Some Interval-Valued Neutrosophic Hesitant Fuzzy Uncertain Linguistic Bonferroni Mean Aggregation Operators and Their Application in Multiple Attribute Decision Making

Some interval-valued neutrosophic hesitant fuzzy uncertain linguistic
Bonferroni mean aggregation operators and their application in multiple

attribute decision making
Peide Liu a,b,*, Fei Tenga,
a,
School of Management Science and Engineering, Shandong University of Finance and
Economics, Jinan Shandong 250014, China
b
School of Economics and Management, Civil Aviation University of China, Tianjin 300300,
China
*The corresponding author: peide.liu@gmail.com, Tel: +86-531-82222188
Abstract: Interval-valued neutrosophic hesitant fuzzy uncertain linguistic set (IVNHFULS) has the
advantages of both neutrosophic numbers and interval-valued hesitant fuzzy uncertain linguistic
variable. In this paper, we firstly introduce the definition, the operational laws, and the score function
of IVNHFULS. Then, we combine interval-valued neutrosophic hesitant fuzzy uncertain linguistic set
with Bonferroni mean operator and propose some new aggregation operators, such as the
interval-valued neutrosophic hesitant fuzzy uncertain linguistic Bonferroni mean (IVNHFULBM)
operator, the interval-valued neutrosophic hesitant fuzzy uncertain linguistic weighted Bonferroni mean
(IVNHFULWBM) operator, the interval-valued neutrosophic hesitant fuzzy uncertain linguistic
geometric Bonferroni mean (IVNHFULGBM) operator, the interval-valued neutrosophic hesitant fuzzy
uncertain linguistic weighted geometric Bonferroni mean (IVNHFULWGBM) operator. At the same
time, the related properties of these operators are discussed. Furthermore, we propose two multiple
attribute decision making methods based on IVNHFULWBM operator and IVNHFULWGBM operator.
Finally, we give an illustrative example to demonstrate the practicality and effectiveness of the
proposed methods.
Keywords: Multiple attribute decision making; Interval-valued neutrosophic hesitant fuzzy uncertain
linguistic set; Bonferroni mean; Interval-valued neutrosophic hesitant fuzzy uncertain linguistic
Bonferroni mean aggregation operator.
1. Introduction
Multiple attribute decision making (MADM) has been a hot research topic. In the real decision
making, because of the complexity and the uncertainty of decision making problems, how to
realistically express the fuzzy and uncertain information is very important. Zadeh [1] firstly proposed
the Fuzzy set (FS), which has been successfully applied in various fields. Due to the FS only consider
the membership degree, Atanassov [2] further developed the intuitionistic fuzzy set (IFS) based on the
FS. In recent years, Smarandache [3] proposed the neutrosophic set (NS) which consisted of
membership degree, non-membership degree and indeterminacy -membership degree. In addition,
some subsets of neutrosophic set have been proposed. For example, Wang et al. [4][5] further proposed
the interval-valued neutrosophic set (IVNS) and the single-valued neutrosophic set (SVNS). Due to the
membership degree of FS, the membership degree and non-membership degree of IFS, the membership
degree and non-membership degree and indeterminacy-membership degree of NS are a single value
ranged from 0 to 1. As we all known, the single values are asymmetric and not comprehensive so that
they inappropriately describe the incomplete information. Based on this, Torra and Narukawa [6-7]
introduced the hesitant fuzzy sets (HFS). In HFS, the membership degree is a set of several possible
values between 0 and 1. The obvious difference between FS and HFS is that the membership degree is
1
a single value or a set of possible values ranged from 0 to 1. When the decision makers evaluate the
alternatives, HFS is a useful tool to describe the evaluation values with hesitance. Chen et al. [8]
extended HFS to interval-valued hesitant fuzzy sets (IVHFS), in which the membership degree is a set
of possible interval values between 0 and 1. Ye [9] presented a neutrosophic hesitant fuzzy set (NHFS),
which is the combination HFS with SVNS. In real decision making, especially for qualitative
information, the linguistic variables are suitable to express the evaluation values and it can improve the
flexibility and reliability of the decision making information. So far, the linguistic variables have been
widely studied [10-12], which could be expressed as single values [13] or interval values [14]. The
interval linguistic terms are also known as uncertain linguistic variables. Rodriguez et al. [15-16]
originally gave the hesitant fuzzy linguistic term sets (HFTLS) which is used to describe the linguistic
variables. Lin et al. [17] developed the hesitant fuzzy linguistic sets (HFLS) and the hesitant fuzzy
uncertain linguistic set (HFULS) on the basis of HFS and linguistic variables or uncertain linguistic
variables. The characteristics of the HFLS and HFULS are that the membership degree can be
expressed as a set of hesitant fuzzy linguistic variables or hesitant fuzzy uncertain linguistic variables.
Similar to the HFS, HFLS doesnt take the non-membership degree, indeterminacy-membership degree
into account so that HFLS has the weaknesses in avoiding the information distortion and losing
