Content uploaded by Zheng Pei
Author content
All content in this area was uploaded by Zheng Pei on Nov 04, 2020
Content may be subject to copyright.
Received: 29 February 2020
|
Revised: 6 July 2020
|
Accepted: 2 August 2020
DOI: 10.1002/int.22271
RESEARCH ARTICLE
Complex q‐rung orthopair fuzzy 2‐tuple
linguistic Maclaurin symmetric mean
operators and its application to emergency
program selection
Yuan Rong
1
|Yi Liu
2,3
|Zheng Pei
1
1
School of Science, Xihua University,
Sichuan, China
2
Data Recovery Key Lab of Sichuan
Province, Neijiang Normal University,
Sichuan, China
3
Numerical Simulation Key Laboratory
of Sichuan Province, Neijiang Normal
University, Sichuan, China
Correspondence
Zheng Pei, School of Science, Xihua
University, Chengdu, 610039 Sichuan,
China.
Email: pqyz@263.net
Funding information
National Natural Science Foundation of
China, Grant/Award Number: 61372187;
Scientific and Technological Project of
Sichuan Province,
Grant/Award Number: 2019YFG0100;
Sichuan Province Youth Science and
Technology Innovation Team,
Grant/Award Number: 2019JDTD0015;
Application Basic Research Plan Project
of Sichuan Province,
Grant/Award Number: 2017JY0199;
Scientific Research Project of
Department of Education of Sichuan
Province, Grant/Award Numbers:
18ZA0273, 15TD0027; Scientific
Research Project of Neijiang Normal
University, Grant/Award Number:
18TD08; Innovation Fund of
Postgraduate Xihua University,
Abstract
This essay designs an innovate approach to work
out linguistic multiattribute group decision‐making
(MAGDM) issues with complex q‐rung orthopair
fuzzy 2‐tuple linguistic (Cq‐ROF2TL) evaluation
information. To begin with, the conception of
Cq‐ROF2TL set is propounded to express uncertain
and fuzzy assessment information. Meanwhile, the
score and accuracy function, a comparison approach,
Cq‐ROF2TL weighted averaging, and Cq‐ROF2TL
weighted geometric operator are put forward. Fur-
thermore, to take into consideration the correlation
among multiple input data, the Cq‐ROF2TL Maclaurin
symmetric mean (MSM) operator, the Cq‐ROF2TL
dual MSM operator and their weighted forms are pre-
sented. Several attractive characteristics and particular
instances of the developed operators are also explored
at length. Later, an innovative MAGDM methodology
is designed based upon the propounded operators to
settle the emergency program evaluation issue under
the Cq‐ROF2TL circumstance. Consequently, the
efficiency and outstanding superiority of the created
approach are severally substantiated by parameter ex-
ploration and detailed comparative analysis.
Int J Intell Syst. 2020;1–42. wileyonlinelibrary.com/journal/int © 2020 Wiley Periodicals LLC
|
1
Grant/Award Number: YCJJ2020028;
University Students Innovation and
Entrepreneurship Project of Xihua Cup,
Grant/Award Number: 2020107
KEYWORDS
complex q‐rung orthopair fuzzy 2‐tuple linguistic set, emergency
management program evaluation, Maclaurin symmetric mean
operator, multiattribute group decision‐making
1|INTRODUCTION
Decision‐making (DM) is a common task associated with intelligent and sophisticated ac-
tivities that take into consideration various kinds of vagueness and uncertainty in which
human beings face situations. During daily life and economic development, various selec-
tion and assessment issues can be deemed as a DM problem. Accordingly, the exploration of
DM approaches for modern decision science and management has received more attention
from multitude scholars.
1‐5
To effectual portray the ambiguous and undetermined assess-
ment data, the propounded fuzzy set (FS)
6
is an important technique during settling DM
issues in uncertain environment. Since it was reported in 1965, FS has attracted a large
number of researchers and attained investigation achievements in various aspects.
7‐11
Nonetheless, one of the deficiencies of FS is that the range of FS is limited to [0, 1], which
will lead to the situation of information loss for expressing evaluation information.
Therefore, Ramot et al.
12
originally introduced the complex fuzzy set (CFS) through ex-
panding the membership grade (MG) from real value to complex value within the unit disc.
In light of its advantages, CFS is omnipresent applied to fuzzy logic, decision science and
other science domains.
13‐15
Because the FS and CFS have a common defect which fail to
take into account the nonmembership grade (NMG) of an element belonging to the given
objective. Then Atanassov
16
advanced an expansion form of FS called intuitionistic fuzzy set
(IFS), which validly remedies the deficiency of FS through attaching a NMG. Since its
introduction, the fundament theory and application on it have gotten numerous attentions,
such as aggregation operator,
17
distance measure,
18
information entropy,
19
decision
approach
20,21
and so forth. Later on, Alkouri et al.
22
proffered the definition of complex
intuitionistic fuzzy set (CIFS) to describe the undetermined and ill‐defined judgement in-
formation in practical issues. The CIFS makes up of complex‐valued MG and complex‐
valued NMG, which are indicated through the polar coordinates. Rani and Garg
23
defined
the basic algorithm of CIFS and proffered several power operators to establish multiple
attribute decision making (MADM) method. Inspired the interval‐valued IFS presented by
Attanassov,
24
Garg and Rani
25
introduced the framework of the complex interval‐valued IFS
and studied its related operational rules and aggregation operators. For the information
measure theory, Garg and Rani
26
introduced several information measures including si-
milarity, entropies measure and so forth, and further presented a clustering algorithm on
the basis of these measures. Garg and Rani
27
propounded generalized Bonferroni mean
(BM) operators based upon Archimedean operations to integrate complex intuitionistic
fuzzy information. The more research results can refer to.
28‐30
However, if DMs judge their
evaluation information as (0.5, 0.7) for MG and NMG, the IFS cannot depict it because of
0
.5 + 0.7 = 1.2 >
1
. Hence, Yager
31
first proffered the Pythagorean fuzzy set (PFS) to por-
tray undetermined opinion for DM issues. It is obvious that the PFS is more universal than
FS and IFS because of 0.5
2
+0.7
2
= 0.74 < 1. Aiming at the PFS, Qin et al.
32
proposed some
ordered weighted distance measures to do with DM problems. Garg
33
defined novel
2
|
RONG ET AL.
logarithmic operations and propounded several logarithmic operators under Pythagorean
fuzzy environment. Liang et al.
34
combined the TOPSIS technique and three‐way decision
theory to construct a new approach to resolve DM problems. Ullah et al.
35
proposed several
distance measures of complex Pythagorean fuzzy set (CPFS) and an algorithm for addres-
sing pattern recognition issues.
Because the PFS has a prerequisite that
≤MG NMG
(
)+( )
1
22
, but when we come cross
several practical situations which the evaluation information provided by DMs in the form of
PFS cannot meet the precondition. For example, the MG and NMG are provided as (0.7, 0.8),
IFS and PFS fail to process it effectively because of
0
.7 + 0.8 >
1
and 0.7
2
+ 0.8
2
> 1. Based on
this restriction, Yager
36
developed the idea of q‐rung orthopair fuzzy set (q‐ROFS), which must
meet ≤MG NMG
(
)+( )
1
qq
. The q‐ROFS can valid dispose of the aforementioned example
because of 0.7
3
+ 0.8
3
= 0.855 < 1. It is evident that q‐ROFS has more evaluation space than IFS
and PFS because they are particular situations of q‐ROFS for
q
=
1
and
q
=
2
, respectively.
Based upon it, Liu and Wang
37
introduced several q‐rung orthopair fuzzy Archimedean
BM operators. Li et al.
38
expanded the EDAS method to q‐rung orthopair fuzzy context to
establish DM approach. Further, Liu et al.
39
presents the concept of complex q‐ROFS
(Cq‐ROFS) and Cq‐ROFLS set (Cq‐ROFLS) and further advanced several Cq‐ROFL Heronian
mean (HM) operators to build up decision model.
The above fuzzy sets can only depict information from the point of quantitative and DMs
are difficult to provide the precise numerical values to express their viewpoint. So Zadeh
40
proposed the linguistic variable (LV) to describe the qualitative information in DM pro-
blems. Thereafter, several novel notions by combining the LV and fuzzy set have been
propounded, such as intuitionistic linguistic numbers,
41
single‐valued neutrosophic lin-
guistic set
42
and linguistic q‐rung orthopair fuzzy number.
43
Furthermore, Herrera and
Martłnez
44
proposed the notion of 2‐tuple fuzzy linguistic established by a LV and a nu-
merical to prevent information loss in the procedure DM. Later on, a lot of scholars com-
bined the 2‐tuple linguistic with other fuzzy sets and propounded intuitionistic 2‐tuple
linguistic model,
45
2‐tuple linguistic PFS,
46
and so forth. These extensions can efficient
express the ill‐defined and fuzzy information in addressing DM issues.
As discussed above‐mentioned, the Cq‐ROFS and 2‐tuple linguistic model are two sig-
nificant techniques for disposing the quantitative and qualitative assessment information.
Motivated by the thought of Pythagorean fuzzy 2‐tuple linguistic, it is meaningful to pro-
pound a novel conception called Cq‐ROF2TLS through synthesizing the Cq‐ROFS and
2‐tuple linguistic to depict the complex uncertain and vague information. Obviously, the
Cq‐ROF2TLS is more universal than existing fuzzy sets because we can attain several
particular examples by taking into some special situations. When we take the parameter
q
=
1
and
q
=
2
in the context of Cq‐ROF2TLS, it will degenerate into the complex in-
tuitionistic fuzzy 2‐tuple linguistic set and complex Pythagorean fuzzy 2‐tuple linguistic set,
respectively. Besides, if the imaginary part of Cq‐ROF2TLS is assigned as zero, then it will
be simplified to q‐rung orthopair fuzzy 2‐tuple linguistic set. From previous research of
linguistic set the Cq‐ROF2TLS is stronger because: (1) it can prevent the information dis-
tortion during the process of linguistic information disposing; (2) it can avoid information
loss through utilizing complex‐valued MG complex‐valued NMG to express assessment
information; (3) it can valid tackle the problems with two dimension information in real‐life
applications.
It is well known that aggregation operator is an indispensable technique for information
fusion field and it has achieved a multitude of researcher results on diverse aspects.
RONG ET AL.
|
3
Xu
47
presented several geometric operators to aggregate intuitionistic fuzzy information.
Liu and Wang
48
propounded weighted average and geometric operator for q‐ROFS to es-
tablished MAGDM approach. However, these operators fail to take into consideration the
interrelationship of discussed attributes in decision problems. In order to conquer this
restriction, the BM and HM operator are proposed to consider the relevance of ant two data.
Nevertheless, the BM and HM operator fail to catch the interconnection among multi‐input
data. Accordingly, Maclaurin
49
originally propounded the Maclaurin symmetric mean
(MSM) operator to conquer the aforementioned defects. Thereafter, Qin and Liu
50
pro-
pounded dual MSM operator to aggregation intuitionistic fuzzy information. Liu and Qin
51
presented several linguistic intuitionistic fuzzy MSM operators to build up MAGDM
method. Wei and Lu
52
extended MSM operator to Pythagorean fuzzy environment to do
with decision problems. As we have seen, the MSM operator is not generalized to complex
fuzzy setting. Consequently, it is valuable to expand MSM operator to Cq‐ROF2TLS to
propound several novel operators.
As aforementioned discussion and analysis, the motivations of this essay are illustrated
as follows: (1) The existing theories about CFS fail to depict uncertain information through
the 2‐tuple linguistic representation madel, because it has stronger capability to describe
linguistic information and it also can avoid information distortion loss in dealing with
linguistic decision issues. Thus, we first proposed the complex q‐rung orthopair fuzzy
2‐tuple linguistic set and related fundamental conceptions, which will further rich the CFS
theories and provide a valid tool for experts to express assessment information; (2) The
information fusion plays an important role in aggregating preference information of deci-
sion experts. In addition, multitude practical issues need to take into consideration the
correlation of the identified attributes. In light of the outstanding superiority of MSM
operator, several complex q‐rung orthopair fuzzy 2‐tuple linguistic MSM operators are
created to address two‐dimensional fuzzy information; (3) The emergency program eva-
luation and selection has been viewed as vital and hot topic in management science. In view
of the complexity and unpredictability of emergency events, different types of assessment
methods need to be explored to valid evaluate emergency programs.
According to the above inspiration, the goal of this essay is to develop a innovative
decision approach to evaluate the emergency management programs. To achieve the
objective, we first find a suitable information expression tool to portray complex cognitive
information. Then we need to explore how to build up the decision algorithm and to
efficiently select the optima emergency program. Eventually, we need to confirm that the
present decision algorithm is valid and advantageous from diverse aspects. Accordingly,
the concrete aims of this article can be summarized as below:
(1) To proffer a novel conception called Cq‐ROF2TLS and several fundamental operations,
then define its score function and comparative approach;
(2) To present Cq‐ROF2TL weighted averaging (Cq‐ROF2TLWA) operator and complex
Cq‐ROF2TL weighted geometric (Cq‐ROF2TLWG) operator;
(3) To propound several MSM operators including the Cq‐ROF2TL Maclaurin symmetric
mean (Cq‐ROF2TLMSM) operator, the Cq‐ROF2TL weighted Maclaurin symmetric
mean (Cq‐ROF2TLWMSM) operator, the Cq‐ROF2TL dual Maclaurin symmetric mean
(Cq‐ROF2TLDMSM) operator and the Cq‐ROF2TL weighted Maclaurin symmetric
mean (Cq‐ROF2TLWDMSM) operator, as well as the fundamental characteristics are
discussed in detail.
