
Nomenclature:
Ai alternatives (objects) (i=1,…, m)
Cj+, Cj- criteria or objects properties (j=1,…, n), (+)benefit, (-) cost
aij elements of decision matrix (DM)
rij normalized elements of decision matrix
𝑟̅j average value of j-th criterion
ajmax maximum element in criteria j
ajmin minimum element in criteria j
wj weight or importance of criteria (j=1,…, n)

2.1. Entropy weighting method (EWM)

The values of the decision matrix are transformed into the segment [0; 1] using Max-
Min normalization (1) with simultaneous inversion (2) of cost criteria values:
aij  a min
rij  , rij  [0;1]
j

a max  a min
j j
, (1)
a max
 aij
rij  , rij  [0;1]
j

a max
 a min
j j
. (2)
The intensity (pij) of the j-th attribute of the i-th alternative is calculated for each
criterion (Sum-method):

79
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
rij m
pij  m
, i  1,..., m, j  1,..., n; p ij 1
r ij
i 1

i 1 . (3)
To calculate the entropy (ej) and the key indicator (qj) of each criterion:
m
1
ej     pij  ln pij , j =1,..., n; ( if pij =0  pij  ln pij = 0 )
ln m i 1 , (4)
q j  1  e j , j  1,..., n
. (5)
To calculate the weight of each criterion:
n
w j  q j /  qk , j  1,..., n
k 1 . (6)
The entropy of the attributes of alternatives for each criterion is a measure of the
significance of this criterion. It is believed that the lower the entropy of the criterion,
the more valuable information the criterion contains.

2.2. CRiteria Importance Through Inter-criteria Correlation (CRITIC)

The values of the decision matrix are transformed based on the concept of the ideal
point. To determine “best” (B=bj) and “worst” (T=tj) solution ([1xn]-vector) for all
attributes and determine relative deviation matrix V[mxn]:
( aij  b j )
rij 
(b j t j )
. (7)
To determine standard deviation (s) ([1xn]-vector) for colls of V:
m
1
s j  std(V )    ( rij  r j ) 2
m  1 i 1
. (8)
To determine the linear correlation matrix (cjk) ([nxn]-matrix) for colls of V is the
(correlation coefficient between the vectors rj and rk):
m

 (r ij  r j )( rik  r k )
c jk  co rr (V )  i 1
, j, k  1,..., n . (9)
m m

 (r
i 1
ij rj) 2
 (r
i 1
ik  rk ) 2

To calculate the key indicator and weight of criteria by the formula (6):
n
q j  s j   (1  c jk ), j  1,..., n
k 1 . (10)
In the CRITIC method, the standard deviation sj is a measure of the significance of
this criterion. Allowance for the relationship between the criteria is determined
through the correlation matrix, which allows you to distribute the weight between
the correlated criteria through the coefficients of reduction (1‒c). The amount
80
 Specific character of objective methods for determining weights of criteria…

shown in expression (10) is a measure of the conflict created by j-th criterion in

relation to the rest of the criteria. Finally, the amount of information contained in the
j-th criterion is determined using multiplicative aggregation of measures by the
formula (10).
The Spearman rank correlation coefficient could be used instead of cjk, in order to
provide a more general measure of the relationship between the rank orders of the
elements included in the vectors rj and rk.

2.3. Standard deviation (SD)

The values of the decision matrix are transformed into the segment [0; 1] using Max-
Min normalization with simultaneous inversion of cost criteria values by formulas
(1) and (2).
To calculate the key indicator and weight of criteria by expression (6):
m
1
q j  s j  std( rij )    ( rij  r j ) 2
m  1 i 1
. (11)
The standard deviation of rj is a measure of the value of that criterion to the decision
making process.

3. Features of methods for evaluating the weights of criteria in MCDM

problems

3.1. Estimating the probabilities of the object states in the Entropy method
One of the main notes for the Entropy method is that the pij in the entropy method
determines the probabilities of m possible and independent states (alternatives) of
an object. It is not clear how the probabilities of states and intensities, obtained by
the Sum-method according to expression (3), are related.
If we combine expressions (1)‒(3) for profit criteria (similarly for cost criteria), we
get:
aij  a min
pij 
j

 (a ij  a min
j )
i . (12)
The pij values are the relative distances of the j-th attribute of alternatives to the
alternative with the smallest value.
Obviously, a discrete random variable is determined by a pair (xi, pi) and the values
of pi are needed to calculate the entropy. However, expression (3) determines pi with
respect to xi. In Sections 4.2 and 4.3, to determine the probabilities of possible states
of alternatives, it is proposed to use the distribution functions of the attributes of
alternatives built on the values of the decision matrix and to use relative dispositions
of attributes instead of pij.

81
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
3.2. Difference in weights of criteria when changing many alternatives
Consider a decision-making problem in which some or all of the set of alternatives
are different, and the attributes are assessed within the same criteria. Taking the
principle of uniformity and continuity of changes in the properties of an object, such
groups of alternatives exist and have attributes from a certain interval of the area of
change. Table 1 presents decision matrices for two groups of alternatives to one
problem in the context of the same criteria.
Let's form a mix-matrix of solutions DM-3 obtained by replacing the first alternative
A1 of the matrix DM-1 with the first alternative B1 from DM-2, and form more one
decision matrix DM-12 of dimension [16x5], obtained by the concatenation of the
matrices DM-1 and DM-2. Both sets of alternatives Ai and Bi are related to the context
of the problem: alternatives are objects of the same nature,
Table 1. Decision matrices DM-1 and DM-2 for two groups of alternatives
Ai and Bi in the context of the same criteria.
DM-1, criteria: benefit(+)/cost(–) DM-2, criteria: benefit(+)/cost(–)
C1+ C2 – C3+ C4+ C5 – C1+ C2 – C3+ C4+ C5 –
A1 71 4500 150 1056 478 B1 83 5322 170 1682 500
A2 85 5800 145 2680 564 B2 84 6021 155 2140 513
A3 76 5600 135 1230 620 B3 76 4219 155 1613 454
Alter
A4 74 4200 160 1480 448 B4 73 6154 157 2047 582
nativ
A5 82 6200 183 1350 615 B5 78 5453 173 2136 598
es
A6 81 6000 178 2065 580 B6 82 6030 174 1238 587
A7 80 5900 160 1650 610 B7 80 4344 172 1365 507
A8 85 6500 140 1650 667 B8 77 6114 158 1585 592
std 5.1 813 17.3 518 74.6 3.8 788 8.7 347 54.5
having similar properties within the same criteria and having approximately the
same range of variation. It should be expected that the criterion weights calculated
by the same method for different decision matrices will be approximately the same.
However, the weights of the criteria (and the ranks of alternatives) differ
significantly (Table 2) not only for different methods, but within the same method
for a different set of alternatives.
To characterize the deviation of the weights obtained in various examples and
methods, the relative error is used (the calculation is made for each j-th
components) δs-k=|w(s)‒w(k)|/w(s)·100,%.