effectively in handling the indeterminate, incomplete and inconsistent problems. In order to overcome
the deficiencies, in this paper, we extend NHFS to interval-valued neutrosophic hesitant fuzzy set
(IVNHFS) and cobmine IVNHFS with uncertain linguistic variables. Consequently, we finally give the
concept of interval-valued neutrosophic hesitant fuzzy uncertain linguistic set (IVNHFULS), which is
composed of two main parts: an uncertain linguistic term and an interval-valued neutrosophic hesitant
fuzzy element. The two parts respectively express an evaluation value and the hesitancy for the given
evaluation value. Certainly, the second part also represents the interval-valued membership degree,
interval-valued non-membership degree and interval-valued indeterminacy-membership degree
associated with the specific uncertain linguistic term.
Up to now the information aggregation operators are studied deeply, and they have been widely
used in various field. All sorts of operators have been proposed to deal with many different situations.
For example, Xu [18-19] developed the uncertain linguistic weighted averaging (ULWA) operator, the
uncertain linguistic ordered weighted averaging (ULOWA) operator, the uncertain linguistic hybrid
averaging (ULHA) operator, the uncertain linguistic geometric averaging (ULGA) operator, uncertain
linguistic weighted geometric averaging (ULWGA) operator, uncertain linguistic ordered weighted
geometric averaging (ULOWGA) operator, and induced uncertain linguistic ordered weighted
geometric averaging (IULOWGA) operator under uncertain linguistic environment. Wei [20] proposed
an uncertain linguistic hybrid geometric averaging (ULHGA) operator. In order to more accurately and
flexibly deal with the hesitant fuzzy information, Lin et al. [17] developed some hesitant fuzzy
uncertain linguistic arithmetic aggregation operators, such as hesitant fuzzy uncertain linguistic
weighted averaging (HFULWA) operator, hesitant fuzzy uncertain linguistic ordered weighted
averaging (HFULOWA) operator, hesitant fuzzy uncertain linguistic hybrid averaging (HFULHA)
operator. Ju and Liu [21] gave the interval-valued hesitant uncertain linguistic weighted averaging
(IVHULWA) operator, the interval-valued hesitant uncertain linguistic weighted geometric averaging
(IVHULWGA) operator. Liu et al. [22] proposed some interval-valued hesitant uncertain linguistic
generalized aggregation operators, including the generalized interval-valued hesitant uncertain
linguistic weighted averaging (GIVHULWA) operator, the generalized interval-valued hesitant
uncertain linguistic ordered weighted averaging (GIVHULOWA) operator and the generalized
2
interval-valued hesitant uncertain linguistic hybrid averaging (GIVHULHA) operator. In order to better
handle the interrelationship between input arguments, Bonferroni [23] originally introduced the
Bonferroni Mean (BM) operator. The extensions of the BM operator have been constantly proposed
since it was initially introduced. Yager [24][25] and Beliakov et al. [26] proposed some generalized
BM operators. Zhu et al.[27] combined BM operator with geometric operator and finally proposed the
geometric Bonferroni mean (GBM) operator. In this paper, we will extend the BM operator to the
interval-valued neutrosophic hesitant fuzzy uncertain linguistic environment for the sake of possessing
the advantages of both BM operator and IVNHFULS, and flexibly handling the indeterminate,
incomplete, inconsistent information and effectively considering the interrelationship of the
interval-valued neutrosophic hesitant fuzzy uncertain linguistic information.
The remainder of this paper is shown as follows. In section 2, we mainly introduce the basic
concepts of the interval-valued neutrosophic hesitant fuzzy uncertain linguistic set (IVNHFULS), and
the operational rules and the characteristics of IVNHFULS. In section 3, we propose some
interval-valued neutrosophic hesitant fuzzy uncertain linguistic BM aggregation operators, such as the
operator, the interval-valued neutrosophic hesitant fuzzy uncertain linguistic weighted Bonferroni mean
(IVNHFULWBM) operator, the interval-valued neutrosophic hesitant fuzzy uncertain linguistic
geometric Bonferroni mean (IVNHFULGBM) operator, the interval-valued neutrosophic hesitant fuzzy
uncertain linguistic weighted geometric Bonferroni mean (IVNHFULWGBM) operator. In section 4,
two multiple attribute decision making methods based on IVNHFULWBM operator and
IVNHFULWGBM operator are proposed. In section 5, we give a numerical example to prove the
effectiveness of the proposed methods.
2. Preliminaries
2.1 The uncertain linguistic variables
Suppose that S ( s0 , s1 , , sl 1 ) consists of the finite and complete ordered elements, where l is
an odd value. In general, l is equal to 3,5,7,9, etc. when l 9 ,
S ( s0 , s1 , s2 , s3 , s4 , s5 , s6 , s7 , s8 ) =(extremely poor, very poor, poor, slightly poor, fair, slightly good,
good, very good, extremely good). Here, s ( 1,2,..., l 1) can be defined as a linguistic variable.
Let si and s j are any two elements in linguistic set S , they must meet the following conditions
[28][29]:
(1) If i j , then si s j (i.e., si is better than s j );
(2) Negative operator: neg ( si ) sl 1i ;
(3) If si s j , then max( si , s j ) si ;
(4) If si s j , then min( si , s j ) si .