4
|
RONG ET AL.
(4) To design a MAGDM methodology on the basis of these operators;
(5) To expound the validation and performance of the developed approach by an empirical
example for assessing emergency project.
To accomplish the aforementioned objectives, the overall structure of the essay is allo-
cated as below. In Section 2, we succinctly retrospect several fundamental concepts and
definitions including 2‐Tuple linguistic, Cq‐ROFS and MSM operator. Section 3propounds
the notion of Cq‐ROF2TLS, basic operation rules, comparison methodology, fundamental
operators. Section 4presents the Cq‐ROF2TLMSM, Cq‐ROF2TLWMSM, Cq‐ROF2TLDMSM,
Cq‐ROF2TLDWMSM operator, and also study several worthwhile features and particular
cases of them. Section 5is concerned with the novel MAGDM methodology on the basis of
Cq‐ROF2TLWMSM and Cq‐ROF2TLWDMSM operator. In Section 6,anevaluationproblem
of emergency program is utilized to show the efficiency and a contrastive study is performed
to highlight the merits of the developed method. Several conclusion remarks are listed in
the end.
2|PRELIMINARIES
This part succinctly retrospected several fundamental knowledge, such as 2‐tuple linguistic,
Cq‐ROFS and MSM operators.
2.1 |2‐tuple linguistic set
To efficiently dispose the vague linguistic information, Herrera and Martinez
44
first prof-
fered the 2‐tuple linguistic model based upon the symbolic transformation and LV, which is
stated as below.
Definition 1 (Herrera and Martinez
44
). Suppose
ss s={ , ,…, }
l01 be a LTS. ∈
β
l[0, ] be
a value denoting the result of a symbolic aggregation operation, then the mapping
Δ
is
utilized to acquire the 2‐tuple that expresses the equivalent information to
β
depicted as
→lS βs
Δ
:[0, ] ×[−0.5, 0.5)Δ()=(,ϱ)
.
i(1)
where
∈
i
round ββi=(),ϱ=−,ϱ[−0.5, 0.5)
,andround denotes the usual round
operation.
Definition 2 (Herrera and Martinez
44
). Suppose
ss s={ , ,…, }
l01 be a LTS and
s
(
,ϱ)
i
be
a2‐tuple. there exists a mapping
Δ
−
1
, which can attain ∈
β
l[0, ] on the basis of 2‐tuple,
described as follows:
→Sl
si β
Δ:×[−0.5, 0.5) [0, ]
Δ(,ϱ)= +ϱ=.
i
−1
−1(2)
RONG ET AL.
|
5
Based upon the aforementioned notions, we can easily find that a linguistic term sican be
shifted to a 2‐tuple linguistic, that is, ss
Δ
()=(,0)
ii
.
2.2 |Cq‐ROFS
Liu et al.
39
propounded the conception of Cq‐ROFS, which is defined as below.
Definition 3 (Liu et al.
39
). Given a fixed set
Ξ
, the Cq‐ROFS is defined as follows:
ab〈〉∣∈
Q
ζζ ζζ={ ,˙(),˙() Ξ}
,
(3)
in which
a
aa
ζe
˙=()
iπ2ζ(
)
and
b
b
b
ζe
˙=()
iπ2
ζ()
indicate complex‐valued MG and
NMG, respectively, meeting the following characteristic ab≤≤
≤
ζζ
0
()+ () 1,0
ab
≤≥q+1(1
)
ζ
q
ζ
q
() () . The indeterminacy degree is expressed as μζ()
=
ab ab
()
()
ζζe
(
1−( ( ) + ( )))
qq
iπ21−+
qζ
q
ζ
qq
1() ()
1
. Additionally, ab
ab
()
ζeζe() , ()
iπiπ22
ζζ() () is
called complex q‐rung orthopair fuzzy number (Cq‐ROFN). Conveniently, we rewrite
as ab
ab
(
)
ee,
iπiπ22
.
Definition 4 (Liu et al.
39
). Given two Cq‐ROFNs ab
ab
()
Q
ee=,
iπiπ
11
212
11
and
ab
ab
()
Q
ee=,
iπiπ
22
222
22
with ≥
λ1
, the basic algorithms between
Q1
and
Q2
are
given as follows:
aaaa bb
a
aaa bb
⊕
()
()
QQ e e
(
1) = + −,
;
qqqq
iπiπ
12 12122+−12 2
qqqqq
q
1
1212
1
12
⎛
⎝
⎜⎞
⎠
⎟(4)
aa bbbb b
aa bbb
⊗
()
()
QQ e e
(
2) = , + −
;
iπqqqq
iπ
1212
212122+−
qqq qqq
12
1
1212
1
⎛
⎝
⎜⎞
⎠
⎟(5)
ab
ab
()
()
()
()
λQee
(
3) = 1 −1−,;
qλiπλiπ
11
21−1−
12
qqλqλ
1
1
1
1
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟(6)
ab
ab
()
()
()
()
Qe e
(
4) = , 1 −1−
.
λλ
iπqλiπ
11
21
21−1−
λqqλq
1
1
1
1
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
(7)
Definition 5 (Liu et al.
39
). Given a Cq‐ROFN ab
ab
()
Q
ee=,
iπiπ22
, the score index and
accuracy index are defined as below:
ab ab
()
S
Q()=
1
2−+−
,
qq qq(8)
6
|
RONG ET AL.
ab ab
()
HQ()=
1
2++ + .
qq qq(9)
2.3 |MSM operator
Definition 6 (Maclaurin
49
). The mathematical expression of MSM operator is defined as
below:
∑∏
≤⋯ ≤
()
M
SM e e e
e
C
(, ,…, )=
kn
rrnj
kr
n
k
() 12
1<< =1
kj
k
1
1
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
(10)
in which
k
is a parameter and knrrr=1,2,…, , , ,…,
k
12 are
k
integer values taken from
the collection n{1, 2, …, } of ninteger values,
C
n
k
stands for the binomial coefficient whose
expression is C=
n
kn
kn k
!
!( −)!.
The MSM operator has the following properties:
(1)
M
SM (0,0,…,0)=
0
k() ;
(2)
M
SM e e e
e
(, ,…, )=
k()
,ifeet n=(=1,2,…,
)
t;
(3)
≤
M
SM e e e MSM e e e(, ,…, ) (
˜,˜,…, ˜
)
kk
n
() () 12 ,if ≤ee
˜
tt
. for all
t
;
(4) ≤≤eMSMeee emin { } ( , , …, ) max { }
tt kntt
() 12 .
Based upon the definition of MSM operator, the dual form of MSM operator is propounded
by Qin and Liu,
50
which is stated as below.
Definition 7. The mathematical expression of DMSM operator is defined as below:
∏∑
≤⋯ ≤
DMSM e e e ke(, ,…, )=
1
.
kn
rrnj
k
r
() 12
1<< =1
k
j
Cn
k
1
1
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟(11)
The DMSM operator has the same characteristics with the MSM operator.
3|COMPLEX Q‐RUNG ORTHOPAIR FUZZY 2‐TUPLE
LINGUISTIC SET
Inspired by the conception of 2‐tuple linguistic model and Cq‐ROFS, we shall first propound a
novel notion named Cq‐ROF2TLS, which can be viewed as a valid expansion of Cq‐ROFS.
Moreover, the fundamental algorithms, score function, accuracy function and basic operators
of Cq‐ROF2TLS are developed in detail.
RONG ET AL.
|
7
Definition 8. Given a fixed set
Ξ
, the Cq‐ROF2TLS is expressed as follows
ab
〈〉∣∈sζζζ={ ( ,ϱ), ˙(),˙() Ξ}
θζ() (12)
where aa a
∈∈sζe,ϱ[−0.5, 0.5), ˙=()
θζ iπ
() 2ζ(
)
and
b
b
b
ζe
˙=()
iπ2
ζ()
indicate complex‐
valued MG and NMG of the element
ζ
belonging to the LV s
(
,ϱ)
θζ() , respectively, meeting
the following characteristic ab≤≤ζζ
0
()+ ()
1
and ab
≤≤≥q
0
+1(1
)
ζ
q
ζ
q
() () . For
any ab ab
∈
()
()
ζμζ ζ ζeΞ,()=(1−( ( ) + ( )))
qq
iπ21−+
qζ
q
ζ
qq
1() ()
1
is called the refusal MG of
the element
ζ
belonging to the LV s
(
,ϱ)
θζ() Additionally, ab
ab
(( )
)
see=( ,ϱ),
θiπiπ
() 22
is called complex q‐rung orthopair fuzzy 2‐tuple linguistic number (Cq‐ROF2TLN).
Definition 9. Given two Cq‐ROF2TLNs ab
ab
(( ))
se e t=(,ϱ), (=1,2
)
tt
ttiπtiπ22
tt
with ≥
λ1
, the basic algorithms between
1
and
2
are given as follows:
aaaa bb
aa
aa bb
⊕
()()
()
ss e e
(1)
=ΔΔ (,ϱ)+( ,ϱ), + −,;
qqqq
iπiπ
12
−1112212122+−12 2
qqqqqq
1
1212
1
12
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟⎞
⎠
⎟
(13)
aa bbbb bb
aa bb
⊗
()()
()
ss e e
(2)
=ΔΔ (,ϱ)×( ,ϱ), , + −;
iπqqqq
iπ
12
−1112212 212122+−
qqq qqq
12
1
1212
1
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟⎞
⎠
⎟
(14)
ab
ab
()
()()
()
()
λλsee
(
3) = ΔΔ(,ϱ), 1−1−,;
qλiπλiπ
1−1111
21−1−
12
qqλqλ
1
1
1
1
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟(15)
ab
ab
()( )
() ()
()
()
se e
(
4) = ΔΔ(,ϱ), ,1−1−
.
λλλiπqλiπ
1−111121
21−1−
λqqλq
1
1
1
1
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
(16)
Definition 10. Given a Cq‐ROF2TLN ab
ab
(( )
)
see=( ,ϱ),
θiπiπ
() 22
the score index
and accuracy index are defined as below:
ab
ab
()
S
s
˜()=
−+−×Δ(,ϱ)
2
,
qq qqθ
−1() (17)
8
|
RONG ET AL.
ab
ab
()
H
s
˜()=
++ + ×Δ(,ϱ)
2
.
qq qqθ
−1() (18)
Definition 11. Given two Cq‐ROF2TLNs ab
ab
(( ))
se e t=(,ϱ), (=1,2
)
tt
ttiπtiπ22
tt ,
then the comparison algorithm of
1
and
2
is given as follows:
1. If
S
S
˜()>
˜(
)
12
, then
>
1
2
2. If
S
S
˜()=
˜(
)
12
, then
•If HH
˜()<
˜(
)
12
, then
<
1
2
•If HH
˜(=
˜(
)
12
, then
=
1
2
.
To enhance the comparison analysing, in the next, we shall present two basic
weighted operators, namely, complex q‐rung orthopair fuzzy 2‐tuple linguistic weighted
averaging and complex q‐rung orthopair fuzzy 2‐tuple linguistic weighted geometric
operator.
Definition 12. Let ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a family
of Cq‐ROF2TLNs. A mapping:
→Φ
Φ
called Cq‐ROF2TLWA operator is defined
by
⊕Cq ROF TLWA w−2(,,…,)=
nt
n
t
t
12 =1
(19)
where
Φ
stands for the collection of Cq‐ROF2TLNs,
w
t
be the weight of
t, meeting
∈
w
[0, 1]
tand
∑
w=
1
t
nt
=1 .
Theorem 1. Let ab
ab
〈〉
()
se e t n=(,ϱ), (=1,2,…,)
tt
ttiπtiπ22
tt be a family of
Cq‐ROF2TLNs,
w
t
be the weight of
t, meeting ∈
w
[0, 1]
tand
∑
w=
1
t
nt
=1 . Then the
fused result from Cq‐ROF2TLWA operator is also a Cq‐ROF2TLN and
ab
ab
∏∏ ∏
∏∏
() ()
()
Cq ROF TLWA
ws e e
−2(,,…,)
=ΔΔ(,ϱ), 1−1−,.
n
t
n
tt
t
t
n
t
qwiπ
t
n
t
wiπ
12
=1
−1
=1
21−1−
=1
2
t
q
t
n
t
qwtq
tt
n
t
wt
1
=1
1
=1
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
(20)
The proof of this theorem can be easily proved through mathematical induction, so we omit
it here. Additionally, the following properties can be easily attained.