Table 2. Weights of the criteria for the decision matrix DM-1 and DM-2 of Table
1, the mix-matrix DM-3, and the concatenation matrix DM-12.
Entropy w1 w2 w3 w4 w5 Rank (SAW)
DM-1 0.146 0.228 0.211 0.223 0.191 A2>A4>A6>…
DM-2 0.107 0.287 0.240 0.134 0.232 B3>B7>B1>…
DM-3 0.144 0.216 0.200 0.249 0.191 A2>A4>B1>…
DM-12 0.159 0.280 0.165 0.220 0.176 B3>A4>A2>…
mean 0.139 0.253 0.204 0.207 0.198
std 0.022 0.036 0.031 0.050 0.024
CRITIC
DM-1 0.144 0.147 0.322 0.214 0.173 A2>A6>A4>…
DM-2 0.194 0.199 0.229 0.203 0.176 B3>B5>B1>…
DM-3 0.179 0.147 0.292 0.219 0.164 A2>A6>A1>…
DM-12 0.187 0.184 0.254 0.193 0.182 A2>B3>B1>…
mean 0.176 0.169 0.274 0.207 0.174

82
 Specific character of objective methods for determining weights of criteria…

std 0.022 0.026 0.041 0.012 0.007

SD
DM-1 0.210 0.203 0.207 0.183 0.196 A2>A4>A6>…
DM-2 0.175 0.206 0.231 0.195 0.192 B3>B1>B5>…
DM-3 0.216 0.181 0.220 0.189 0.193 A2>B1>A4>…
DM-12 0.208 0.225 0.192 0.176 0.199 A2>B3>A4>…
mean 0.202 0.204 0.213 0.186 0.195
std 0.018 0.018 0.017 0.008 0.003
relative error of means δs-k. %
Entropy-CRITIC 19.1 40.9 15.1 31.4 36.2
CRITIC-SD 79.7 36.4 23.1 31.0 54.3
Entropy-SD 45.3 19.4 4.4 10.1 1.5

Table 2. (Continued)
DM-1 w1 w2 w3 w4 w5 Rank (SAW)
Entropy 0.146 0.228 0.211 0.223 0.191 A2>A4>A6>…
CRITIC 0.144 0.147 0.322 0.214 0.173 A2>A6>A4>…
SD 0.210 0.203 0.207 0.183 0.196 A2>A4>A6>…
DM-2
Entropy 0.107 0.287 0.240 0.134 0.232 B3>B7>B1>…
CRITIC 0.194 0.199 0.229 0.203 0.176 B3>B5>B1>…
SD 0.175 0.206 0.231 0.195 0.192 B3>B1>B5>…
DM3
Entropy 0.144 0.216 0.200 0.249 0.191 A2>A4>B1>…
CRITIC 0.179 0.147 0.292 0.219 0.164 A2>A6>A1>…
SD 0.216 0.181 0.220 0.189 0.193 A2>B1>A4>…
DM-12
Entropy 0.159 0.280 0.165 0.220 0.176 B3>A4>A2>…
CRITIC 0.187 0.184 0.254 0.193 0.182 A2>B3>B1>…
SD 0.208 0.225 0.192 0.176 0.199 A2>B3>A4>…
In this particular example, the largest variation in weights in the Entropy method is
185%, in the CRITIC method — 52% and SD — 24%. Such a wide spread leads to a
change in the ranking of alternatives (Simple Additive Weighting method — SAW).
On the one hand, all three matrices represent assessments of alternatives within the
same criteria. However, the relative difference in the estimates of the weights for the
given example for some criteria is up to 50% and more. Even when replacing one
alternative, the weights of the criteria change significantly.

3.3. Decision matrix transformations and influence on key parameters of

CRITIC and SD methods
Since key indicators (q) should not depend on scales of measurements of the
attributes, the attributes should be normalized or reduced to dimensionless values.
Considering that the number of methods of normalization and inversion totals about
20 (Jahan et al., 2015), we should expect the ambiguity of the results. 6 basic linear
methods of normalization, presented in Table 3, have a meaningful interpretation.
The IZ-method (Mukhametzyanov, 2018, 2019, 2021) is a generalization of the Max-
Min method and will be used below in the Entropy-method.

83
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
Table 3. Basic linear methods for the normalization of the decision matrix.

non displacement: rij  ai j / k j with displacement: rij  (ai j  a*j ) / k j

Max *) Sum Vec Max-Min dSum IZ

ZI
kj  ,
a max
j  a min
j

*)The short name of the normalization methods is determined by the semantic value of the compression
ratio k. The method abbreviation is also used as the name of a function that converts values in accordance
with the normalization method. For example, rij=Max(aij)= aij/ajmax. dSum-method is a combination of Max-
Min and Sum methods (see the formula (12)) and 1‒r inversion. In the IZ-method, I and Z define a segment of
normalized values that is fixed for all criteria (0 ≤ I < Z ≤ 1).

The use of linear normalization methods is due to the fact that they retain the
relative dispositions between the natural and normalized values of the alternatives:
ai j  a *j rp j  rq j a p j  aq j
if ri j    ,  p, q  1, m
kj rng( rj ) rng( a j )
, (13)
where rng(rj)=rjmax‒rjmin is the range of values for each attribute.
Preserving the “dispositions” of alternatives after normalization means that the
normalized values preserve information about the system. For linear normalization
without displacement, one degree of freedom is lost, and for normalization with
displacement, two degrees of freedom are lost. Any nonlinear normalization leads to
information distortion.
When using a general linear transformation of natural values rij=( aij‒ aj*)/kj, the
standard deviation is scaled:
s j ( aij )
s j ( rij ) 
kj
, (14)
and the matrix of pair-wise correlations does not change (invariant):
corr ( rij )  corr ( aij )
. (15)
This means that the result of CRITIC and SD depends on the choice of the decision
matrix normalization method, since the proportions between kj in different
normalization methods for different criteria will be different, and the intensity of the
qj key indicators will be different.
Figure 2 shows the domains of the normalized values of the attributes of the
alternatives and shows the intervals (m‒sj; m+sj) of deviation from the mean m. For
normalization methods, except for Max-Min and IZ, domains for different criteria
have a different range of values rng(rj). The standard deviation s correlates with the
range of domain sizes.

84
 Specific character of objective methods for determining weights of criteria…

Figure 2. Domains of normalized values of attributes of alternatives and

intervals (m‒s; m+s) for various transformations of the decision matrix DM-1.

Of the six normalization options in Table 3, the Max-Min and IZ method is preferable,
since in this case the range of values of all attributes is the same and the standard
deviation for the normalized values sj(aij)/kj of various attributes does not depend
on the range and has a natural interpretation as the degree of dispersion of values.
In all other cases of normalization, the key indicators qj = sj for various criteria are
not consistent with each other, and the result depends on the transformation
method of criteria and measurement scales. The numerical results of the example
under consideration are shown in Table 4.
The results demonstrate a significant difference in criterion weights for different
methods, which indicates the importance of choosing the correct normalization
method. For the Max-Min and IZ normalization methods, the standard deviation is
proportional to the factor (Z‒I). Therefore, the criteria weights will be equal
(highlighted in the table).