For any linguistic set S ( s0 , s1 , , sl 1 ) , the element s is a strictly monotonically increasing
function of its subscript [30]. In order to minimize the loss of linguistic information, the discrete
linguistic set S ( s0 , s1 , , sl 1 ) is extended to a continuous linguistic set S {s | [0, q ]} , q is
sufficiently large number. The operational laws are defined as follows:[28][29]
3
(1) si s i 0 (1)
(2) si s j si j (2)
(3) si / s j si / j (3)
(4) si n si n 0
n
(4)
Definition 1 [31]. Let ~s [ s a , sb ] , sa , sb S and b a 0 , s a , sb are the lower limit and upper limit
~
of s , respectively, then, ~s is called an uncertain linguistic variable.
Let ~ s [s , s ] , ~
1 a1 s [ s , s ] be any two uncertain linguistic variables, the operation rules are
b1 2 a2 b2
defined as follows[32][33]:
(1) s1 s2 [ sa1 , sb1 ] [ sa 2 , sb 2 ] [ sa1 a 2 , sb1 b 2 ] (5)
(2) s1 s2 [ sa1 , sb1 ] [ sa 2 , sb 2 ] [ sa1a 2 , sb1b 2 ] (6)
(3) s1 / s2 [ sa1 , sb1 ] / [ sa 2 , sb 2 ] [ sa1/ b 2 , sb1/ a 2 ]if a 2 0, b2 0 (7)
(4) s1 [ sa1 , sb1 ] [ s *a1 , s *b1 ] 0 (8)
2.2 The neutrosophic set

Definition 2 [34]. Let X be a universe of discourse, with a generic element in X denoted by x. A
neutrosophic number A in X is
A( x ) x | (T A ( x), I A ( x), FA ( x)) (9)
where, TA ( x) is the truth-membership function, I A ( x) is the indeterminacy-membership function,
and FA ( x) is the falsity-membership function. TA x , I A x and FA x are real standard or
nonstandard subsets of 0 ,1 .
There is no restriction on the sum of TA ( x) , I A ( x) and FA ( x) , so 0 TA ( x) I A ( x) FA ( x) 3 .
Definition 3 [35]. Let X be a universe of discourse, with a generic element in X denoted by x. A single
valued neutrosophic number A in X is
A( x ) x | (T A ( x), I A ( x), FA ( x)) (10)
where TA ( x) is the truth-membership function, I A ( x) is the indeterminacy-membership function,