Theorem 2. Let ab
ab
(( ))
se e=(,ϱ),
tt
ttiπtiπ22
tt
and ab
ab
(( ))
se et n′=(′,ϱ′)′,′(=1,2,…, )
tt
ttiπttiπt
2′2′
be two families of Cq‐ROF2TLNs, then the Cq‐ROF2TLWA operator possesses following
features:
RONG ET AL.
|
9
(1) Idempotentency: If all Cq‐ROF2TLNs are equal, that is,
=
tfor all
t
, then
Cq ROF TLWA−2(,,…,)=
.
n12
(2) Monotonicity: If aabb aa
≤≤≥≤ss
(
,ϱ)(′,ϱ′), ′,′,
′
tttttttt ttand
bb
≥
′
tt
. Then
≤Cq ROF TLWA Cq ROF TLWA−2(,,…,)−2(′,′,…, ′)
.
nn12 1 2
(3) Boundedness: Let
tn(=1,2,…, )
tbe a family of Cq‐ROF2TLNs, and
=min{ }, =max{ }
tt tt
−+
. Then
≤≤Cq ROF TLWA−2(,,…,).
n
−12 +
Definition 13. Let ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a family of
Cq‐ROF2TLNs. A mapping:
→Φ
Φ
called Cq‐ROF2TLWG operator is defined by
⊗Cq ROF TLWG−2(,,…,)=
nt
nt
w
12 =1
t
(21)
where
Φ
stands for the collection of Cq‐ROF2TLNs,
w
t
is the weight of
tmeeting
∈
w
[0, 1]
tand
∑
w=
1
t
nt
=1 .
Theorem 3. Let ab
ab
〈〉
()
se e t n=(,ϱ), (=1,2,…,)
tt
ttiπtiπ22
tt be a family of
Cq‐ROF2TLNs,
w
t
is the weight of
tmeeting ∈
w
[0, 1]
tand
∑
w=
1
t
nt
=1 . Then the
fused result from Cq‐ROF2TLWG operator is also a Cq‐ROF2TLN and
ab
ab
∏∏ ∏
∏∏
() ()
()
Cq ROF TLWG
se e
−2(,,…,)
=ΔΔ(,ϱ), ,1−1−.
n
t
n
tt
w
t
n
t
wiπ
t
n
t
qwiπ
12
=1
−1
=1
2
=1
21−1−
ttt
n
t
wtt
q
t
n
t
qwtq
=1
1
=1
1
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
(22)
The Cq‐ROF2TLWG operator also meets the properties of idempotentency, monotonicity,
and boundedness.
4|THE CQ‐ROF2TLMSM OPERATOR AND
CQ‐ROF2TLDMSM FOR CQ‐ROFLNS
Under this part, we generalize the MSM operator to the propounded Cq‐ROF2TL en-
vironment to build up the Cq‐ROF2TLMSM operator and Cq‐ROF2TLDMSM operator.
Moreover, we develop their weighted form named Cq‐ROF2TLWMSM operator and
Cq‐ROF2TLWDMSM operator.
10
|
RONG ET AL.
4.1 |The Cq‐ROF2TLMSM and Cq‐ROF2TLDMSM operator
Definition 14. Let ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a family of
Cq‐ROF2TLNs. A mapping:
→Φ
Φ
called Cq‐ROF2TLMSM operator is defined by
⊕⊗
≤⋯ ≤
()
Cq ROF TLMSM C
−2(,,…,)=
kn
rrnt
kr
n
k
() 12
1<< =1
kt
k
1
1
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟(23)
where
Φ
stands for the collection of Cq‐ROF2TLNs,
k
is a parameter and
knrrr=1,2,…, , , ,…,
k
12 are
k
integer values taken from the collection n{1, 2, …, } of n
integer values,
C
n
k
stands for the binomial coefficient.
Theorem 4. Let ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a family of
Cq‐ROF2TLNs, the fusion result of
{, ,…, }
n12 by utilizing Cq‐ROF2TLMSM
operator is expressed as
a
b
a
b
∏∏
∏∏
∑∏
∏∏
∏∏
()
()
()
()
()
()
Cq ROF TLMSM
s
C
e
e
−2(,,…,)
=Δ
Δ(,ϱ)
,
1−1−,
1−1−1−1−
,
kn
ξt
krr
n
k
ξt
k
r
qiπ
ξt
k
r
q
iπ
() 12
=1 −1
=1
21−1−
=1
21−1−1−1−
tt
k
t
Cn
kqk
ξt
k
rt
qCn
kqk
t
Cn
kkq
ξt
k
rt
qCn
kkq
1
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
(24)
where
ξ
denotes the subscript ≤⋯≤rrn
(
1<<
)
k1.
RONG ET AL.
|
11
Proof. In light of the basic algorithms of Cq‐ROF2TLNs, we have
a
b
a
b
∏∏ ∏
∏
∏
∏
()
()
() ()
()
se
e
=ΔΔ(,ϱ), ,
1−1−
t
k
r
t
k
rr
t
k
r
iπ
t
k
r
qiπ
=1 =1
−1
=1
2
=1
21−1−
tt
ttt
k
rt
t
q
t
k
rt
qq
=1
1
=1
1
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
and
a
b
a
b
∑∏ ∑∏
∏∏
∏∏
∏∏
∏∏
≤⋯ ≤
()
()
() ()
()
s
e
e
=ΔΔ(,ϱ),
1−1−,
1−1−
rrnt
k
r
ξt
k
rr
ξt
k
r
qiπt
ξt
k
r
qiπ
1<< =1 =1
−1
=1
21−−
=1
21−1−
k
tt
t
t
q
ξt
k
rt
qq
t
q
ξt
k
rt
qq
1
1
=1
1
1
=1
1
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
Then
a
b
a
b
∑∏ ∑∏
∏∏
∏∏
∏∏
∏∏
≤⋯ ≤
()
()
()
() ()
()
C
s
C
e
e
=Δ
Δ(,ϱ)
,
1−1−,
1−1−
rrnt
k
r
n
k
ξt
k
rr
n
k
ξt
k
r
qiπ
ξt
k
r
qiπ
1<< =1 =1
−1
=1
21−1−
=1
21−1−
kttt
t
Cn
kq
ξt
k
rt
qCn
kq
t
qCn
k
ξt
k
rt
qqCn
k
1
11
=1
11
11
=1
11
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
Accordingly
12
|
RONG ET AL.
a
b
a
b
∏∏
∏∏
∑∏
∑∏
∏∏
∏∏
≤⋯ ≤
()
()
()
()
()
()
C
s
C
e
e
=Δ
Δ(,ϱ)
,
1−1−,
1−1−1−1−
.
rrnt
kr
n
k
ξt
krr
n
k
ξt
k
r
qiπ
ξt
k
r
q
iπ
1<< =1
=1 −1
=1
21−1−
=1
21−1−1−1−
kt
k
tt
k
t
Cn
kqk
ξt
k
rt
qCn
kqk
t
Cn
kkq
ξt
k
rt
qCn
kkq
1
1
1
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
□
Theorem 5. Assume ab
ab
(( ))
se e=(,ϱ),
tt
ttiπttiπt
22
and ab
ab
(( ))
se et n′=(′,ϱ′)′,′(=1,2,…, )
tt
ttiπttiπt
2′2′
be two families of Cq‐ROF2TLNs, then the Cq‐ROF2TLWG operator possesses following
features:
(1) Idempotentency: If all Cq‐ROF2TLNs are equal, that is,
=
t
ab
ab
(( )
)
se e
=
(,ϱ), ,
iπiπ22
, for all
t
, then
Cq ROF TLMSM−2(,,…,)=.
n12
(2) Monotonicity: For
tand
tn′(=1,2,…,
)
t,if
≤
′
tt
, for all
t
. Then
≤
Cq ROF TLMSM
Cq ROF TLMSM
−2(,,…,)
−2(′,′,…, ′)
.
n
n
12
12
(3) Boundedness: Let
tn(=1,2,…, )
tbe a family of Cq‐ROF2TLNs, and
=min{ }, =max{ }
tt tt
−+
. Then
≤≤Cq ROF TLMSM−2(,,…,).
n
−12 +
RONG ET AL.
|
13
Proof.
(1) Idempotentency:
□
Cq ROF TLMSM−2(,,…,)
k()
a
b
a
b
a
b
a
b
ab
b
a
b
a
b
a
a
b
ab
∏∏
∏∏
∏∏
∏∏
∑∏
∑∏
∏∏
∏∏
∏∏
∏∏
∏
()
()
()
()
s
C
e
e
s
C
e
e
Cs
C
e
e
se
e
se e
=Δ
Δ(,ϱ)
,
1−1−,
1−1−1−(1−())
=Δ(Δ(,ϱ)),
1−1−,
1−1−1−(1−())
=Δ((Δ(,ϱ))),
1−(1−() ),
1−1−(1−(1−()))
=Δ(Δ(,ϱ)),(1−(1−() )) ,
(())
=((,ϱ), (,)).
qk
1−1
ξt
k
rr
n
k
ξt
k
r
q
iπ
ξt
k
rq
iπ
ξt
k
n
k
ξt
kq
iπ
ξt
k
q
iπ
n
kk
n
k
qk
C
iπ
qkC
iπ
qk iπ
qiπ
iπiπ
=1
−1=1
21−1−
=1
21−1−1−1−
=1
−1=1
21−1−
=1
21−1−1−(1−()
)
−1
×
21
−(1−() )
21−1−(−()
)
−1×2(1−(1−() ))
2(()
)
22
tt
kt
Cn
k
qk
ξt
k
rt
qCn
k
qk
t
Cn
kk
q
ξt
k
rt
qCn
kk
q
k
Cn
k
qk
ξt
kqCn
k
qk
Cn
kk
q
ξt
kqCn
kk
q
k
n
kCn
k
qkξ
qk
Cn
kCn
k
qk
n
kCn
kk
q
Cn
kCn
kk
q
qkqkqk
qqq
1
111
=1
111
111
=1
111
1
111
=1
111
111
=1
111
1
111×
111
111
111
11×11
11
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
14
|
RONG ET AL.
Proof.
(2) Monotonicity: Because
≤
′
tt
, then we have aa aa
≤≤≤ss
(
,ϱ)(′,ϱ′), ′,
′
tttttt tt
and
b
b
bb
≥≥′,
′
tt ttfor all
t
tn(=1,2,…, )
. Then, for the 2‐tuple linguistic
information
∑∏ ∑∏
≤
≤⋯ ≤ ≤⋯ ≤
()
() ()
CsCs
1Δ(,ϱ)1Δ′,ϱ′
.
n
k
rrnt
k
rr
n
k
rrnt
k
rr
1<< =1
−1
1<< =1
−1
k
tt
k
k
tt
k
1
1
1
1
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
Further, for the real‐valued membership degrees, because
a
a≤
′
tt
, one has
aa a a
a
a
a
a
∏∏∏∏∏∏
∏∏
∏∏
∏∏
∏∏
≤⇒ ≥
⇒
≤
⇒
≤
′1−1−′
1−1−
1−1−′
1−1−
1−1−′ .
t
k
t
q
t
k
t
q
ξt
k
t
q
ξt
k
t
q
ξt
k
t
q
ξt
k
t
q
ξt
k
t
q
ξt
k
t
q
=1 =1 =1 =1
=1
=1
=1
=1
Cn
k
Cn
k
Cn
kqk
Cn
kqk
1
1
111
111
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
Analogously, for the imaginary‐valued membership degrees, we attain
aa
∏∏ ∏∏
≤1−1−1−1−′
.
ξt
kq
ξt
kq
=1 =1
t
Cn
kqk
t
Cn
kqk
111
111
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
Accordingly, we get
RONG ET AL.
|
15
a
a
a
a
∏∏
∏∏
≤
∏∏
∏∏
e
e
1−1−
1−1−′ .
ξt
k
t
qiπ
ξt
k
t
qiπ
=1
21−1−
=1
21−1−′
Cn
kqk
ξt
k
t
qCn
kqk
Cn
kqk
ξt
k
t
qCn
k
qk
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
Similar to the testify of membership degrees, we attain
b
b
b
b
∏∏
∏∏
≥
∏∏
∏∏
()
()
()
()
()
()
e
e
1−1−1−1−
1−1−1−1−′
.
ξt
k
t
q
iπ
ξt
k
t
q
iπ
=1
21−1−1−1−()
=1
21−1−1−1−(′)
Cn
kkq
ξt
k
tqCn
kkq
Cn
kkq
ξt
k
t
qCn
kk
q
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
In light of the Definition 5, we attain ≤
S
S
˜() ˜(′
)
tt
,thatis,Cq ROF TLMS
M
−2
≤
(
)
Cq ROF TLMSM(, ,…, ) −2′,′,…, ′
nn
12 12
. That proves the monotonicity
of Cq‐ROF2TLMSM operator.
□
Proof.
(3) Boundedness: Due to the idempotentency and monotonicity of the propounded
Cq‐ROF2TLMSM operator, the following results can be attained:
For
≥min={}
ttt
−
, one has
≥Cq ROF TLMSM Cq ROF TLMSM−2(,,…,)−2(,,…,)=
n12 −− − −
For
≤max={}
ttt
+
, one has
≤Cq ROF TLMSM Cq ROF TLMSM−2(,,…,)−2(,,…,)=
n12 ++ + +
Accordingly,
≤≤Cq ROF TLMSM−2(,,…,)
n
−12 +.
16
|
RONG ET AL.
□
In the next, several novel operators shall be attained through assigning diverse values
of
k
.