Table 4. The values of the standard deviation and weights of the criteria of the
decision matrix DM-1 for various transformation methods.
Norm. Standard deviation Weights of criteria
method C1+ C2 – C3+ C4+ C5 – w1 w2 w3 w4 w5
Max 0.060 0.125 0.095 0.193 0.112 0.103 0.214 0.162 0.331 0.191
Sum 0.008 0.018 0.014 0.039 0.016 0.084 0.190 0.144 0.411 0.170
Vec 0.023 0.051 0.039 0.107 0.046 0.086 0.192 0.147 0.403 0.172
Max-Min 0.366 0.354 0.361 0.319 0.341 0.210 0.203 0.207 0.183 0.196
dSum 0.111 0.111 0.081 0.063 0.099 0.239 0.239 0.175 0.135 0.213
IZ(0.2; 0.7) 0.183 0.177 0.180 0.160 0.170 0.210 0.203 0.207 0.183 0.196
For the CRITIC method, the general transformation (7), in contrast to the Max-Min
normalization by the formula (1) is not correct, because the “best” (B) and “worst”
(T) solution may differ from the maximum and/or minimum value. For example, if
“best” and “worst” are absolute (from the entire set of possible alternatives, and not
from a private sample in the investigated decision-making problem), then the
normalized values are within the interval (0; 1), and for each criterion the domain
85
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
range will be different. The normalized values will be offset relative to the segment
[0; 1] with all the consequences of changing the standard deviation of the attributes
of individual criteria and changing their weight. Therefore, the correct application of
the CRITIC method is possible only with the Max-Min transformation.
To invert the attributes of cost criteria, it is recommended the ReS-algorithm in two
forms (Mukhametzyanov, 2020):
1) for natural values of attributes

a ij*   aij*  a max
j *  a j * , j  C j
min *
. (16)
2) for the normalized values of attributes
1) rij  Norm( aij ), j  1,..., n

2) r ij*   rij*  rjmax  rjmin
* , j  C j
*
*
, (17)
where Norm() is one of the linear normalization methods applied to both benefit and
cost attributes; the j* index meets the cost criteria. The abbreviation of the
normalization method (Table 3) or inversion is also used as the name of the function
that transforms values in accordance with the selected method: rij=Vec(rij), vij=ReS(rij).
ReS-algorithm is universal for any linear and non-linear transformations and ReS-
inversion algorithm preserves dispositions of natural and normalized attribute
values. For the Max-Min normalization method, the inversion by the expression (2)
and the inversion using the ReS-algorithm have the same results.
According to the standard deviation expression, inverting cost attributes does not
affect the outcome.
However, in the matrix of pair-wise correlations, the sign of some of the coefficients
changes according to the rule:
T
corr (ReS( rij ))  corr ( rij )  g  g
, (18)
where  is the element-wise product of two matrices of dimension [nxn];
g  (sgn C1 , sgn C2 ,..., sgn Cn )
[1xn]-vector

 1, if C j  benefit criteria


sgn C j  
 1, if C j  cos t criteria .


For example, for the test matrix DM-1 used in this paper:
 1   1 1 1 1 1 
 1   1 1 1 1 1 
T
  
g  g   1   (1  1 1 1  1)   1 1 1 1 1 
   
 1   1 1 1 1 1 
 1   1 1 1 1 
   1
.
Negative correlation should not increase the weight of the key factor in expression
(10). Therefore, in expression (10), the correlation must be absolute and this is the
correct expression for CRITIC.

86
 Specific character of objective methods for determining weights of criteria…

3.4. How much do the weights of the Entropy method vary depending on the
distribution in the data?
To answer this question, let us turn to the graphical illustration (Fig. 3), which shows
the function f (x) = ‒x·lnx; the values pij of attributes of each alternative after
transformation Sum; the values of the components of the entropy of the j-th criterion
eij = –pij·lnpij of each alternative; and domains of values of eij for each criterion,
allowing to observe the nature of the distribution. The values –pij·lnpij for all j are
actually on the line of the function ‒x·lnx and, for clarity the images, are separated by
a parallel translation along the 0Y axis.
The entropy function f (x) has a maximum at the point 1/e≈0.368. For m states, the
probabilities pi have, on average, less value than 1/e, and are concentrated in the
vicinity of the point 1/m. Accordingly, the entropy component –pi·lnpi is
concentrated in the vicinity of the point 1/m·ln(1/m).
The values of the weight of the criterion and the entropy of the criterion are
opposite. In accordance to the expression (4), the maximum value of the criterion
entropy is equal to 1, and the criterion weight is equal to 0. This state is achieved
when all probabilities pij = 1/m for a fixed j.

Figure 3. Entropy and entropy components (‒x·lnx) for the DM-1 matrix.

This is possible provided that all alternatives have the same value for the j-th
attribute pij = Const. The greater the difference between the values of the j-th
attribute for various alternatives, the lower the value of the criterion entropy and, in
accordance to the expression (5), the greater the value of the criterion weight. This
is due to the specifics of data normalization by the Sum-method.

87
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
Therefore, the criterion j, the attributes of the alternatives of which are concentrated
closer to the smallest value ajmin, will have more weight. This is believed to indicate
that the criterion contains valuable information. A very controversial assumption.
The weak point of the entropy method for assessing weight is the high sensitivity
(hypersensitivity) of weight to the values of the entropy of various criteria. This
hypersensitivity is due to the exponential behavior of the logarithm in the vicinity of
0. Indeed, the rate of change of entropy along the i-th component is determined by
expression:

e j 1
  (ln pij  1)
pi ln m
, (19)
therefore, small changes in the values of the j-th attribute of the i-th alternative can
change the weights. For example, (see Table 2 and Fig. 3) the entropies for criteria 2
and 3 differ by 1.5%, while the weights differ by 8%. A slight error in the estimates
of the attribute of alternatives will lead to a significant change in the estimates of the
weights.
Another weak point of the entropy method is the sensitivity to the distribution of
data in the domain of pij values. The concentration of the normalized values of one of
the attributes in the vicinity of 0 sharply decreases the entropy and leads to an
increase in the weight of this criterion. Figure 4 illustrates this feature. When
replacing only one element a12 in the decision matrix DM-1 from 85 to 72, entropy of
the first criterion decreased from 0.8897 to 0.8065, and the weight of the first
criterion increased from 0.146 to 0.231 (by 158%). The ranking of alternatives has
also changed.

Figure 4. Weights of criteria and entropy with decreasing variance of the

1st attribute in the decision matrix DM-1.

Note that when the data changed, the standard deviation for the first criterion
increased from 5.120 to 5.125 (natural). The SD-method shows weight changes of
only 0.1% — from 0.2102 to 0.2104. This example demonstrates the high sensitivity
(hypersensitivity) of EWM-method to the distribution of data in the domain.

88
 Specific character of objective methods for determining weights of criteria…

The apparent efficiency of the entropy method can most likely be attributed to a
situation in which the attributes of the alternatives have approximately the same
distribution and approximately the same variance. In such a situation, the weights of
the criteria differ less significantly.

3.5. Do you need to transform (normalize) the natural values of the attributes
before applying the Entropy method?
Some studies when estimating weights do not use preliminary data transformation
(by formulas (1) and (2)), believing that it is obvious that the transformation Sum
(formula (3)) is sufficient, or they use normalization methods other than Max-Min.
For example, in (Wu et al., 2011; Li et al., 2011), values are normalized and inverted
for cost criteria using the Max and Vec methods.
The peculiarity of the transformation according to the expression expression (3)
(Sum-method) is that the normalized pij values are converted into the interval (0; 1)
with the preservation of the dispositions of natural values, but the domains for
different criteria have different sizes and are shifted relative to each other. This is
clearly seen in the diagrams for pij in Figure 5. For more details on the displacement
of domains during normalization, see the study by Mukhametzyanov (2020, 2021).
The displacement of the pij values by the j-th criterion, according to the analysis
carried out above in Section 3.4, strongly affects the entropy of the ej criteria, which
is reflected in the weights.
Normalization based on linear methods without displacement (Max, Sum, Vec) fully
preserves the proportions of natural and normalized values of attributes. However,
after the re-normalization, the equalities are valid (Mukhametzyanov, 2020):
pij =Sum(aij/kj)=Sum(Max(aij))=Sum(Sum(aij))= Sum(Vec(aij))= Sum(aij). (20)

Figure 5. Probabilities (p) of possible states of alternatives and entropy

components (‒p·lnp) of the Entropy method for the DM-1 matrix and the normalized
decision matrix DM-1 by the Max-Min and dSum methods. Two-axis technique.