and FA ( x) is the falsity-membership function. For each point x in X, we
have TA ( x), I A ( x), FA ( x) [0,1] , and 0 TA ( x) I A ( x ) FA ( x ) 3 .
Definition 4 [35]. Let X be a universe of discourse, with a generic element in X denoted by x. A
interval neutrosophic set A in X is
A( x ) x | (T A ( x), I A ( x), FA ( x)) (11)
4
where, TA ( x) is the truth-membership function, I A ( x) is the indeterminacy-membership function,
and FA ( x) is the falsity-membership function. For each point x in X, we have
TA ( x), I A ( x), FA ( x) [0,1] , and 0 sup(TA ( x)) sup( I A ( x)) sup( FA ( x)) 3 .
For convenience, we can use x ([T L , T U ],[ I L , I U ],[ F L , F U ]) to represent an element in INS, and
can call an interval neutrosophic number (INN).
2.3 Interval-valued hesitant fuzzy set

Definition 5 [36][37]. A hesitant fuzzy set (HFS) E on X is defined in terms of a function that
when applied to X , which returns a finite subset of [0, 1], to be easily understood, the HFS can be
expressed by mathematical symbol as follows:
E { x, h( x) | x X } (12)
where h(x) is a set of some different values in [0, 1], representing the possible membership
degrees of the element x X to E .From now on, it will be convenient to call h(x) a hesitant fuzzy
element (HFE) and E the set of all hesitant fuzzy elements (HFEs).
For any three HFEs h, h1 and h2 , Torra [22] defined some basic operations shown as follow:
(1) h c 1 , (13)
h
(2) h1 h2 max 1 , 2 , (14)

1 h1 , 2 h2
(3) h1 h2 min 1 , 2 . (15)

1h1 , 2 h2
After that, Xia and Xu [23] defined four operations about the HFEs h, h1 , h2 with a positive
scale n :
(1) h n n ,
h
(16)
h

(2) nh 1 (1 ) n , (17)
(3) h1 h2 1 2 1 2 , (18)
1h1 , 2 h2
(4) h1 h2 1 2 . (19)
1h1 , 2 h2
~
Definition 6 [38]. An interval-valued hesitant fuzzy set (IVHFS) E on X is defined in terms of a
function that when applied to X , which returns a set of finite closed sub-intervals in[0,1], the IVHFS
can be expressed by mathematical symbol as follows:
~ ~
E { x, h ( x) | x X } (20)
~
where h ( x) is a set of some different interval values in [0,1], representing the possible
~ ~
membership degrees of the element x X to E . From now on, it will be convenient to call h ( x)
~
an interval-valued hesitant fuzzy element (IVHFE) and E the set of all hesitant fuzzy elements
5
(IVHFEs).
2.4 interval-valued neutrosophic hesitant fuzzy uncertain linguistic set
~
Definition 7 [39]. Let X is a fixed set and S is an uncertain linguistic set, then the hesitant fuzzy
uncertain linguistic set (HFULS) on X can be defined as follows:
A { x, [ s ( x ) , s ( x ) ], h( x)) | x X } (21)
~
where [ s ( x ) , s ( x ) ] S and h(x) is a set of some values in [0,1], denoting the possible
membership degree of the element x X to the uncertain linguistic variable [ s ( x ) , s ( x ) ] .From now
on, it will be convenient to call a [ s ( a ) , s ( a ) ], h( a ) a hesitant fuzzy uncertain linguistic element
(HFULE) and A the set of all HFULEs.

~
Definition 8. Let X be a fixed set and S be an uncertain linguistic set, then an interval-valued
neutrosophic hesitant fuzzy uncertain linguistic set (IVNHFULS) on X can be expressed by:
~ ~ ~
A { x, [ s ( x ) , s ( x ) ], ( t ( x), i ( x), f ( x)) | x X } (22)
~

where [ s ( x ) , s ( x ) ] S , ~t ( x) ~ ~ ~t ( x), ~i ( x) ~ ~ ~i ( x) , and f ( x) ~ ~ f ( x) are three
~ ~