(1) If k=
1
, the Cq‐ROF2TLMSM operator is yielded to Cq‐ROF2TL arithmetic averaging
(Cq‐ROF2TLAA) operator, displayed as below:
a
b
a
b
∑∏
∏
⊕⊕
∏
∏
≤≤
()
() ()
()
Cq ROF TLMSM nn
rr
ns
e
e
−2,,…,=
1=1(let = )
=Δ1Δ(,ϱ),
1−1−,
nrn rr
n
r
r
n
rrr
n
r
q
iπ
r
n
r
iπ
(1) 12 1=1
1
=1
−1=1
21−1−
=1
2
t
n
q
r
n
r
q
nq
nr
n
r
n
1
1
1
=1
11
1
=1
1
⎛
⎝
⎜⎞
⎠
⎟⎛
⎝
⎜⎞
⎠
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
(25)
(2) If k=
2
, the Cq‐ROF2TLMSM operator is yielded to Cq‐ROF2TL Bonferroni mean
(Cq‐ROF2TLBM) operator, displayed as below:
⊕⊗ ⊕⊗
≤≤
≠
() ()
Cq ROF TLMSM Cnn
−2(,,…,)= =
1
(−1)
n
rr n t r
nrr
rr
k
rr
(2) 12
1<< =1
2
2,=1
11
t
12
1
2
12
12
12
1
2
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
aa
bb
aa
bb
∑
∏
∏
∏
∏
≠
≠
≠
≠
≠
()()
()
()()
()
()()
()
()()
nn ss
e
e
Cq ROF TLBM
=
Δ1
(−1) Δ,ϱΔ,ϱ,
1−1−,
1−1−1−1−1−
=−2(,,…,).
rr
rr
k
rrrr
rr
rr
k
r
qr
q
iπ
rr
rr
k
r
qr
q
iπ
n
,=1
−1−1
,=1
21−(1−)
,=1
21−1−1−1−1−
(1,1) 12
nn q
rr
rr
k
r
qr
q
nn q
nn
q
rr
rr
k
r
q
r
q
nn
q
12
12
1122
1
2
12
12
12
2
(−1)
11
2
1,2=1
12
12
1
(−1)
11
2
12
12
12
1
(−1)
1
2
1
1,2=1
12
12
1
(−1)
1
2
1
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
(26)
RONG ET AL.
|
17
(3) If
k=3
, the Cq‐ROF2TLMSM operator is yielded to Cq‐ROF2TL generalized Bonferroni
mean (Cq‐ROF2TLGBM) operator, displayed as below:
⊕⊗
⊕⊗⊗
≤≤
≠≠
()
() ()
Cq ROF TLMSM C
nn n
Cq ROF TLBM
−2(,,…,)=
=1
(−1) −2
=−2(,,…,).
n
rrrn t r
n
rrr
rr r
k
rrr
n
(3) 12
1<< =1
3
3
,,=1
111
(1,1,1) 12
t
ijp
123
1
3
123 123
1
3
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
(27)
(4) If kn=, the Cq‐ROF2TLMSM operator is yielded to Cq‐ROF2TL geometric mean
(Cq‐ROF2TLGM) operator, displayed as below:
a
b
a
b
∏
∏
∏
⊗
∏
∏
()
() ( )
()
()
Cq ROF TLMSM
s
e
e
−2,,…,
==ΔΔ(,ϱ),
,
1−1−
.
nn
t
nt
t
n
tt
t
n
t
iπ
t
n
t
q
iπ
() 12
=1
=1
−1
=1
2
=1
21−1−
n
n
n
t
n
t
n
nq
t
n
t
qnq
1
1
1
=1
1
11
=1
11
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
(28)
In what follows, the dual operator of Cq‐ROF2TLMSM operator named Cq‐ROF2TLDMSM
operator is presented on the basis of combining the Cq‐ROF2TLS and DMSM operator named
Cq‐ROF2TLDMSM operator.
Definition 15. Let ab
ab
()
()
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a family of
Cq‐ROF2TLNs. A mapping:
→Φ
Φ
called Cq‐ROF2TLDMSM operator is defined by
⊗⊕
≤⋯ ≤
Cq ROF TLDMSM k
−2(,,…,)=
1
,
knrrn
t
k
r
() 12 1<< =1
kt
Cn
k
1
1
⎛
⎝
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟(29)
where
Φ
stands for the collection of Cq‐ROF2TLNs,
k
is a parameter,
rr r,,…,
k
12
are
k
integer
values taken from the collection n{1, 2, …, } of ninteger values,
C
n
k
stands for the binomial
coefficient.
Theorem 6. Let ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt be a family of
Cq‐ROF2TLNs, the fusion result of
{, ,…, }
tn2by using Cq‐ROF2TLMSM
operator is expressed as follows:
18
|
RONG ET AL.
a
b
a
b
∏∑
∏∏
∏∏
∏∏
∏∏
≤⋯ ≤
()
()
()
()
()
Cq ROF TLDMSM
ks
e
e
−2(,,…,)
=Δ1Δ(,ϱ),
1−1−1−1−,
1−1−
,
kn
rrn
t
k
rr
ξt
k
r
q
iπ
ξt
k
r
qiπ
() 12
1<< =1
−1
=1
21−1−1−1−
=1
21−1−
k
tt
Cn
k
t
Cn
kkq
ξt
k
rt
qCn
kkq
t
Cn
kqk
ξt
k
rt
qCn
kqk
1
1
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
(30)
where
ξ
denotes the subscript ≤⋯≤rrn
(
1<<
)
k1.
The proof is omitted because of the similarity with Theorem 4.
Theorem 7. Assume ab
ab
(( ))
se e=(,ϱ),
tt
ttiπtiπ22
tt
and ab
ab
se e
′=(( ′,ϱ′)( ′,′))
tt
ttiπtiπ22
tt
′′
be two families of Cq‐ROF2TLNs, then the Cq‐ROF2TDMSM operator possesses following
features:
(1) Idempotentency: If all Cq‐ROF2TLNs are equal, that is,
==
t
ab
ab
(
()
)
se e(,ϱ), ,
iπiπ22
, for all
t
, then
Cq ROF TLDMSM−2(,,…,)=
.
n12
(2) Monotonicity: If
aabb
aa
≤≤≥≤
()
ss
(
,ϱ)′,ϱ′,′,′,
tttttttt′
t
t
and . Then
≥
()
Cq ROF TLDMSM
Cq ROF TLDMSM
−2(,,…,)
−2′,′,…, ′
.
n
n
12
12
(3) Boundedness: Let
tn(=1,2,…, )
tbe a family of Cq‐ROF2TLNs, and
=min{ }, =max{ }
tt tt
−+
. Then
≤≤Cq ROF TLDMSM−2(,,…,).
n
−12 +
RONG ET AL.
|
19
The proof is omitted because of the similarity with Theorem 6.
Similar to Cq‐ROF2TLMSM operator, a few novel operators shall be attained through
assigning diverse parameter values of
k
.
(1) If k=
1
, the Cq‐ROF2TLDMSM operator is yielded to Cq‐ROF2TLGM operator, displayed
as below:
a
b
a
b
∏
∏
∏
⊗
∏
∏
()
()
()
()
Cq ROF TLDMSM
s
e
e
−2(,,…,)=
=ΔΔ(,ϱ),
,
1−1−
.
nt
nt
t
n
tt
t
n
t
iπ
t
n
t
q
iπ
(1) 12 =1
=1
−1
=1
2
=1
21−1−
n
n
nt
n
t
n
nq
t
n
t
qnq
1
1
1
=1
1
11
=1
11
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
(31)
(2) If k=
2
, the Cq‐ROF2TLDMSM operator is yielded to Cq‐ROF2TL geometric Bonferroni
mean (Cq‐ROF2TLGBM) operator, displayed as below:
aa
bb
aa
bb
∑
∏
∏
⊗⊕ ⊗⊕
∏
∏
≤⋯≤
≠
≠
≠
≠
≠
≠
()()
()
()
() ( )
()()
()
()
()
()
()
Cq ROF TLDMSM k
ss
e
e
Cq ROF TLDBM
−2(,,…,)=
1=1
2
=
Δ1
2Δ,ϱ+Δ,ϱ,
1−1−1−1−1−
,
1−1−
=−2(,,…,).
nrrn
t
k
rrr
rr
krr
rr
rr
k
rrrr
rr
rr
k
r
qr
q
iπ
rr
rr
k
r
qr
q
iπ
n
(2) 12 1<< =1 ,=1
,=1
−1−1
,=1
21−1−1−1−1−
,=1
21−1−
(1,1) 12
kt
Cn
knn
nn
nn
q
rr
rr
k
r
qr
q
nn
q
nn qrr
rr
k
r
q
r
q
nn
q
1
1
12
12
12
1
(−1)
12
12
1122
1
(−1)
12
12
12
1
(−1)
1
2
1
1,2=1
12
12
1
(−1)
1
2
1
12
12
12
1
(−1)
11
2
1,2=1
12
12
1
(−1)
11
2
⎛
⎝
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
(32)
20
|
RONG ET AL.
(3) If
k=3
, the Cq‐ROF2TLDMSM operator is yielded to Cq‐ROF2TL generalized geometric
Bonferroni mean (Cq‐ROF2TLGGBM) operator, displayed as below:
⊗⊕
⊗⊕⊕
≤⋯ ≤
≠≠
()
Cq ROF TLDMSM k
Cq ROF TLBM
−2(,,…,)=
1
=1
3
=−2(,,…,).
()
nrrn
t
k
r
rrr
rrr
krrr
n
(3) 12 1<< =1
,,=1
(1,1,1) 12
kt
Cn
k
nn n
1
1
122
123
123
1
(−1) −2
⎛
⎝
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
(33)
(4) If kn=, the Cq‐ROF2TLDMSM operator is yielded to Cq‐ROF2TLAM operator, displayed
as below:
a
b
a
b
∑∏
∏
⊕⊕
∏
∏
≤≤
() () ()
Cq ROF TLDMSM nn
rr
ns
e
e
−2(,,…,)=
1=1(set = )
=Δ1Δ(,ϱ),
1−1−,
.
nnrn rr
n
r
r
n
rrr
n
r
q
iπ
r
n
r
iπ
() 12 1=1
1
=1
−1=1
21−1−
=1
2
t
n
q
r
n
r
q
nq
nr
n
r
n
1
1
1
=1
11
1
=1
1
⎛
⎝
⎜⎞
⎠
⎟⎛
⎝
⎜⎞
⎠
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
(34)
4.2 |The Cq‐ROF2TLWMSM and Cq‐ROF2TLWDMSM operator
The weight of attribute is an indispensable indicator in settling practical decision issues. In
light of the situation in which Cq‐ROF2TLMSM operator fails to take into account the im-
portance of the fused data, we propound the Cq‐ROF2TLWMSM operator and Cq‐
ROF2TLWDMSM operator to make up the faultiness of Cq‐ROF2TLMSM operator as below.
Definition 16. Assume ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a set of
Cq‐ROF2TLNs,
w
t
is the weight of
tmeeting ∈
w
[0, 1]
tand
∑
w=
1
t
nt
=1 . The
Cq‐ROF2TLWMSM operator is expressed as below:
⊕⊗
≤⋯ ≤
()
Cq ROF TLWMSM C
−2(,,…,)=
()
,
kn
rrnt
krw
n
k
() 12
1<< =1
ktrtk
1
1
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟(35)
RONG ET AL.
|
21
where
Φ
stands for the collection of Cq‐ROF2TLNs,
k
is a parameter and
knrrr=1,2,…, , , ,…,
k
12 are
k
integer values taken from the collection n{1, 2, …, } of n
integer values,
C
n
k
stands for the binomial coefficient.
Theorem 8. Let ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a collection of
Cq‐ROF2TLNs, the fusion result of
{, ,…, }
tn2by Cq‐ROF2TLMSM operator is
expressed as follows:
a
b
a
b
∏∏
∏∏
∑∏
∏∏
∏∏
()
()
()
()
()
()
()
Cq ROF TLWMSM
s
C
e
e
−2(,,…,)
=Δ
Δ(,ϱ)
,
1−1−,
1−1−1−1−
kn
ξt
krr
w
n
k
ξt
k
r
w
qiπ
ξt
k
r
qw
iπ
() 12
=1 −1
=1
21−1−
=1
21−1−1−1−
tt
rtk
t
rt
Cn
kqk
ξt
k
rt
wrt
qCn
kqk
t
rtCn
kkq
ξt
k
rt
qwrtCn
kkq
1
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
(36)
where
ξ
denotes the subscript ≤⋯≤rrn
(
1<<
)
k1.