89
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
Therefore, the result of calculating the weights of the criteria will not change. In
Figure 4, in particular, pij coincides for the case when the preliminary data
transformation is not performed and for the cases of linear transformation without
displacement (1 fragment).
Normalization based on linear methods with displacement (Max-Min, dSum) retains
only the dispositions of the values. However the probabilities change:
pij =Sum( (aij‒aj*)/kj ) # Sum(aij), (21)
and therefore the entropies of the criteria will change.
In a significant part of the studies (Liang et al., 2006; Dong et al., 2016; Zhao et al.,
2017; Chen et al., 2019; Wu et al., 2018), the entropy method uses preliminary
transformation of the initial data using normalization of Max-Min with simultaneous
inversion of the values of the cost criteria.
The Max-Min transformation maps attribute values of all criteria to the interval [0;
1], which aligns the domains for all criteria and eliminates their relative
displacement. The Max-Min and inverse transformation for Max-Min also preserve
the relative proportions of the values.
The subsequent transformation by the Sum-method leads to different values of the
intensities (probabilities) pij than for the natural data of the decision matrix. At least
one of the values after transformation by the Sum method is equal to 0. The
transformed system is not identical to the original one. Despite the fact that
agreement of domains for different criteria has been achieved, within the domain
there is a shift in data towards a decrease in the contribution of individual attributes
to the entropy of the criterion and a decrease in the entropy of the criterion. At least
one of the values after the stepwise transformation by the Max-Min and Sum method
is equal to 0. The transformations are uneven for various criteria (except for the
distribution in the data that is symmetric with respect to transformation). Therefore,
the previously transformed system is a different system, not identical to the original
one, and the results of the weight estimation differ significantly.
Table 5 and Figure 4 shows an example of such calculations for the first criterion of
the DM-1 matrix.

Table 5. Probabilities of states of alternatives (p), entropy (e) and weights of

criterion (w) for matrix DM-1 and normalized matrix of solutions DM-1 by Max-
Min and dSum methods. EWM-method.
pi1
DM-1 Max-Min dSum transform e1 e2 e3 e4 e5
A1 0.112 0 0.099 DM-1 0.9991 0.9949 0.9974 0.9804 0.9961
A2 0.134 0.212 0.143 Max-Min 0.8897 0.8280 0.8410 0.8323 0.8558
A3 0.120 0.076 0.115 dSum 0.9965 0.9961 0.9982 0.9989 0.9971
A4 0.117 0.046 0.109 w1 w2 w3 w4 w5
A5 0.129 0.167 0.134 (1) DM-1 0.028 0.160 0.079 0.612 0.121
A6 0.129 0.152 0.130 (2) Max-Min 0.146 0.228 0.211 0.223 0.191
A7 0.126 0.136 0.127 (3) dSum 0.265 0.295 0.137 0.081 0.222
A8 0.134 0.212 0.143 δ1-2 421.4 42.5 167.1 63.6 57.9
δ2-3 81.5 29.4 35.1 63.7 16.2

The values of the entropy and weights of the criteria calculated after performing
the preliminary data transformation can significantly differ from the analogous

90
 Specific character of objective methods for determining weights of criteria…

values without transformation. The largest variation in weights in the Entropy

method is over 400%, with no strong case for data transformation.

3.6. Do you need to apply value inversion for cost criteria before applying
Entropy?
The answer to this question is not obvious. On the one hand, the criteria are
independent and the entropy (ej) of the criterion is not related to the direction of
improving the values of alternatives. On the other hand, attribute values are ordered
from lowest to highest value. When the direction of the coordinate axis and the
origin is changed, the state probability distribution function is reflected. If the
distribution does not have symmetry with respect to such a transformation, then
formally the entropy of the system will change due to the relative displacement of
values within the domain (even in the case of using the best of the inversion
algorithms — ReS (Mukhametzyanov, 2020), which preserves the positions of the
boundaries of the domain of inverted values). For example, for the example
considered in Table 2, the weights calculated without preliminary inversion of the
values of the cost criteria (Table 6) will differ from the weights calculated with
preliminary inversion.
Table 6. Weights and ranks of decision matrices, calculated without
preliminary inversion of the values of the cost criteria. EWM-method.
w1 w2 w3 w4 w5 w1 w2 w3 w4 w5
DM-1 0.028 0.148 0.081 0.625 0.117 DM-2 0.030 0.278 0.036 0.525 0.132
rank A2 > A6 > A4 >… (SAW) rank B2 > B5 > B4 >… (SAW)
Thus, the second argument is strongly pointing in favor of pre-inverting the cost
attribute values. For the inverse of values of cost criteria, the ReS-algorithm
presented in 3.3 above is recommended.
It is not desirable to use non-linear inversion. For example, in the works of Li, et al.,
(2011), Wu, et al., (2011), a nonlinear transformation iMax=aij*min/aij*, is used, which
transforms the original system to another system that is not equivalent to the
original one, the weights of which are different.

3.7. IZ-transformation of natural values of attributes for the Entropy method

Applying linear operations of shift to a fixed point, tension-compression and
subsequent displacement, it is possible to transform the natural values of all criteria
into a fixed (or predetermined) interval of values [a; b] (Mukhametzyanov, 2018,
2019, 2021) while maintaining the relative dispositions of natural values for each
attribute. A special case of such a transformation is the Max-Min normalization
method, which converts the attribute values for all criteria into the segment [0; 1].
An example of IZ-transformation of the matrix of solutions DM-1 into the segment
[0.2; 0.7] is shown above in Figure 2. IZ-transformation is in no way inferior, and
even more correct than transformation in [0; 1], which has problems with null
values. For example, there is no contribution of min-value of attributes to entropy.
As a result of the IZ-transformation, in comparison with the Max-Min
transformation, the standard deviation and the matrix of pairwise correlations will
not change. This means that such data transformation does not affect the result of
the CRITIC and SD methods. However, pij will change, and due to the nonlinearity of
91
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
the entropy components (–pij·lnpij), both the entropy and the weights of the criteria
will change.
It is natural to ask how the choice of the IZ-transformation interval [a; b] affects the
entropy and weights of the criteria?
Due to the strict monotonicity of the entropy components on the interval [0, 1/e]
(see Fig. 3), all changes will also be strictly monotonic, except for the cases when any
of the values of pij are greater than 1/e. Figure 6 shows a color map of entropy values

Figure 6. Color map of entropy values for the 5th criterion of DM-1 depending on
the size and position of the transformation interval [a; b] with respect to [0; 1].

for the 5th criterion (the results are similar for other criteria) depending on the size
and position of the transformation interval [a; b] with respect to [0; 1]. According to
such a map, entropy changes nonlinearly and strictly monotonically depending on
changes in the size and position of the domain of normalized values.
Non-linear changes in entropy leads to the fact that the dynamics of the weight
coefficients, firstly, is also non-linear (Figure 7). Secondly, there are points of change
in the priority of criteria (intersection points), for which the priority of the criterion
(its weight) changes. This indicates the dependence of the weighing result on the
choice of the size and position of the domain.