sets of some values in [0,1], which represents the possible truth-membership degrees,
indeterminacy-membership degrees, and falsity-membership degrees of the element x X to the
uncertain linguistic variable [ s ( x ) , s ( x ) ] , and satisfies these limits :
~ ~
~ [0,1] [0,1]~ [0,1] and 0 sup ~ sup sup~ 3 ,
where ~ ~~t ( x ) max~ , ~ ~~i ( x ) max~, and ~ ~ ~f ( x ) max~ for x X .
~ ~ ~
Hence, a [ s ( a ) , s ( a ) ],{t (a), i (a), f (a)} can be described as an interval-valued neutrosophic
hesitant fuzzy uncertain linguistic element (IVNHFULE). So, A is the set of all IVNHFULEs.
~ ~ ~ ~ ~ ~ ~ ~ ~
Let a~ [ s ( a~ ) , s ( a~ ) ], (~t (a~ ), i (a~ ), f (a~ )) and b [ s (b~ ) , s (b~ ) ], ( t (b ), i (b ), f (b )) be any
two IVNHFULEs, the operation rules are defined as follows:

~
a~ b [s (a~) (b~) , s (a~) (b~) ], ([~
ra~ L ~
rb~ L ~
ra~ L ~
rb~ L , ~
ra~U ~
rb~U ~
ra~U ~
rb~U ],
~ ~ ~ ~ ~
r ( a~)t (a~ ),~r (b )t (b )
~ ~ ~ ~ ~ ~
( a~)i ( a~), (b )i (b )
~ ~ ~
~ (a~ ) f ( a~ ),~ (b ) f (b)
~ ~ ~ ~
(1) [ a~ L b~ L , a~U b~U ],[~a~ L~b~ L ,~a~U~b~U ]) (23)
~ ~ ~ ~ ~ ~ ~ ~ ~
a~ b [s (a~) (b~) , s (a~) (b~) ], ([~
ra~ L ~
rb~ L , ~
ra~U ~
rb~U ],[ a~ L b~ L a~ Lb~ L , a~U b~U a~Ub~U ]
~ ~ ~ ~~
r (a~)t (a~),~
r (b )t (b )
~ ~ ~~ ~~
(2) (a~)i (a~), (b )i (b )
~ ~ ~
~(a~) f (a~),~(b ) f (b)
[~a~ L ~b~ L ~a~ L~b~ L ,~a~U ~b~U ~a~U~b~U ]) (24)
6
a~ [ s ( a~ ) , s ( a~ ) ], ra~ L ) ,1 (1 ~
([1 (1 ~ ra~ U ) ],
(3) ~ ~ ~ ~
r ( a~ )t ( a~ ), ( a~ )i ( a~ ),~ ( a~ ) f ( a~ )
~
~ ~
[( a~ L ) , ( a~ U ) ], [(~a~ L ) , (~a~ U ) ]) (25)
~ ~
a~ [ s ( a~ ) , s ( a~ ) ], ra~ L ) , (~
([(~ ra~ U ) ], [1 (1 a~ L ) ,1 (1 a~ U ) ],
(4) ~ ~ ~ ~
r ( a~ )t ( a~ ), ( a~ )i ( a~ ),~ ( a~ ) f ( a~ )
~
[1 (1 ~a~ L ) ,1 (1 ~a~ U ) ]) (26)
~ ~ ~ ~ ~ ~ ~ ~ ~
Let a~ [ s ( a~ ) , s ( a~ ) ], (~
t (a~ ), i (a~ ), f (a~ )) , b [ s (b~ ) , s (b~ ) ], ( t (b ), i (b ), f (b )) are any two
IVNHFULEs, and , 1 , 2 0 , then we have:

~ ~
(1) a~ b b a~ (27)
~ ~
(2) a~ b b a~ (28)
~ ~
(3) (a~ b ) a~ b (29)
(4) 1a~ 2 a~ (1 2 )a~ (30)
~ ~
(5) a~ b (a~ b ) (31)
(6) a~ 1 a~ 2 a~ 1 2 (32)
Proof.
(1) The formula (27) is obviously right according to the operational rule (1) expressed by (23)
(2) The formula (28) is obviously right according to the operational rule (2) expressed by (24)
(3) For the left of the formula (29), we have
~
(a~ b ) ( [s ~ ,s ~ ],
(a~ ) (b ) ( a~) (b ) ([~ra~ L ~rb~ L ~
ra~ L ~
rb~ L , ~
ra~U ~
rb~U ~
ra~U ~
rb~U ],
~ ~ ~ ~ ~
r ( a~)t (a~ ),~
r (b )t (b )
~ ~ ~ ~ ~ ~
( a~)i ( a~), (b )i (b )
~ ~ ~
~ (a~ ) f ( a~),~(b ) f (b)
~ ~ ~ ~
[ a~ L b~ L , a~U b~U ],[~a~ L~b~ L ,~a~U~b~U ]) )
[ s ( ( a~ ) (b~ )) , s ( ( a~ ) (b~ )) ], ra~ L ) (1 ~