Proof. In light of the basic algorithms of Cq‐ROF2TLNs, we have
a
b
a
b
∏∏ ∏
∏
∏
∏
()
()()
() ()
()
()
se
e
()=ΔΔ(,ϱ), ,
1−1−
t
k
rw
t
k
rr
w
t
k
r
wiπ
t
k
r
qwiπ
=1 =1
−1
=1
2
=1
21−1−
trttt
rt
t
rtt
k
rt
wrt
t
rt
q
t
k
rt
qwrtq
=1
1
=1
1
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
22
|
RONG ET AL.
and
a
b
a
b
∑∏ ∑∏
∏∏
∏∏
∏∏
∏∏
≤⋯≤ ≤⋯≤
()
()
()
()
()
()
()
s
e
e
() =ΔΔ(,ϱ),
1−1−,
1−1−
rrnt
k
rw
rrnt
k
rr
w
ξt
k
r
w
qiπ
ξt
k
r
qwiπ
1
<< =1 1 << =1
−1
=1
21−1−
=1
21−1−
k
trt
k
tt
rt
t
rt
q
ξt
k
rt
wrt
qq
t
rt
q
ξt
k
rt
qwrtq
11
1
=1
1
1
=1
1
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
Then
a
b
a
b
∏∏
∏∏
∑∏ ∑∏
∏∏
∏∏
≤⋯ ≤
()
()
()
()
()
()
()
()
()
C
s
C
e
e
()
=Δ
Δ(,ϱ)
,
1−1−,
1−1−
rrnt
krw
n
k
ξt
krr
w
n
k
ξt
k
r
w
qiπ
ξt
k
r
qwiπ
1<< =1 =1 −1
=1
21−1−
=1
21−1−
ktrttt
rt
t
rt
Cn
kq
ξt
k
rt
wrt
qCn
kq
t
rt
qCn
k
ξt
k
rt
qwrtqCn
k
1
11
=1
11
11
=1
11
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
RONG ET AL.
|
23
Accordingly,
a
b
a
b
∏∏
∏∏
∑∏ ∑∏
∏∏
∏∏
≤⋯ ≤
()
() ()
()
()
()
()
()
C
s
C
e
e
()
=Δ
Δ(,ϱ)
,
1−1−,
1−1−1−1−
.
rrnt
krw
n
k
ξt
krr
w
n
k
ξt
k
r
w
qiπ
ξt
k
r
qw
iπ
1<< =1 =1 −1
=1
21−1−
=1
21−1−1−1−
ktrtk
tt
rtk
t
rt
Cn
kqk
ξt
k
rt
wrt
qCn
kqk
t
rtCn
kkq
ξt
k
rt
qwrtCn
kkq
1
11
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
□
In the next, several novel operators shall be attained through assigning diverse values of
k
.
(1) If k=
1
, the Cq‐ROF2TLWMSM operator is yielded to Cq‐ROF2TL weighted averaging
operator, displayed as below:
a
b
a
b
∑
∏
∏
⊕⊕
∏
∏
≤≤
()
()
()
()
()
() ()
()
()
()
()
Cq ROF TLMSM nn
rr
ns
e
e
−2(,,…,)=
1=1(=)
=Δ1Δ(,ϱ),
1−1−,
1−1−
nrn
r
w
r
n
r
w
r
n
rr
w
r
n
r
qw iπ
r
n
r
qw iπ
(1) 12 1=1
1
=1
−1
=1
21−1−
=1
21−1−
rr
r
r
nq
r
n
r
qwrnq
rqn
r
n
r
qwrqn
1
1
1
11
=1
11
11
=1
11
⎛
⎝
⎜⎞
⎠
⎟⎛
⎝
⎜⎞
⎠
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
(37)
24
|
RONG ET AL.
(2) If k=
2
, the Cq‐ROF2TLWMSM operator is yielded to Cq‐ROF2TL weighted BM operator,
displayed as below:
⊕⊗
⊕⊗
≤≤
≠
()
()
Cq ROF TLWMSM C
nn
−2(,,…,)=
=1
(−1)
n
rr n t r
w
n
rr
rr
k
r
w
r
w
(2) 12
1<< =1
2
2
,=1
t
rt
rr
12
1
2
12
12
1
1
2
2
1
2
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
aa
bb
bb
aa
∑
∏
∏
∏
∏
≠
≠
≠
≠
≠
()
()()
()
()
() ()
()()
()()
nn ss e
e
Cq ROF TLWBM
=
Δ1
(−1) Δ,ϱΔ,ϱ,
1−1−
,
1−1−1−1−1−
=
−2(,,…,).
r
qwr
r
qwr
−1−1
11−2
2
rr
rr
k
rr
w
rr
w
rr
rr
k
r
w
r
wq
iπ
rr
rr
k
r
qw
r
qw
iπ
n
,=1
−1−1
,=1
21−(1−(( ) ( ) ) )
,=1
21−1−
(1,1) 12
rr
rr
nn q
rr
rr
k
rwrrwrq
nn
q
rr
nn
q
rr
rr
k
nn
q
12
12
11
1
22
2
1
2
12
12
1
1
2
2
2
(−1)
11
2
1,2=1
12
1122
2
(−1)
11
2
12
12
1
1
2
2
2
(−1)
1
2
1
1,2=1
12
1
2
(−1)
1
2
1
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
(3) If
k=3
, the Cq‐ROF2TLWMSM operator is yielded to Cq‐ROF2TL generalized BM
(Cq‐ROF2TLGBM) operator, displayed as below:
⊕⊗
⊕⊗⊗
≤≤
≠≠
()
() ()
Cq ROF TLWMSM C
nn n
Cq ROF TLWBM
−2(,,…,)=
=1
(−1) −2
=−2(,,…,).
n
rr n t r
w
n
rrr
rr r
k
r
wr
wr
w
n
(3) 12
1<< =1
3
3
,,=1
(1,1,1) 12
t
rt
ijp
rrr
12
1
3
123
1
1
2
2
3
3
1
3
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎞
⎠
⎟
⎟
⎟
(38)
RONG ET AL.
|
25
(4) If kn=, the Cq‐ROF2TLMSM operator is yielded to Cq‐ROF2TL geometric mean operator,
displayed as below:
a
b
a
b
∏
∏
∏
⊗
∏
∏
()
()
()
()
()
Cq ROF TLWMSM
s
e
e
−2(,,…,)=
=ΔΔ(,ϱ),
() ,
1−1−
.
nnt
nt
w
t
n
tt
wt
n
twiπ
t
n
t
qwiπ
() 12 =1
=1
−1
=1
2
=1
21−1−
tn
t
n
t
n
t
n
t
wtn
t
nq
t
n
t
qwtnq
1
1
1
=1
1
11
=1
11
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
(39)
In what follows, the dual operator of Cq‐ROF2TLWMSM operator named
Cq‐ROF2TLWDMSM operator is presented on the basis of combining the Cq‐ROF2TLS and
DMSM operator named Cq‐ROF2TLDMSM operator.
Definition 17. Assume ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a family of
Cq‐ROF2TLNs,
w
t
is the weight of
tmeeting ∈
w
[0, 1]
tand
∑
w=
1
t
nt
=1 . The
Cq‐ROF2TLWDMSM operator is expressed as below:
⊗⊕⊗
≤⋯ ≤
Cq ROF TLWDMSM kw−2(,,…,)=
1
.
nrrn
t
k
rr12 1<< =1
ktt
Cn
k
1
1
⎛
⎝
⎜
⎜
⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
⎟(40)
In view of the fundamental operations of Cq‐ROF2TLNs, we can attain the following
Theorems.
Theorem 9. Let ab
ab
(( ))
se e t n=(,ϱ), (=1,2,…,
)
tt
ttiπtiπ22
tt
be a family of
Cq‐ROF2TLNs, the aggregation value of
{, ,…, }
n12 produced by Cq‐ROF2TLWDMSM
operator is still a Cq‐ROF2TLN and
26
|
RONG ET AL.
a
b
a
b
∏∑
∏∏
∏∏
∏∏
∏∏
≤⋯≤
()
()
()
()
()
()
Cq ROF TLWDMSM
kws
e
e
−2(,,…,)
=Δ1Δ(,ϱ),
1−1−1−1−,
1−1−
,
kn
rrn
t
k
rr
r
ξt
k
r
qw
iπ
ξt
k
r
w
qiπ
() 12
1<< =1
−1
=1
21−1−1−1−
=1
21−1−
k
ttt
Cn
k
t
rtCn
kkq
ξt
k
rt
qwrtCn
kkq
t
rtCn
kqk
ξt
k
rt
wrt
qCn
kqk
1
1
111
=1
111
111
=1
111
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
⎛
⎝
⎜
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎞
⎠
⎟
⎟
⎟
⎟
⎟
(41)
where
ξ
denotes the subscript ≤⋯≤rrn
(
1<<
)
k1.
The proof is omitted because of the similarity with Theorem 9.
5|THE PROPOUNDED MAGDM APPROACH
This part gives the depiction of traditional MAGDM issue and designs an innovative MAGDM
algorithm based upon the propounded Cq‐ROF2TLWMSM operators to cope with actual de-
cision issues.
This section shall establish a MAGDM approach on the basis of the propounded
Cq‐ROF2TLMSM operator under the Cq‐ROF2TL setting. For a classical MAGDM issues: suppose
ϒ={ϒ,ϒ,…,ϒ}
m12 be a family of alternatives,
={ , ,…, }
n12
be a family of attributes,
and the associated weights of attributes are
w
ww w={ , ,…, }
tn
T
12
with
∑
w=
1
t
nt
=1 .Supposethe
set of evaluators is
={ , ,…, }
p12 , and the weights of evaluators are
ω
ωω ω={ , ,…, }
tp
T
12
with
∑
ω=
1
ς
pς
=1 . The evaluator
ςp(=1,2,…,)
ςprovides his (her) assessment information for
alternative jmϒ(=1,2,…, )
jwith respect to the attribute tn(=1,2,…, )
tby the form of
Cq‐ROF2TLN, which is denoted as ab
ab
()
()
()
see=,ϱ,,
jt
ςjt
ς
jt
ςjt
ςiπjt
ςiπ
22
jt
ς
jt
ς. The decision
matrices are indicated as
()
Z
=
ςjt
ς
mn×. In order to attain the satisfies alternative, we build up a
novel MAGDM method to sort the alternatives based upon the presented Cq‐ROF2TLMSM
operator. The detailed procedure includes the following steps:
RONG ET AL.
|
27
Step 1: Standardize the DM matrices by the following transformation manner;
ab
ba
ab
ba
()
()
()
()
()
()
see
see
=
,ϱ, , , is benefit attribute ;
,ϱ, , , , is cost attribute .
jt
ςjt
ς
jt
ςjt
ςiπjt
ςiπt
jt
ς
jt
ςjt
ςiπjt
ςiπt
22
22
jt
ς
jt
ς
jt
ς
jt
ς
⎧
⎨
⎪
⎩
⎪
(42)
Step 2: Utilize the Cq‐ROF2TLWMSM operator or Cq‐ROF2TLWDMSM to integrate all
individual DM matrices
()
Z
ςp=(=1,2,…,)
ςjt
ς
mn×into a single DM matrix
Z
=( )
jt m n×:
(
)
Cq ROF TLWMSM=−2,,…,
jt jt jt jt
p
12 (43)
or
()
Cq ROF TLWDMSM=−2,,…,
jt jt jt jt
p
12
(44)
Step 3: Utilize the Cq‐ROF2TLWMSM operator or Cq‐ROF2TLWDMSM to fuse linguistic
evaluation information
tn(=1,2,…, )
jt into the comprehensive assessment value of alter-
natives jmϒ(=1,2,…, )
j:
Cq ROF TLWMSM=−2(,,…,
)
jjtjtjt
(45)
or
Cq ROF TLWDMSM=−2(,,…,)
.
jjtjtjt
(46)
Step 4: Compute the score function
S
˜(
)
jof each assessment value jmϒ(=1,2,…, )
j
through Definition 11, if the score values of diverse alternatives are same, we further calculate
the accuracy value of them;
Step 5: Sort jmϒ(=1,2,…, )
jin light of the comparison approach in Definition 12 and
choice the optimal alternative(s).
Step 6: End.
6|NUMERICAL EXAMPLE
In this section, a practical example is provided to verify the applicability of the designed
approach. Then, the sensitiveness of the
k
and
q
is studied. Also, the comparison analysing is
implemented to show the merits of the developed method.
Example 1. Emergency management is a proper term for dealing with the risk of
major accidents and disasters, which refers to the government and other public
institutions in the process of prevention, response, disposal and recovery of emergencies,
through the establishment of necessary response mechanism, take a series of necessary
measures, the use of science, technology, planning and management means, to ensure
the safety of emergencies related activities public life, health and property, and
promoting the harmonious and healthy development of society. In recent years, the
28
|
RONG ET AL.
frequent occurrence of natural disasters has caused great harm and loss to human life
and global economy. In order to effectively reduce the losses caused by major accidents
and disasters, the emergency management center will formulate different emergency
alternatives according to the types of accidents and invite experts in various fields to evaluate
the alternative emergency plans. The evaluation of emergency alternative is an important
process in emergency management. Its core is the classic DM problem, which has received a
multitude of attention from a large number of scholars. Therefore, in this part, we will apply
the propound approach to handle the evaluation problems of selecting the optimal
emergency alternative for the emergency management centre. After a series of screening,
four better alternatives will conduct further assessment. The four alternatives
{ϒ,ϒ,ϒ,ϒ}
123 4
will be assessed through three evaluators
{, , }
123
utilizing the linguistic term
sssss= { = very bad, = bad, = fair, = good, =very good }
01234
.Thefollowingfour
attributes are taken into consideration for valid modeling the characteristic of alternatives
by discussing from experts, which are displayed as:
:
1preparation ability, :
2rescuing
ability,
:
3restore ability, and :
4reaction capacity. Suppose the weight vector of
{, , , }
123 4
is
w
= {0.22, 0.30, 0.20, 0.28}
Tand the weight vector of
{, , }
123
is
ω
= {0.40, 0.35, 0.25}
T
. The individual decision matrices are provided by
{, , }
123
in
accordance to their knowledge level and viewpoint, which are displayed as Tables 1,2,and3.
6.1 |DM procedure
Step 1: The standardization process is omitted because all attributes are the benefit type.
Step 2: Utilize the Cq‐ROF2TLWMSM operator or Cq‐ROF2TLWDMSM to integrate all
individual DM matrices into a single DM matrix, which are displayed in Tables 4and 5
(set k=
2
and
q
=3
).