92
 Specific character of objective methods for determining weights of criteria…

Figure 7. Dynamics of entropy and weight coefficients depending on the size and
position of the domain [a; b] at IZ-transformation (correspondence according to the
Figure 6). DM-1.

The conclusion from the analysis is not encouraging. When manipulating data
transformation in the Entropy-method, you can manipulate the criteria weights.

3.8. Integrated methods

The essence of integrated methods (or aggregation methods) is to combine weighing
results obtained by different methods. For example, weights obtained using
subjective and objective weighing methods such as AHP and Entropy are often
corrected. It is assumed that the combined weighing method will reduce the
potential bias of a single subjective or objective weight, or can make up for the
deficiency of the subjective weight.
Let the weights wj(1) of the j-th criterion be obtained by one of the available methods,
and the weights wj(2) be obtained by another method. A common aggregation
procedure has the following multiplicative form:

q j  w(1)
j  wj
(2)
(22)

Aggregation (21) strengthens (weakens) the weights for “strong” (“weak”) criteria,
when the criterion has a greater (lower) weight in both weighing methods. In other
cases, the weight of the criteria is smoothed, when the criterion had greater weight

93
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
in one of the methods and less weight in the other weighing method. In some cases,
such a procedure is dangerous from the point of view of amplifying the error.
The qj value can be arranged in various forms, for example, as different variants of
the average value in N weighing methods:
‒ harmonic mean (HM)
N
1
qj  N /  (k )
k 1 w j
(23)
‒ geometric mean (GM)
1/ N
 N 
q j    w(jk ) 
 k 1  (24)
‒ arithmetic mean (AM)
N
1
qj 
N
w (k )
j
k 1 (25)
‒ root sum of squares (RSSq)
1/ 2
N 2
q j    w(jk ) 
 k 1  (26)
The weights are calculated as the intensities of the integrated indicator qj by formula
(6):
n
wj  qj /  qj
j 1
(27)
For different means, the following inequalities hold:
min ≤ HM ≤ GM ≤ AM ≤ max ≤ RSSq (28)
The choice of the method for the mean is not formalized and is determined by the
context of the problem.
The weights aggregation method is compensating. It compensates for low criterion
weights in some intermediate weighing methods with high values in others. If
compensation is acceptable, then a higher weight value for the same criterion in a
different weighing method will be required to obtain a higher weight value. With
multiplicative aggregation, a higher value of the weight of one of the criteria can be
obtained by reducing the weight of the other criterion.
Given the variability of aggregation and compensation, the integrated methods are
not general. Rather, they give the researcher a formal opportunity to eliminate the
discrepancy between the criteria weights in a particular problem.

94
 Specific character of objective methods for determining weights of criteria…

4. EWM modification

4.1. Modification of the entropy method using correlation: EWM-Corr

As discussed in section 3.8 above, key indicators of different methods can be
combined to assess criterion weights. The combination of entropy and correlation in
the following multiplicative form is relevant:
n
q j  (1  e j )   (1  c jk ), j  1,..., n
k 1 (29)
where ej is the entropy of the criterion determined by formulas (1)‒(4), cjk is the
matrix of pairwise correlations determined by the formula (9).
The first factor determines the dispersion of the attribute values, and the second
factor allows the weight to be distributed among the correlated criteria in terms of
reduction factors (1-c).
The weights of the criteria are determined by the intensity of the key indicator qj
using the Sum-normalization by the formula (6).
Table 7 shows the results of calculating the weights by the proposed EWM-Corr
method.

Table 7. Weights of the criteria and rank for the decision matrix DM-1, DM-2,
DM-3, DM-12, obtained using the EWM-Corr method.
Decision matrix w1 w2 w3 w4 w5 Rank (SAW)
DM-1 0.098 0.161 0.321 0.255 0.165 A2>A6>A4>…
DM-2 0.121 0.281 0.241 0.141 0.216 B3>B7>B1>…
DM-3 0.118 0.173 0.263 0.285 0.160 A2>A6>AB1>…
DM-12 0.145 0.230 0.219 0.244 0.163 A2>B3>B5>…
mean 0.139 0.253 0.204 0.207 0.198
std 0.022 0.036 0.031 0.050 0.024
Analysis of the results shows the redistribution of weights due to correlation
between the attributes of the alternatives. There was an increase in the weight of
3rd criterion (weak correlation of attributes with the rest) and a weakening in the
weight of 1st criterion (significant correlation of attributes with the rest).
Comparison of the results with similar estimates by the Entropy and CRITIC
methods (Table 2) does not reveal any definite pattern.
Since there are no criteria for the adequacy of various weight assessment methods,
the effectiveness of the EWM-Corr modification is relative. The choice of a particular
weighing method and its appropriateness is assessed by the presence of arguments
“for and against”. This applies to all methods without exception. In our case, the
argument is the redistribution of weights between correlated and uncorrelated
attributes.

4.2. Assessment of the entropy of the criterion using the distribution function of
the attribute of alternatives: EWM.df modification
After the decision matrix is normalized using the Max-Min transformation, the
domains of the normalized values of attribute are consistent with each other. The
priority of individual criteria due to different measurement scales is eliminated.
95
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
Information about the states of an object is determined by the distribution of values
within the domain, not by the values themselves. Therefore, the calculation of
entropy by values is not correct. A discrete random variable is determined by the
pair (xi, pi), and the values of the probabilities pi are needed to calculate the entropy.
To estimate the probabilities of possible states of alternatives for a fixed criterion, it
is necessary to construct a discrete distribution using a statistical approach by
performing the transition to frequency characteristics.
For the case of small samples, the distribution of a random variable can be
determined based on estimates of the parameters of a hypothetical distribution of a
random variable (for example, normal). The distribution assessment includes a
statistical test for testing a hypothetical distribution law (for example, the Jarque-
Bera test for checking normality (Jarque & Bera, 1987). Next, it is necessary to divide
the domain of observed values into m intervals [xij, xi+1,j] and calculate the
probability of the attribute falling into a certain interval. It does not matter that
some of the intervals obtained by dividing the domain of values remain
“unoccupied” by the actual values of the decision matrix. The frequency of the
attribute values is taken into account through the distribution parameters. This does
not exclude the existence of alternatives with such states. Potential probabilities of
states are calculated by the expression:
xi 1 j

pij   f ( x, 1 ,  2 ,...)dx
xij
, (30)
where f is the density of the feature distribution, θ1, θ2, ... are the distribution
parameters.
Below is a fragment of the calculation procedure for the case of a normal feature
distribution law, performed in MatLab using built-in functions:
% define DM(m,n)
mu=mean(DM); sigma=std(DM);
mx=max(DM); mn=min(DM); h=(mx-mn)/(m-1);
for j=1:n
[hJB, pJB] = jbtest(DM(:,j)); % Jarque-Bera test
if hJB==0
xN=mn(j)-h(j)/2 : h(j) : mx(j)+h(j)/2;
xN1=xN(1:m);
xN2=xN(2:m+1);
pd = makedist('Normal',mu(j),sigma(j));
y = cdf(pd,xN2)-cdf(pd,xN1); % cumulative distribution functions
end
pij(:,j)=y';
end

The calculations and illustration for the described algorithm are shown below in
Figure 8 and Table 8.
When calculating the probability of possible states of alternatives, the mean value,
standard deviation, the smallest and largest values of the attribute of alternatives
are used. The results of evaluating the weight of the same criteria based on a
different set of alternatives (DM-1, DM-2, DM-3, DM-12) are not so much different, as
in the case of calculating the probabilities of possible states of alternatives using
expression (1), in which aij is used to determine pij .
96
 Specific character of objective methods for determining weights of criteria…

What to do in the absence of information about the distribution of features or

rejection of the statistical test 0 hypothesis of distribution?
It remains to apply an expert approach. The subjectivity of the assessment with a
good expert is minimal. The expert's preferences are a reflection of the frequency of
the feature, which is formed intuitively on the basis of experience. Expert scores are,
in our opinion, better than statistics, which rarely have integrity and comprehensive
coverage.

Fugure 8. Probabilities of states of attributes of alternatives of the first criterion

(normal distribution, Decision Matrix from Table 1).

Table 8. Estimation of the entropy and weithts of the criterion using the EWM.df
method.
Decision EWM.df
Rank (SAW)
matrix w1 w2 w3 w4 w5
DM-1 0.205 0.205 0.187 0.221 0.182 A2>A6>A8>…
DM-2 0.156 0.233 0.238 0.177 0.196 B5>B4>B2>…
DM-3 0.205 0.220 0.174 0.229 0.173 A2>A6>A8>…
DM-12 0.211 0.231 0.171 0.215 0.173 A2>A6>A8>…
mean 0.194 0.222 0.192 0.210 0.181
std 0.025 0.013 0.031 0.023 0.011
Entropy (base)
DM-1 0.146 0.228 0.211 0.223 0.191 A2>A4>A6>…
DM-2 0.107 0.287 0.240 0.134 0.232 B3>B7>B1>…
DM-3 0.144 0.216 0.200 0.249 0.191 A2>A4>B1>…
DM-12 0.159 0.280 0.165 0.220 0.176 B3>A4>A2>…
mean 0.139 0.253 0.204 0.207 0.198
std 0.022 0.036 0.031 0.050 0.024
97
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
4.3. Estimation of the entropy of the criterion using the relative disposition of
alternatives: EWM.dsp modification
Let us arrange the values of the j-th attribute of alternatives in ascending order and
calculate the relative disposition of the ordered set of alternatives for each j-th
criterion using expression (13):
ui j  sort ( ai j , ' ascend ')
i , (31)
uq j  u p j
d jpq  ,  p, q  1,..., m
u max  u min
j j
. (32)
We get a discrete set (m‒1) of values:
( m 1), m
d j  { d 1,2 2,3
j , d j ,..., d j }
. (33)
djpq are the normalized distances between adjacent alternatives (in an ordered list)
for the j-th attribute.
To estimate the probabilities of the states of the j-th attribute for the EWM-method,
we apply not the normalized values of the attributes pij, but a discrete set (33)
consisting of (m‒1) the values of the relative dispositions of alternatives for each
criterion.
Let's justify this decision:
1) set (33) characterizes the degree of order. If the distribution of the values of
the j-th attribute is uniform:
1
d 1,2  d 2,3  ...  d (j m 1),m 
m 1 ,
j j
(34)
then the entropy of the j-th criterion is ej = 1, and its weight is 0. The order “in the
line” is ideal when the distance between neighboring elements is equal. Any
violation of the order leads to a decrease in entropy,
2) in accordance with expression (12), the pij values are the relative distances of
the ordered list of alternatives of the j-th attribute to the alternative with the
smallest value, i.e. tied to a reference point. In contrast to this, djpq characterizes not
the position, but the state, and do not depend on the position of the point on the
scale [0; 1],
3) if the distribution of system states is uniform, then the values pij determine the
probability that the random value X of the j-th attribute belongs to the interval
[ajmin; aij]:
aij  a min
pij  P(a min
j
j <X<aij )= P(0<X<pij )=
 (a ij  a min
j )
i . (35)
pij depends on the position of the attribute on the scale [0; 1] and contradicts the
concept of the probability of a state.
The djpq values determine the probability that the random value X of the j-th
attribute belongs to the interval [ujp; ujp] for ordered set of alternatives:

98
 Specific character of objective methods for determining weights of criteria…

uq j  u p j
d jpq =P( u pj <X< uqj )=
u max  u min
j j
. (36)
and correspond to the concept of the probability of a state as the probability of
belonging to a localized range of values,
4) as noted above, the relative disposition of the attribute of alternatives is
invariant with respect to linear data transformation (according to the formula (13))
or retains almost all information about the system (minus two degrees of freedom),
5) each of the elements of the set (33) represents the intensity of the state
m 1
0  d ij ( i 1)  1, d i  ( i 1)
j 1
i 1 , (37)
6) set (33) represents a differentiated series or the relative rate of change of
values in individual sections,
7) unlike pij by the formula (12) the sets {dj} for different j are not subject to
displacement relative to each other, and the size of the domains does not depend on
the normalization of Sum,
8) set (33) can be considered as an analogue of the distribution histogram.

Thus, to calculate the entropy of each criterion, we use the quantities {dj}:
m 1
1
ej     d ij,( i 1)  ln( d ij,( i 1) ), ( if d jpq =0  d jpq  ln( d jpq )= 0 )
ln( m j  1) i 1
. (38)
where mj is dimension of the set {djpq } of different attribute (without 0) values by
the j-th criterion (without zeros).
Next, we calculate the weights of the criteria using expression (6).
Notes:
i. in the EWM.dsp method we used the concept that the probability of a state
is proportional to the length of the localized gap. This concept is adequate
for the even distribution of attributes. If the distribution is not uniform (or
very different from uniform), then it is better to use the EWM.df approach
described in section 4.2 above. In any case, EWM.dsp is better than
estimates of the probability of states using expression (35) for the basic
EWM-method.
ii. the proposed approach does not require inversion of values for cost criteria,
iii. if ej = 1 for all j, then the weights should be taken equal,
iv. if k alternatives have the same attributes, then djpq = 0. (k‒1) values are
excluded from the calculations and the entropy is calculated from the (m‒k)
values.
The results of evaluating the entropy and weight of criteria by the proposed
EWM.dsp method for matrices DM-1, DM-2, DM-3, DM-12 are presented in Table 9.
The weights of the criteria are not as different as in the case of using the basic EWM-
method (Table 2).

99
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
Figure 9 shows the summary probability diagrams of alternative states and entropy
components (‒p∙lnp) for the decision matrix DM-1, obtained in various modifications
of the EWM-method.

Table 9. Estimation of the entropy and weights of the criterion using the
EWM.dsp method.
decision
matrix e1 e2 e3 e4 e5
DM-1 0.934 0.813 0.914 0.885 0.825
DM-2 0.949 0.671 0.686 0.793 0.700
DM-3 0.916 0.806 0.946 0.822 0.857
DM-12 0.977 0.802 0.917 0.756 0.844
mean 0.944 0.773 0.866 0.814 0.807
std 0.026 0.068 0.121 0.054 0.072
w1 w2 w3 w4 w5 Rank (SAW)
DM-1 0.105 0.298 0.137 0.182 0.278 A4>A2>A1>…
DM-2 0.043 0.274 0.261 0.172 0.250 B3>B7>B1>…
DM-3 0.128 0.297 0.082 0.273 0.220 A2>A4>B1>…
DM-12 0.033 0.281 0.118 0.347 0.221 A2>B3>A4>…
mean 0.077 0.288 0.150 0.244 0.242
std 0.046 0.012 0.078 0.083 0.028
Due to the equality of the attributes for some alternatives, in the calculations of the
EWM.dsp method of the presented example, the number of states involved is less
than (m‒1). The problem under study, defined by the DM-1 matrix, is weakly
sensitive to weight variation. Therefore, for all modifications of the EWM-method,
the ranking for the studied example is the same.