Step 3: Utilize the Cq‐ROF2TLWMSM operator or Cq‐ROF2TLWDMSM to fuse linguistic
evaluation information
tn(=1,2,…, )
jt in Tables 4or 5and attain the comprehensive as-
sessment value of alternatives jmϒ(=1,2,…, )
j, which is displayed in Table 6(set k=
2
and
q
=3
).
Step 4: Calculate score values
S
˜(
)
jof each alternative jmϒ(=1,2,…, )
jthrough Definition
11, which are listed in Table 7.
TABLE 1 The individual decision making matrix
Z1
provided by
1
12
3
4
ϒ
1se
e
(,0),0.30 ,
0.20
iπ
iπ
3
2 (0.50)
2 (0.40)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.25 ,
0.35
iπ
iπ
4
2 (0.56)
2 (0.40)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.20 ,
0.45
iπ
iπ
2
2 (0.70)
2 (0.28)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.25 ,
0.15
iπ
iπ
4
2 (0.65)
2 (0.20)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
2
se
e
(,0),0.45 ,
0.28
iπ
iπ
2
2 (0.55)
2 (0.42)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.26 ,
0.32
iπ
iπ
3
2 (0.46)
2 (0.36)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.31 ,
0.43
iπ
iπ
3
2 (0.31)
2 (0.48)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.30 ,
0.18
iπ
iπ
4
2 (0.58)
2 (0.32)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
3
se
e
(,0),0.40 ,
0.28
iπ
iπ
4
2 (0.29)
2 (0.22)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.29 ,
0.35
iπ
iπ
1
2 (0.64)
2 (0.29)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.35 ,
0.51
iπ
iπ
3
2 (0.41)
2 (0.28)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.37 ,
0.19
iπ
iπ
2
2 (0.39)
2 (0.42)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
4
se
e
(,0),0.43 ,
0.20
iπ
iπ
2
2 (0.65)
2 (0.12)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.29 ,
0.30
iπ
iπ
3
2 (0.56)
2 (0.29)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.38 ,
0.44
iπ
iπ
4
2 (0.61)
2 (0.33)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.33 ,
0.10
iπ
iπ
3
2 (0.48)
2 (0.39)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
RONG ET AL.
|
29
Step 5: Sort the alternatives jmϒ(=1,2,…, )
jand obtain the optimal alternative, which is
displayed in Table 8.
6.2 |Sensitivity analysis
In the above‐mentioned computations, the parameters have a pivotal role for the decision
procedure and strategy affects the final decision result. Therefore, we shall conduct a parameter
analysing for the parameter
k
and
q
in this subsection. For convenience, we utilize
Cq‐ROF2TLWMSM operator to resolve the mentioned example to achieve the analysis of this
subsection. To reveal the sensitivity of parameters
k
and
q
for the ultimate decision outcomes,
we utilize the propounded Cq‐ROF2TLWMSM operator with dissimilar parameter values to
cope with the aforementioned actual example and attain the ranking results, which are re-
spectively listed in Table 9(when
q
=3
) and Table 10 (when k=
2
).
From Table 9, we can discover that the sorting outcomes of alternatives are slightly diverse
when the parameter
k
is taken dissimilar values through the risk preference of decision makers.
TABLE 2 The individual decision making matrix
Z2
provided by
2
12
3
4
ϒ
1se
e
(,0),0.20 ,
0.26
iπ
iπ
3
2 (0.80)
2 (0.17)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.15 ,
0.42
iπ
iπ
3
2 (0.55)
2 (0.35)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.20 ,
0.69
iπ
iπ
2
2 (0.37)
2 (0.48)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.22 ,
0.24
iπ
iπ
4
2 (0.56)
2 (0.39)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
2
se
e
(,0),0.37 ,
0.39
iπ
iπ
2
2 (0.34)
2 (0.28)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.26 ,
0.24
iπ
iπ
3
2 (0.38)
2 (0.19)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.43 ,
0.32
iπ
iπ
1
2 (0.19)
2 (0.47)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.20 ,
0.17
iπ
iπ
4
2 (0.47)
2 (0.28)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
3
se
e
(,0),0.10 ,
0.37
iπ
iπ
3
2 (0.27)
2 (0.38)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.36 ,
0.46
iπ
iπ
1
2 (0.28)
2 (0.55)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.41 ,
0.37
iπ
iπ
4
2 (0.46)
2 (0.29)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.28 ,
0.36
iπ
iπ
3
2 (0.59)
2 (0.37)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
4
se
e
(,0),0.39 ,
0.18
iπ
iπ
4
2 (0.64)
2 (0.25)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.17 ,
0.27
iπ
iπ
3
2 (0.60)
2 (0.34)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.18 ,
0.55
iπ
iπ
1
2 (0.49)
2 (0.38)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.24 ,
0.57
iπ
iπ
2
2 (0.45)
2 (0.41)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
TABLE 3 The individual decision making matrix
Z
3provided by
3
12
3
4
ϒ
1se
e
(,0),0.61 ,
0.33
iπ
iπ
1
2 (0.40)
2 (0.38)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.37 ,
0.24
iπ
iπ
3
2 (0.65)
2 (0.22)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.29 ,
0.59
iπ
iπ
2
2 (0.57)
2 (0.32)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.22 ,
0.21
iπ
iπ
3
2 (0.42)
2 (0.39)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
2
se
e
(,0),0.47 ,
0.36
iπ
iπ
2
2 (0.36)
2 (0.27)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.32 ,
0.23
iπ
iπ
2
2 (0.38)
2 (0.40)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.17 ,
0.30
iπ
iπ
1
2 (0.58)
2 (0.44)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.29 ,
0.15
iπ
iπ
4
2 (0.49)
2 (0.46)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
3
se
e
(,0),0.37 ,
0.30
iπ
iπ
3
2 (0.36)
2 (0.39)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.28 ,
0.45
iπ
iπ
4
2 (0.29)
2 (0.45)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.21 ,
0.43
iπ
iπ
4
2 (0.57)
2 (0.31)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.19 ,
0.27
iπ
iπ
2
2 (0.11)
2 (0.74)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
4
se
e
(,0),0.22 ,
0.10
iπ
iπ
4
2 (0.64)
2 (0.25)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.36 ,
0.29
iπ
iπ
3
2 (0.19)
2 (0.56)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.29 ,
0.56
iπ
iπ
3
2 (0.53)
2 (0.83)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.38 ,
0.57
iπ
iπ
2
2 (0.56)
2 (0.26)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
30
|
RONG ET AL.
TABLE 4 Collective decision making matrix by utilizing Cq‐ROF2TLWMSM operator
123
4
ϒ
1
se
e
( , 1.3292), 0.6949 ,
0.1822
iπ
iπ
1
2 (0.8289)
2 (0.2353)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.4964), 0.6626 ,
0.5473
iπ
iπ
2
2 (0.8354)
2 (0.2404)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.2605), 0.6119 ,
0.4156
iπ
iπ
1
2 (0.8194)
2 (0.2584)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,−0.4446), 0.6129 ,
0.1399
iπ
iπ
2
2 (0.8399)
2 (0.2333)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
2
se
e
( , 0.2605), 0.7548 ,
0.2393
iπ
iπ
1
2 (0.7505)
2 (0.2345)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.3990), 0.6551 ,
0.1879
iπ
iπ
1
2 (0.7441)
2 (0.2297)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.1696), 0.6718 ,
0.2515
iπ
iπ
1
2 (0.6909)
2 (0.3266)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,−0.4096), 0.6728 ,
0.1171
iπ
iπ
2
2 (0.7907)
2 (0.2464)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
3
se
e
(,−0.4964), 0.6544 ,
0.2220
iπ
iπ
2
2 (0.6739)
2 (0.2329)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.1297), 0.6802 ,
0.2929
iπ
iπ
1
2 (0.7403)
2 (0.3092)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,−0.4711), 0.6891 ,
0.3107
iπ
iπ
1
2 (0.7795)
2 (0.2026)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.3234), 0.6334 ,
0.1947
iπ
iπ
1
2 (0.6821)
2 (0.3706)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
4
se
e
( , 0.4499), 0.7060 ,
0.1201
iπ
iπ
1
2 (0.8643)
2 (0.1478)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.4440), 0.6440 ,
0.1993
iπ
iπ
1
2 (0.7672)
2 (0.2795)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.3352), 0.6575 ,
0.3626
iπ
iπ
1
2 (0.8183)
2 (0.3868)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.3341), 0.7059 ,
0.3353
iπ
iπ
1
2 (0.8138)
2 (0.2583)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
RONG ET AL.
|
31
TABLE 5 Collective decision making matrix by utilizing Cq‐ROF2TLWDMSM operator
123
4
ϒ
1se
e
(,−0.1907), 0.2726 ,
0.6397
iπ
iπ
1
2 (0.4379)
2 (0.6782)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.1190), 0.1849 ,
0.6989
iπ
iπ
1
2 (0.4141)
2 (0.6902)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,−0.3362), 0.1585 ,
0.8045
iπ
iπ
1
2 (0.4076)
2 (0.7215)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.2374), 0.1608 ,
0.5886
iπ
iπ
1
2 (0.4031)
2 (0.6868)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
2
se
e
(,−0.3362), 0.3009 ,
0.7014
iπ
iπ
1
2 (0.3046)
2 (0.6901)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,−0.0951), 0.1928 ,
0.6450
iπ
iπ
1
2 (0.2880)
2 (0.6773)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,−0.4476), 0.2330 ,
0.7141
iπ
iπ
1
2 (0.2655)
2 (0.7608)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.3276), 0.1856 ,
0.5556
iπ
iπ
1
2 (0.3663)
2 (0.7017)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
3
se
e
( , 0.1190), 0.2289 ,
0.6845
iπ
iπ
1
2 (0.2117)
2 (0.6895)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,−0.4383), 0.2183 ,
0.7489
iπ
iπ
1
2 (0.3099)
2 (0.7539)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
( , 0.1972), 0.2417 ,
0.7513
iπ
iπ
1
2 (0.3347)
2 (0.6757)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,−0.2247), 0.2087 ,
0.6509
iπ
iπ
1
2 (0.3077)
2 (0.7841)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
4
se
e
( , 0.0591), 0.2614 ,
0.5456
iπ
iπ
1
2 (0.4608)
2 (0.5925)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,−0.0043), 0.1960 ,
0.6623
iπ
iπ
1
2 (0.3693)
2 (0.7244)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,−0.1427), 0.2076 ,
0.7520
iπ
iπ
1
2 (0.3883)
2 (0.7315)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,−0.2146), 0.2220 ,
0.7382
iπ
iπ
1
2 (0.3471)
2 (0.7109)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
32
|
RONG ET AL.
The dissimilar parameter values
k
reflect the interrelationship of different attributes during the
decision making process. For instance, when we take k=
1
in Cq‐ROF2TLWMSM operator, the
order relation of alternatives is ϒ>ϒ>ϒ>ϒ
4123
, which is different from other situations.
Because the Cq‐ROF2TLWMSM operator shall transform into Cq‐ROF2TLWA operator when
the parameter is assigned k=
1
, the correlation of discussed attributes fail to consider in the
course of dealing decision issues. When decision issues need to take consideration the
TABLE 9 The score value and order relation of alternatives based upon diverse
k
values
Score value of alternative
Parameter value
ϒ
1
ϒ
2
ϒ
3
ϒ4
Sorting Optimal selection
k
=
1
0.6021 0.5471 0.4973 0.6111 ϒ>ϒ>ϒ>ϒ
412
3
ϒ
4
k
=
2
0.8752 0.8474 0.8268 0.8720 ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
k
=
3
1.0839 1.0668 1.0465 1.0727 ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
TABLE 6 The integrated assessment information by utilizing Cq‐ROF2TLWMSM and Cq‐ROF2TLWDMSM
operator
Alternative Cq‐ROF2TLWMSM operator Cq‐ROF2TLWDMSM operator
ϒ
1
se
e
( , 0.1098), 0.8935 ,
0.1606
iπ
iπ
1
2 (0.9552)
2 (0.1521)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.2419), 0.9082 ,
0.1241
iπ
iπ
0
2 (0.9130)
2 (0.2637)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
2
se
e
( , 0.0908), 0.9109 ,
0.1275
iπ
iπ
1
2 (0.9300)
2 (0.1632)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.2164), 0.8985 ,
0.1439
iπ
iπ
0
2 (0.9169)
2 (0.1958)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
3
se
e
( , 0.0972), 0.9031 ,
0.1628
iπ
iπ
1
2 (0.9209)
2 (0.1839)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.2171), 0.9179 ,
0.1408
iπ
iπ
0
2 (0.9248)
2 (0.1872)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
4
se
e
( , 0.0927), 0.9080 ,
0.1701
iπ
iπ
1
2 (0.9502)
2 (0.1767)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
( , 0.2293), 0.9070 ,
0.1398
iπ
iπ
0
2 (0.9124)
2 (0.2479)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
TABLE 7 The score values of alternatives
Operator
ϒ
1
ϒ
2
ϒ
3
ϒ4
Cq‐ROF2TLWMSM operator
S
˜( ) = 0.875
2
1
S
˜( ) = 0.847
4
2
S
˜( ) = 0.826
8
3
S
˜( ) = 0.8720
4
Cq‐ROF2TLWDMSM operator
S
˜( ) = 0.180
2
1
S
˜( ) = 0.160
7
2
S
˜( ) = 0.168
8
3
S
˜( ) = 0.170
6
4
TABLE 8 The order relation of alternatives
Operator Ranking relation of j
ϒ
( = 1, 2, 3, 4)
jOptimal selection
Cq‐ROF2TLWMSM operator ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
Cq‐ROF2TLWMSM operator ϒ>ϒ>ϒ>ϒ
143
2
ϒ
1
RONG ET AL.
|
33
interconnection among any input data in assessment process, evaluators can take k=
2
or
k=3
in Cq‐ROF2TLWMSM operator to aggregation assessment information. For different
parameter values of
k
, DMs can choose an appropriate parameter value through their pre-
ference attitude and further attain a reasonable and satisfied decision result.
From Table 10, it is obvious that the orders of alternatives are the same for diverse para-
meter
q
by utilizing Cq‐ROF2TLWMSM operator, which testify the decision procedure is stable.
The values of
q
stand for the space of the assessment information for DMs. As the parameter
q
increases, DMs can provide more evaluation information according to their preference.
Moreover, we can easily acquire that the score value of alternatives on the basis of
Cq‐ROF2TLWMSM operator gets smaller and smaller with the value of
q
increases.
6.3 |Comparison analysis
6.3.1 |Verification of validity
To elaborate the validity and practicability of the created method in this essay, we conduct a
collection of comparative analysing with other previous decision methodologies including the
method based upon complex intuitionistic fuzzy weighted averaging (CIFWA) operator pro-
posed by Garg,
27
the method based upon complex intuitionistic BM (CIFBM) operator proposed
by Garg
28
and the method based upon complex q‐rung orthopair fyzzy linguistic Heronian
mean (Cq‐ROFLHM) operator proposed by Liu et al.
39
We utilize these methods to cope with
the Example 1in this paper, the score values and ranking of alternatives are displayed in
Table 11. From it, we can attain the same sorting results of alternatives based on the previous
methods and the designed method in this article, which can verify the availability of the
propounded method.
6.3.2 |Generalization analysis
In the next, a numerical case is utilized to expound the generalization of the presented method.
The evaluation information provided by DMs in respect of complex q‐rung orthopair linguistic
numbers, which are displayed in Table 12, and the weights of attributes are the same as with
Example 1. Then the case is settled through our method in this article and the decision results
are displayed in Table 13.
TABLE 10 The score value and order relation of alternatives based upon diverse
q
values
Score value of alternative
Parameter value
ϒ
1
ϒ
2
ϒ
3
ϒ4
Sorting Optimal selection
q
=
1
0.8764 0.8483 0.8275 0.8750 ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
q
=
2
0.8762 0.8481 0.8269 0.87430 ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
q
=3
0.8736 0.8474 0.8268 0.8720 ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
q
=
4
0.8752 0.8464 0.8248 0.8661 ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
q
=5
0.8704 0.8440 0.8208 0.8588 ϒ>ϒ>ϒ>ϒ
142
3
ϒ
1
34
|
RONG ET AL.
TABLE 11 The score value and order relation of alternatives based upon diverse
q
values
Score value of alternative
Approaches
ϒ
1
ϒ
2
ϒ
3
ϒ4
Sorting
CIFWA operator presented in Garg and Rani
27
0.7027 0.6900 0.6663 0.7023 ϒ>ϒ>ϒ>ϒ
1423
CIFBM operator presented in Garg and Rani
28
(pq==
1
) 0.8752 0.8474 0.8268 0.8720 ϒ>ϒ>ϒ>ϒ
1423
Cq‐ROFLHM operator presented in Liu et al.
39
(
st q==1, =3
) 0.3848 0.3509 0.3280 0.3577 ϒ>ϒ>ϒ>ϒ
1423
Cq‐ROF2TLMSM operator presented in this essay (
q
=
1
) 0.8764 0.8483 0.8275 0.8750 ϒ>ϒ>ϒ>ϒ
1423
Cq‐ROF2TLMSM operator presented in this essay (
q
=3
) 0.8736 0.8474 0.8268 0.8720 ϒ>ϒ>ϒ>ϒ
1423
Cq‐ROF2TLMSM operator presented in this essay (
q
=5
) 0.8704 0.8440 0.8208 0.8588 ϒ>ϒ>ϒ>ϒ
1423
RONG ET AL.
|
35
From Table 13, we can find that the previous approaches like CIFWA operator,
27
CIFBM operator,
28
distance measure
35
can not address the Example 2, but the present
method can effectively address it. That proves the presented method is more universal
than previous approaches because the CIFS and CPFS are the special examples of our
propounded method.
6.3.3 |Further contrastive analysing
In the next, we will administer a detailed contrast between the previous works with the
designed method, the significant supremacies of our propounded method are summarized as
below.
✓Compare with the approach based upon CIFWA operator propounded by Garg and
Rani.
27
The CIFWA operator as a fundamental aggregation technique to integrate com-
plex intuitionistic fuzzy information assumes that the considered attributes in real‐life
problems are unrelated, that is, it deems the relevance of attributes, which the decision
result ambiguous and unreasonable. The Cq‐ROF2TLMSM operator can valid conquer
the aforementioned defect and take into consideration the relationship of attributes.
TABLE 13 The score value and order relation of alternatives based upon diverse
q
values
Approaches Score value of alternative Sorting
CIFWA operator presented in Garg &
Rani
27
Cannot be computed Cannot be computed
CIFBM operator presented in Garg &
Rani
28
(pq==
1
)
Cannot be computed Cannot be computed
Distance measure on CPFS presented in
Ullah et al.
35
Cannot be computed Cannot be computed
Cq‐ROF2TLMSM operator presented in
this essay (
q
=3
)
S
S
˜(ϒ) = 0.8924, ˜(ϒ) = 0.7290
,
12
S
S
˜(ϒ) = 0.7036, ˜(ϒ) = 0.622
8
34
ϒ>ϒ>ϒ>ϒ
123
4
TABLE 12 The decision making matrix from Example 2
12
3
4
ϒ
1se
e
(,0),0.70 ,
0.36
iπ
iπ
3
2 (0.80)
2 (0.95)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.55 ,
0.62
iπ
iπ
4
2 (0.55)
2 (0.35)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.89 ,
0.69
iπ
iπ
3
2 (0.37)
2 (0.48)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.78 ,
0.84
iπ
iπ
4
2 (0.56)
2 (0.39)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
2
se
e
(,0),0.67 ,
0.69
iπ
iπ
3
2 (0.34)
2 (0.28)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.56 ,
0.59
iπ
iπ
3
2 (0.38)
2 (0.85)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.69 ,
0.78
iπ
iπ
3
2 (0.59)
2 (0.47)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.87 ,
0.97
iπ
iπ
2
2 (0.47)
2 (0.73)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
3
se
e
(,0),0.90 ,
0.96
iπ
iπ
4
2 (0.27)
2 (0.67)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.68 ,
0.85
iπ
iπ
3
2 (0.28)
2 (0.79)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.65 ,
0.76
iπ
iπ
4
2 (0.46)
2 (0.488)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.81 ,
0.77
iπ
iπ
3
2 (0.59)
2 (0.37)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
ϒ
4
se
e
(,0),0.67 ,
0.78
iπ
iπ
3
2 (0.64)
2 (0.25)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.57 ,
0.48
iπ
iπ
2
2 (0.60)
2 (0.88)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟se
e
(,0),0.33 ,
0.66
iπ
iπ
3
2 (0.49)
2 (0.38)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
se
e
(,0),0.24 ,
0.87
iπ
iπ
3
2 (0.45)
2 (0.81)
⎛
⎝
⎜⎛
⎝
⎜⎞
⎠
⎟
⎞
⎠
⎟
36
|
RONG ET AL.
Additionally, it can reflect DM's individual favourites and show the dynamic trend of
the order relation of alternatives through the adjustable parameter. Accordingly, the
Cq‐ROF2TLMSM operator are more efficient and general to process decision analysis
problems.
✓Compare with the approach based on CIFBM operator proffered via Garg and Rani.
28
Although CIFBM operator can be utilized to aggregation complex intuitionistic fuzzy in-
formation, it only takes into consideration the correlation between any two attributes. The
Cq‐ROF2TLMSM operator proffered in this essay not only catches the correlation among
multiple attributes, but also reduces the computational complexity during information
aggregation process. Furthermore, our propounded method can address actual issues from a
qualitative point of view and it settles several problems that CIFBM operator can not
resolve. Hence, the presented methodology is more universal and realistic to attain a ra-
tional decision result.
✓Compared with the approach on the basis of Cq‐ROFLHM operator proposed by
Liu et al.,
39
the following three aspects are given highlight the difference between
Cq‐ROFLHM and Cq‐ROF2TLMSM operator. (1) For linguistic representation: the
integration information of LV may not be matched by appropriate linguistic terms. For
instance, as= ( , (0.6, 0.8)
)
1 1.432 . Aiming at this situation, the virtual linguistic term s1.43
2
is
only utilized to comparative and computation process, but does not have any semantics to
match with it. That will lead to the information loss in information fusion process.
Nevertheless, the 2‐tuple linguistic representation model can valid prevent information
loss because the linguistic terms in 2‐tuple linguistic are continuous. (2) For the corre-
lation of attributes: the HM operator can only take into the interrelationship between any
two attributes, which will produce a irrational decision result. But the MSM operator can
consider the correlation among multiple input attributes in the procedure of information
integration. (3) For the computational‐complexity: the HM operator has two parameters,
which make the computational process become more difficult, and it is arduous for DMs
to determinate two satisfied parameter values. However, the MSM operator only has one
parameter, which is more convenient for DMs to allocate appropriate parameter
value according to actual needs and their favourites. To sum up, the designed approach
based upon Cq‐ROF2TLMSM operator is more universal and flexible than the
Cq‐ROFLHM operator.
From the above comprehensive comparative analysis, we summarize the marked char-
acteristics between the propounded method with other existing approaches, which are dis-
played in Table 14. From it, we can attain that the current methods like CIFs, CPFS and
Cq‐ROFLS are the particular cases OF Cq‐ROF2TLS. The propounded approach on the basis of
Cq‐ROF2TLMSM operators is more powerful than other previous methods to fuse fuzzy in-
formation. Consequently, the merits of the developed operators are outlined as below: (a) the
proposed operators can efficiently seize the interrelationship of multiple input arguments,
which will reduce information loss during the real‐life decision procedure and attain more
rational decision outcome; (b) the presented operators can ponder the correlation between
different quantitative parameters through assigning diverse parameter, which greatly increases
the flexibility of decision process; (c) the developed can degenerate into several extant operators
via taking dissimilar values of parameter, which signifies that the propounded operators are
more generalized and suitable to dispose actual decision issues.
RONG ET AL.
|
37
TABLE 14 Characteristic comparison with existing approaches
Approaches
Capability
to fuse
information
Capture correlation
between two
attributes
Capture correlation
among multiple
attributes
Capability to tackle
two‐dimensional
information
Capability to express
information by
complex numbers
Flexibility of
decision
procedure
The method presented in
Rani and Garg
23
✓✓
×
✓✓
×
The method presented in
Garg and Rani
27
✓
××
✓✓
×
The method presented
in Garg and Rani
28
✓✓
×
✓✓ ✓
The method presented in
Ullah et al.
35
×
××
✓✓
×
The method presented in
Liu
41
✓
××
×
×
×
The method presented in
Liu et al.
39
✓✓
×
✓✓ ✓
The method presented in
Li et al.
53
✓✓
×
×
×
✓
The presented
method (
q
=
1
)
✓✓ ✓ ✓ ✓ ✓
The presented
method (
q
=
2
)
✓✓ ✓ ✓ ✓ ✓
The presented
method (
q
=3
)
✓✓ ✓ ✓ ✓ ✓
38
|
RONG ET AL.
7|CONCLUSION
The Cq‐ROFS and 2‐tuple LV are two efficient models that can not only portray the complex
vague assessment information but also reduce information loss in MAGDM issues. Inspired by
the mentioned notion, we first propounded a novel conception called Cq‐ROF2TLS to express
ill‐defined and uncertain evaluation information in actual problems, which synthetically
consider the merits of Cq‐ROFS and 2‐tuple LV. Moreover, we present several aggregation
operators including Cq‐ROF2TLWA, Cq‐ROF2TLWG, Cq‐ROF2TLMSM, Cq‐ROF2TLWMSM,
Cq‐ROF2TLDMSM and Cq‐ROF2TWDLMSM operator to integrate Cq‐ROF2TL information
and explore several characteristics of them at length. Furthermore, we designed a new
MAGDM method based upon the Cq‐ROF2TLWMSM and Cq‐ROF2TLWDMSM operator, as
well as a numerical case that is used to show the efficiency and feasibility of the novel
approach. In the end, a comparative analysis between the extant methods and our approach is
administered to highlight the superiorities of the designed approach. Although this research
originally develops a novel conception to describe ambiguous and ill‐defined information and
expands its theory and application in real‐life, the following two defects of this research need to
further explore. It's easy to see that the different parameter can cause diverse rankings, thus the
first problem is how to determine the most suitable value of parameter for the presented
algorithm. The other is to use appropriate methods for attaining the weight of attribute
and expert. The presented approach assumes that the weights of attributes and experts
are provided by decision specialists in advance. For the expert weight, it only considers the
subjectively and ignores the weight information produced from the decision matrix. Thus, the
associative weight method to ascertain weight of attribute should be explored for handling
actual decision problems.