Fugure 9. Probabilities (p) of possible states of alternatives and entropy

components (‒p∙lnp) of the Entropy method for the decision matrix DM-1
for various modifications of the EWM-method. Two-axis technique.

100
 Specific character of objective methods for determining weights of criteria…

5. Conclusion
The importance of the weights when aggregating the attributes of alternatives is
such that alternatives with a priority on the attribute and a higher weight will
receive priority in the performance indicator. Therefore, the quality of weight
estimation for MCDM tasks is critical.
The conducted comprehensive analysis shows that the rationality of all objective
methods for evaluating the criteria weights for MCDM tasks is questionable. Specific
algorithms for objective methods for evaluating the weights of criteria still require
further study.
The first decision that needs to be made and which will greatly affect the final
results, is the choice between equal and different weights. Equal weighting is the
preferred procedure in most applications and makes the weights estimation less
subjective.
Differential weighing requires choosing the most appropriate approach for
determining the weights. The solution should be supported by theoretical
considerations that give meaning to each indicator or take into account its effect on
synthesis in accordance with the structure of the problem.
It is important to compare weighting results for different methods. Since there are
no criteria for the effectiveness of weighing methods, the discrepancy in the results
of weighting requires serious analysis. The design and determination of weights can
be interpreted in terms of value judgments, that methods based on the subjective
opinions of individual experts are preferred.
It is obvious that the use of various objective (and other classes) methods and
modifications lead in many cases to completely different values in the estimates of
the weights of the criteria. There are also no criteria for the effectiveness of the
methods. Therefore, further research, in our opinion, can be aimed at a constructive
solution to the problem — building a decision support system (DSS) for weight
estimation, including a wide range of methods, a knowledge base and an intelligent
system for analyzing and synthesizing of results. In fact, it is an extended component
of DSS for multi-criteria decision support systems.
All the algorithms described in this paper are implemented by the author in the
MatLab system, posted in a file hosting service “File Exchange MathWorks” and are
available for free use at the link (Math Works, 2021).

Funding: This research received no external funding.

Conflicts of Interest: The authors declare no conflicts of interest.

References
Ahn, B.S., & Park, K.S. (2008). Comparing methods for multiattribute decision making
with ordinal weights. Computers & Operations Research, 35 (5), 1660–1670.
Ali, Z., Mahmood, T., Ullah, K., & Khan, Q. (2021). Einstein Geometric Aggregation
Operators using a Novel Complex Interval-valued Pythagorean Fuzzy Setting with
Application in Green Supplier Chain Management. Reports in Mechanical
Engineering, 2(1), 105-134.
101
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
Alosta, A., Elmansuri, O., & Badi, I. (2021). Resolving a location selection problem by
means of an integrated AHP-RAFSI approach. Reports in Mechanical Engineering,
2(1), 135-142.
Barron, F.H. (1992). Selecting a best multiattribute alternative with partial
information about attribute weights. Acta Psychologica, 80, 91–103.
Barron, F.H., & Barrett, B.E. (1996). Decision quality using ranked attribute weights.
Management Science, 42, 1515–1523.
Barron, H., & Schmidt, C. (1988). Sensitivity analysis of additive multiattribute value
models. Operations Research, 36(1), 122-127.
Bobko P., Roth, P.L., & Buster, M.A. (2007). The Usefulness of Unit Weights in
Creating Composite Scores. A Literature Review, Application to Content Validity, and
Meta-Analysis. Organizational Research Methods, 10(4), 689–709.
Chen, P. (2019). Effects of normalization on the entropy-based TOPSIS method.
Expert Systems with Applications, 136, 33‒41.
Đalić, I., Stević, Ž., Karamasa, C., & Puška, A. (2020) A Novel Integrated Fuzzy
PIPRECIA – Interval Rough Saw Model: Green Supplier Selection. Decision Making:
Applications in Management and Engineering, 3(1), 126‒145.
Deng, M., Xu, W., & Yang, J.B. (2004). Estimating the attribute weights through
evidential reasoning and mathematical programming. International Journal of
Information Technology & Decision Making, 3, 419–428.
Diakoulaki, D., Mavrotas, G., & Papayannakis, L. (1995). Determining objective
weights in multiple criteria problems: The CRITIC method. Computers & Operations
Research, 22(7), 763‒770.
Doyle, J.R., Green, R.H., & Bottomley, P.A. (1997). Judging relative importance: Direct
rating and point allocation are not equivalent. Organizational Behavior and Human
Decision Processes, 70, 55–72.
Ginevicius, R., & Podvezko, V. (2005). Objective and subjective approaches to
determining the criterion weight in multicriteria models. Proceedings of
International Conference RelStat. Transport and Telecommunication, 6(1), 133–137.
Goodwin, P., & Wright, G. (1998). Decision Analysis for Management Judgment. (2nd
ed). Chichester: Wiley.
He, D., Xu, J., & Chen, X. (2016). Information-theoretic-entropy based weight
aggregation method in multiple-attribute group decision-making. Entropy, 18(6),
171.
Hwang, C.L., & Yoon K. (1981). Multiple Attributes Decision Making: Methods and
Applications. A State-of-the-Art Survey. Berlin: Springer-Verlag , Heidelberg, XI.
Jahan, A, & Edwards, K.L. (2015). A state-of-the-art survey on the influence of
normalization techniques in ranking: Improving the materials selection process in
engineering design. Materials & Design, 65, 335–342.
Jahan, A., Mustapha, F., Sapuan, S.M., Ismail, M. Y., Bahraminasab, M. (2012). A
framework for weighting of criteria in ranking stage of material selection process.
The International Journal of Advanced Manufacturing Technology, 58, 411–420.
102
 Specific character of objective methods for determining weights of criteria…

Jarque, C. M., & Bera, A. K. (1987). A test for normality of observations and
regression residuals. International Statistical Review, 55(2), 163–172.
Karande, P., Zavadskas, E., Chakraborty, S. (2016). A study on the ranking
performance of some MCDM methods for industrial robot selection problems.
International Journal of Industrial Engineering Computations, 7, 399–422.
Kersuliene, V., Zavadskas, E. K., & Turskis, Z. (2010). Selection of rational dispute
resolution method by applying new step-wise weight assessment ratio analysis
(SWARA). Journal of Business Economics and Management, 11(2), 243–258.
Kirkwood, C.W., & Corner, J.L. (1993). The effectiveness of partial information about
attribute weights for ranking alternatives in multiattribute decision making.
Organizational Behavior and Human Decision Processes, 54, 456–476.
Li, P., Qian, H., Wu, J., & Chen, J. (2013). Sensitivity analysis of TOPSIS method in
water quality assessment: I. Sensitivity to the parameter weights. Environmental
Monitoring and Assessment, 185, 2453–2461.
Li, X., Wang, K,. Liu, L., Xin J., Yang, H., Gao, C. (2011). Application of the Entropy
Weight and TOPSIS Method in Safety Evaluation of Coal Mines, Procedia Engineering,
First International Symposium on Mine Safety Science and Engineering, 26, 2085-
2091.
Liang, J., Shi, Z. D., Wierman, M.J. (2016). Information entropy, rough entropy and
knowledge granulation in incomplete information systems. International Journal of
General Systems, 35, 641–654.
Lotfi, F. H.; Fallahnejad, R. (2010). Imprecise Shannon’s entropy and multi attribute
decision making. Entropy, 12, 53–62.
Ma, J., Fan, Z. P., & Huang, L. H. (1999). A subjective and objective integrated
approach to determine attribute weights. European Journal of Operational Research,
112, 397–404.
Mareschal, B. (1988). Weight stability intervals in multicriteria decision aid.
European Journal of Operational Research, 33, 54‒64.
Math Works. File Exchange. Weight of Criteria by Irik Mukhametzyanov (2021).
https://www.mathworks.com/matlabcentral/fileexchange/91720-weight-of-
criteria Accessed 06 May 2021.
Milosevic, T., Pamucar, D., Chatterjee, P. (2021). Model for selecting a route for the
transport of hazardous materials using a fuzzy logic system. Military Technical
Courier, 69(2), 355-390.
Mukhametzyanov, I. & Pamučar, D. (2018). Sensitivity analysis in MCDM problems: A
statistical approach. Decision Making: Applications in Management and Engineering,
1(2), 51‒80.
Mukhametzyanov, I. Z. (2018). IZ-norm method of normalization the
multidimensional data, United States Copyright Office, Reg# TXu002125819/2018-
09-28, https://cocatalog.loc.gov/cgi-bin/Pwebrecon.cgi?DB=local&PAGE=First

103
 Mukhametzyanov /Decis. Mak. Appl. Manag. Eng. 4 (2) (2021) 75-105
Mukhametzyanov, I. Z. (2019). A new method for the normalization of
multidimensional data (IZ-Method) for MCDM problems, Proc. the 25st International
Conference on Multiple Criteria Decision Making (MCDM 2019), Istanbul, Turkey
(2019), 96–97.
Mukhametzyanov, I. Z. (2020). ReS-algorithm for converting normalized values of
cost criteria into benefit criteria in MCDM tasks. International Journal of Information
Technology & Decision Making, 19(5), 1389–1423.
Mukhametzyanov, I. Z. (2021). Elimination of the domains’ displacement of the
normalized values in MCDM tasks: the IZ-method. International Journal of
Information Technology & Decision Making, [In Press].
Odu, G. O. (2019). Weighting Methods for Multi-Criteria Decision Making Technique.
Journal of Applied Sciences and Environmental Management, 23(8), 1449–1457.
Pamučar D., Stević Ž., & Sremac S. (2018). A new model for determining weight
coefficients of criteria in MCDM models: Full Consistency Method (FUCOM).
Symmetry, 10(9), 393.
Pamucar, D., & Savin, L.M. (2020). Multiple-criteria model for optimal off-road
vehicle selection for passenger transportation: BWM-COPRAS model. Military
Technical Courier, 68(1), 28-64.
Pamucar, D., Ecer, F. (2020). Prioritizing the weights of the evaluation criteria under
fuzziness: The fuzzy full consistency method – FUCOM-F. Facta Universitatis, series:
Mechanical Engineering. 18(3), pp. 419 - 437.
Pekelman, D., & Sen, S.K. (1974). Mathematical programming models for the
determination of attribute weights. Management Science, 20, 1217–1229.
Roberts, R., & Goodwin P. (2002). Weight approximations in multi-attribute decision
models. Journal of Multi-Criteria Decision Analysis, 11, 291–303.
Saaty, T. L. (1980). The Analytic Hierarchy Process. New York: McGraw-Hill.
Shirland, L.E., Jesse, R.R., Thompson, R.L., & Iacovou, C.L. (2003). Determining
attribute weights using mathematical programming. Omega, 31, 423–437.
Singh, A., Ghadai, R.K., Kalita, K., Chatterjee, P., Pamucar, D. (2020). EDM process
parameter optimization for efficient machining of Inconel-718. Facta Universitatis,
series: Mechanical Engineering. 18(3), pp. 473 - 490.
Solymosi, T., & Dombi, J. (1986). A method for determining the weights of criteria:
The centralized weights. European Journal of Operational Research, 26, 35–41.
Stillwell, W. G., Seaver, D. A., & Edwards, W. (1981). A comparison of weight
approximation techniques in multiattribute utility decision making. Organizational
Behavior and Human Performance, 28, 62–77.
Takeda, E., Cogger, K. O., & Yu, P. L. (1987). Estimating criterion weights using
eigenvectors: A comparative study. European Journal of Operational Research, 29,
360–369.
Triantaphyllou, E. (2000). Multi-criteria Decision Making Methods: A Comparative
Study (Springer US).

104
 Specific character of objective methods for determining weights of criteria…

Tzeng, G-H., & Huang, J-J. (2011). Multiple Attribute Decision Making: Methods and
Application. Chapman and Hall/CRC.
Ustinovičius, L. (2001). Determining integrated weights of attributes. Statyba, 7(4),
321–326.
Vinogradova, I., Podvezko, V., & Zavadskas, E.K. (2018). The recalculation of the
weights of criteria in MCDM methods using the bayes approach. Symmetry, 10(6),
205.
Von Winterfeldt, D., & Edwards, W. (1986). Decision Analysis and Behavioral
Research, Cambridge: Cambridge University Press, MA.
Wang, Y.-M., & Luo, Y. (2010). Integration of correlations with standard deviations
for determining attribute weights in multiple attribute decision making. Math.
Comput. Model., 51, 1–12.
Wolters, W. T. M., & Mareschal, B. (1995). Novel types of sensitivity analysis for
additive MCDM methods, European Journal of Operational Research, 81(2), 281‒
290.
Wu, D., Wang, N., Yang, Z., Li, C., Yang, Y. (2018). Comprehensive Evaluation of Coal-
Fired Power Units Using grey relational analysis and a hybrid entropy-based
weighting method. Entropy, 20, 215.
Wu, J., Sun, J., Liang, L., & Zha, Y. (2011). Determination of weights for ultimate cross
efficiency using Shannon entropy. Expert Systems with Applications, 38(5), 5162–
5165.
Xu, X. (2004). A note on the subjective and objective integrated approach to
determine attribute weights. European Journal of Operational Research, 156, 530–
532.
Zeleny, M. (1982). Multiple Criteria Decision Making. McGraw-Hill, New York.
Zhao, H., Yao, L., Mei, G., Liu, T., Ning, Y. (2017). A Fuzzy comprehensive evaluation
method based on AHP and entropy for landslide susceptibility map. Entropy, 19,
396.
Žižović, M., & Pamucar, D. (2019). New model for determining criteria weights: Level
Based Weight Assessment (LBWA) model. Decision Making: Applications in
Management and Engineering, 2(2), 126‒137.
Žižović, M., Miljković, B., & Marinković, D. (2020). Objective methods for determining
criteria weight coefficients: A modification of the CRITIC method. Decision Making:
Applications in Management and Engineering, 3(2), 149-161.
Zolfani, S.H.; Yazdani, M.; Pamucar, D. & Zarate, P. (2020). A VIKOR and TOPSIS
focused reanalysis of the MADM methods based on logarithmic normalization. Facta
universitatis series: Mechanical Engineering, 18(3), 341-355.

© 2018 by the authors. Submitted for possible open access publication under
the terms and conditions of the Creative Commons Attribution (CC BY)
license (http://creativecommons.org/licenses/by/4.0/).

105