In future research, we shall apply the presented method to deal with several real‐life
problems, for instance, site selection for wind power plants,
54
assessment of the innovative
ability of universities
55
and so forth. Meanwhile, we will continue to study other theories of
CFS such ac fuzzy logic, basic operational, decision analysis and so forth. In addition, the
spherical fuzzy set,
56‐58
as a generalization of picture fuzzy set is originated to process un-
certainty and fuzziness recently. Hence, the research on spherical fuzzy set will become a
momentous key‐point in the following stage.
ACKNOWLEDGMENTS
This work was supported by the National Natural Science Foundation of China under Grant
61372187, the Scientific and Technological Project of Sichuan Province under Grant
2019YFG0100. the Sichuan Province Youth Science and Technology Innovation Team under
Grant 2019JDTD0015, the Application Basic Research Plan Project of Sichuan Province under
Grant 2017JY0199, the Scientific Research Project of Department of Education of Sichuan
Province under Grant 18ZA0273 and Grant 15TD0027, the Scientific Research Project of Nei-
jiang Normal University under Grant 18TD08, the Innovation Fund of Postgraduate Xihua
University under Grant YCJJ2020028, the University Students Innovation and Entrepreneur-
ship Project of Xihua Cup under Grant 2020107. The author would like to thank the editors and
anonymous reviewers for their constructive comments and suggestions, which will help us to
better improve this paper. The author (Yuan Rong) would like to special thank the radio
management technology research center of Xihua University for its great support during the
preparation of the paper.
RONG ET AL.
|
39
CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.
AUTHOR CONTRIBUTIONS
This paper is a result of the common work of the authors in all aspects. All authors read and
agreed to the published version of the manuscript.
DATA AVAILABILITY STATEMENT
The data to sustain the application of this investigation are included within the essay.
REFERENCES
1. Ashraf S, Abdullah S. Spherical aggregation operators and their application in multiattribute group
decision‐making. Int J Intell Syst. 2019;34(3):493‐523.
2. Ashraf S, Abdullah S, Mahmood T, Ghani F, Mahmood T. Spherical fuzzy sets and their applications in
multi‐attribute decision making problems. J Intell Fuzzy Syst. 2019;36(3):2829‐2844.
3. Rong Y, Pei Z, Liu Y. Hesitant fuzzy linguistic hamy mean aggregation operators and their application to
linguistic multiple attribute decision‐making. Math Probl Eng. 2020;2020:1‐22. https://doi.org/10.1155/
2020/3262618
4. Rong Y, Pei Z, Liu Y. Generalized single‐valued neutrosophic power aggregation operators based on
archimedean copula and co‐copula and their application to multi‐attribute decision‐making. IEEE Access.
2020;8:35496‐35519.
5. Jin Y, Ashraf S, Abdullah S. Spherical fuzzy logarithmic aggregation operators based on entropy and their
application in decision support systems. Entropy. 2019;21(7):628.
6. Zadeh LA. Fuzzy sets. Inform Control. 1965;8(3):338‐353.
7. Pei Z, Liu J, Hao F, Zhou B. FLM‐TOPSIS: the fuzzy linguistic multiset TOPSIS method and its application
in linguistic decision making. Inform Fusion. 2019;45:266‐281.
8. Kong M, Pei Z, Ren F, Hao F. New operations on generalized hesitant fuzzy linguistic term sets for
linguistic decision making. Int J Fuzzy Syst. 2019;21(1):243‐262.
9. Liu Y, Liu J, Qin Y. Pythagorean fuzzy linguistic Muirhead mean operators and their applications to
multiattribute decision‐making. Int J Intell Syst. 2020;35(2):300‐332.
10. Rong Y, Liu Y, Pei Z. Novel multiple attribute group decision‐making methods based on linguistic in-
tuitionistic fuzzy information. Mathematics. 2020;8(3):322.
11. Ashraf S, Abdullah S, Smarandache F, Amin NU. Logarithmic hybrid aggregation operators based on single
valued neutrosophic sets and their applications in decision support systems. Symmetry. 2019;11(3):364.
12. Ramot D, Milo R, Friedman M, Kandel A. Complex fuzzy sets. IEEE Trans Fuzzy Syst. 2002;10(2):171‐186.
13. D Ramot, Friedman M, Langholz G, Kandel A. Complex fuzzy logic. IEEE Trans Fuzzy Syst. 2003;11(4):450‐461.
14. Bi L, Dai S, Hu B. Complex fuzzy geometric aggregation operators. Symmetry. 2018;10(7):251.
15. Yazdanbakhsh O, Dick S. A systematic review of complex fuzzy sets and logic. Fuzzy Sets and Syst. 2018;
338:1‐22.
16. Atanassov KT. Intuitionistic fuzzy sets. Fuzzy Sets Syst. 1986;20(1):87‐96.
17. Liu Y, Liu J, Qin Y. Dynamic intuitionistic fuzzy multiattribute decision making based on evidential
reasoning and MDIFWG operator. J Intell Fuzzy Syst. 2019;36(6):5973‐5987.
18. He X, Li Y, Qin K, Meng D. Distance measures on intuitionistic fuzzy sets based on intuitionistic fuzzy
dissimilarity functions. Soft Comput. 2020;24(1):523‐541.
19. Yuan J, Luo X. Approach for multi‐attribute decision making based on novel intuitionistic fuzzy entropy
and evidential reasoning. Comput Ind Eng. 2019;135:643‐654.
20. Luo L, Zhang C, Liao H. Distance‐based intuitionistic multiplicative MULTIMOORA method integrating a novel
weight‐determining method for multiple criteria group decision making. Comput Ind Eng. 2019;131:82‐98.
21. Zhang C, Luo L, Liao H, Mardani A, Streimikiene D, Al‐Barakati A. A priority‐based intuitionistic mul-
tiplicative UTASTAR method and its application in low‐carbon tourism destination selection. Appl Soft
Comput. 2020;88:106026.
40
|
RONG ET AL.
22. Alkouri AMDJS, Salleh AR. Complex Intuitionistic Fuzzy Sets. In Proceedings of the International
Conference on Fundamental and Applied Sciences, Kuala Lumpur, Malaysia, 12‐14 June 2012; Volume 1482,
Chapter 2; 2012:464‐470. ISBN 9780735410947.
23. Rani D, Garg H. Complex intuitionistic fuzzy power aggregation operators and their applications in
multicriteria decision‐making. Expert Syst. 2018;35(6):e12325.
24. Atanassov K, Gargov G. Interval‐valued intuitionistic fuzzy sets. Fuzzy Sets Syst. 1989;31:343‐349.
25. Garg H, Rani D. Complex interval‐valued intuitionistic fuzzy sets and their aggregation operators. Fundam
Inform. 2019;164(1):61‐101.
26. Garg H, Rani D. Some results on information measures for complex intuitionistic fuzzy sets. Int J Intell Syst.
2019;34(10):2319‐2363.
27. Garg H, Rani D. New generalised Bonferroni mean aggregation operators of complex intuitionistic fuzzy
information based on Archimedean t‐norm and t‐conorm. J Exp Theor Artif Intell. 2020;32(1):81‐109.
28. Garg H, Rani D. Some generalized complex intuitionistic fuzzy aggregation operators and their application
to multicriteria decision‐making process. Arab J Sci Eng. 2019;44(3):2679‐2698.
29. Garg H, Rani D. Exponential, logarithmic and compensative generalized aggregation operators under
complex intuitionistic fuzzy environment. Group Decis Negot. 2019;28(5):991‐1050.
30. Garg H, Rani D. Novel aggregation operators and ranking method for complex intuitionistic fuzzy sets and
their applications to decision‐making process. Artif Intell Rev. 2019;53:3595‐3620. https://doi.org/10.1007/
s10462‐019‐09772‐x
31. Yager RR. Pythagorean membership grades in multicriteria decision making. IEEE Trans Fuzzy Syst. 2013;
22(4):958‐965.
32. Qin Y, Liu Y, Hong Z. Multicriteria decision making method based on generalized Pythagorean fuzzy
ordered weighted distance measures. J Intell Fuzzy Syst. 2017;33(6):3665‐3675.
33. Garg H. New logarithmic operational laws and their aggregation operators for Pythagorean fuzzy set and
their applications. Int J Intell Syst. 2019;34(1):82‐106.
34. Liang D, Xu Z, Liu D, Wu Y. Method for three‐way decisions using ideal TOPSIS solutions at Pythagorean
fuzzy information. Inform Sci. 2018;435:282‐295.
35. Ullah K, Mahmood T, Ali Z, Jan N. On some distance measures of complex Pythagorean fuzzy sets and
their applications in pattern recognition. Complex Intell Syst. 2020;6:15‐27.
36. Yager RR. Generalized orthopair fuzzy sets. IEEE Trans Fuzzy Syst. 2016;25(5):1222‐1230.
37. Liu P, Wang P. Multiple‐attribute decision‐making based on Archimedean Bonferroni Operators of q‐rung
orthopair fuzzy numbers. IEEE Trans Fuzzy Syst. 2018;27(5):834‐848.
38. Li Z, Wei G, Wang R, Wu J, Wei C, Wei Y. EDAS method for multiple attribute group decision making
under q‐rung orthopair fuzzy environment. Technol Econ Dev Eco. 2020;26(1):86‐102.
39. Liu P, Ali Z, Mahmood T. A method to multi‐attribute group decision‐making problem with complex
q‐rung orthopair linguistic information based on heronian mean operators. Int J Comput Int Syst. 2019;
12(2):1465‐1496.
40. Zadeh LA. The concept of a linguistic variable and its application to approximate reasoning‐I. Inform Sci.
1975;8(3):199‐249.
41. Liu P. Some generalized dependent aggregation operators with intuitionistic linguistic numbers and their
application to group decision making. J Comput Syst Sci. 2013;79(1):131‐143.
42. Wang J, Yang Y, Li L. Multi‐criteria decision‐making method based on single‐valued neutrosophic lin-
guistic Maclaurin symmetric mean operators. Neural Comput Appl. 2018;30(5):1529‐1547.
43. Liu P, Liu W. Multiple‐attribute group decision‐making based on power Bonferroni operators of linguistic
q‐rung orthopair fuzzy numbers. Int J Intell Syst. 2019;34(4):652‐689.
44. Herrera F, Martinez L. A 2‐tuple fuzzy linguistic representation model for computing with words. IEEE
Trans Fuzzy Syst. 2000;8(6):746‐752.
45. Beg I, Rashid T. An intuitionistic 2‐tuple linguistic information model and aggregation operators. Int J Intell
Syst. 2016;31(6):569‐592.
46. Deng X, Wei G, Gao H, Wang J. Models for safety assessment of construction project with some 2‐tuple
linguistic Pythagorean fuzzy Bonferroni mean operators. IEEE Access. 2018;6:52105‐52137.
47. Xu Z. Intuitionistic fuzzy aggregation operators. IEEE Trans on Fuzzy Syst. 2007;15(6):1179‐1187.
RONG ET AL.
|
41
48. Liu P, Wang P. Some q‐rung orthopair fuzzy aggregation operators and their applications to multiple‐
attribute decision making. Int J Intell Syst. 2018;33(2):259‐280.
49. Maclaurin C. A second letter to martin folkes, Esq.; concerning the roots of equations, with demonstration
of other rules of algebra. Philos Trans Roy Soc London Ser A. 1729;36:59‐96.
50. Qin J, Liu X. Approaches to uncertain linguistic multiple attribute decision making based on dual Ma-
claurin symmetric mean. J Intell Fuzzy Syst. 2015;29(1):171‐186.
51. Liu P, Qin X. Maclaurin symmetric mean operators of linguistic intuitionistic fuzzy numbers and their
application to multiple‐attribute decision‐making. J Exp Theor Artif Intell. 2017;29(6):1173‐1202.
52. Wei G, Lu M. Pythagorean fuzzy Maclaurin symmetric mean operators in multiple attribute decision
making. Int J Intell Syst. 2018;33(5):1043‐1070.
53. Li L, Zhang R, Shang X. Some q‐rung orthopair linguistic Heronian meanoperators with their application to
multi‐attribute group decision making. Arch of Control Sci. 2018;28(4):551‐583.
54. Wu X, Zhang C, Jiang L, Liao H. An integrated method with PROMETHEE and conflict analysis for
qualitative and quantitative decision‐making: Case study of site selection for wind power plants. Cogn
Comput. 2020;12:100‐114.
55. Dong L, Gu X, Wu X, Liao H. An improved MULTIMOORA method with combined weights and its
application in assessing the innovative ability of universities. Expert Syst. 2019;36(2):e12362.
56. Ashraf S, Mahmood T, Abdullah S, Khan Q. Different approaches to multi‐criteria group decision making
problems for picture fuzzy environment. Bull Braz Math Soc, New Series. 2019;50:373‐397.
57. Ashraf S, Abdullah S, Abdullah L. Child development influence environmental factors determined using
spherical fuzzy distance measures. Mathematics. 2019;7(8):661.
58. Jin H, Ashraf S, Abdullah S, Qiyas M, Bano M, Zeng S. Linguistic spherical fuzzy aggregation operators and
their applications in multi‐attribute decision making problems. Mathematics. 2019;7(5):413.
How to cite this article: Rong Y, Liu Y, Pei Z. Complex q‐rung orthopair fuzzy 2‐tuple
linguistic Maclaurin symmetric mean operators and its application to emergency
program selection. Int J Intell Syst. 2020;1‐42. https://doi.org/10.1002/int.22271
42
|
RONG ET AL.