([1 (1 ~ rb~ L ) ,1 (1 ~
ra~ U ) (1 ~
rb~ U ) ],
~ ~ ~ ~ ~
r ( a~ )t ( a~ ), ~
r ( b )t (b )
~ ~ ~ ~ ~ ~
( a~ )i ( a~ ), (b )i (b )
~ ~ ~
~ ( a~ ) f ( a~ ),~ (b ) f (b )
~ ~ ~ ~
[( a~ L b~ L ) , ( a~ U b~ U ) ], [(~a~ L~b~ L ) , (~a~U ~b~ U ) ])
and for the right of the formula (29), we have
[ s ( a~ ) , s ( a~ ) ], ra~ L ) ,1 (1 ~
([1 (1 ~ ra~U ) ]
~ ~ ~ ~
r ( a~ )t ( a~ ), ( a~ )i ( a~ ),~ ( a~ ) f ( a~ )
~
~ ~
[( a~ L ) , ( a~ U ) ], [(~a~ L ) , (~a~U ) ])
[ s (b~ ) , s (b~ ) ], rb~ L ) ,1 (1 ~

([1 (1 ~ rb~U ) ]
~ ~ ~ ~ ~ ~ ~ ~ ~ ~
r (b )t (b ), (b )i (b ),~ (b ) f ( b )
~
~ ~
[( b~ L ) , ( b~U ) ], [(~b~ L ) , (~b~U ) ])
[ s ( ( a~ ) (b~ )) , s ( ( a~ ) (b~ )) ], ra~ L ) (1 ~

[(1 (1 ~ rb~ L ) ,1 (1 ~
ra~ U ) (1 ~
rb~ U ) ],
~ ~ ~ ~ ~
r ( a~ )t ( a~ ), ~
r ( b )t (b )
~ ~ ~ ~ ~ ~
( a~ )i ( a~ ), (b )i (b )
~ ~ ~
~ ( a~ ) f ( a~ ),~ (b ) f (b )
~ ~ ~ ~
[( a~ L b~ L ) , ( a~ U b~ U ) ], [(~a~ L~b~ L ) , (~a~U ~b~ U ) ])
7
So, we can get (a b ) a b , which completes the proof of formula (29).
(4) For formula (30), we have
1a~ 2 a~ [ s1 ( a~ ) , s1 ( a~ ) ], ra~ L ) 1 ,1 (1 ~
([1 (1 ~ ra~U ) 1 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~
~ ( a~ ) f ( a~ )
~ ~
[( a~ L ) 1 , ( a~U ) 1 ], [(~a~ L ) 1 , (~a~U ) 1 ])
[ s2 ( a~ ) , s2 ( a~ ) ], ([1 (1 ~ ra~ L ) 2 ,1 (1 ~ra~U ) 2 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~
~ ( a~ ) f ( a~ )
~ ~
[( a~ L ) 2 , ( a~U ) 2 ], [(~a~ L ) 2 , (~a~U ) 2 ])
[ s( 1 2 ) ( a~ ) , s(1 2 ) ( a~ ) ], ra~ L ) 1 (1 ~
([1 (1 ~ ra~ L ) 2 ,1 (1 ~
ra~U ) 1 (1 ~
ra~U ) 2 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~
~ ( a~ ) f ( a~ )
~ ~ L 1 ~ ~
[( a~ ) ( a~ L ) 2 , ( a~U ) 1 ( a~U ) 2 ], [(~a~ L ) 1 (~a~ L ) 2 , (~a~U ) 1 (~a~U ) 2 ])
[ s( 1 2 ) ( a~ ) , s(1 2 ) ( a~ ) ], ([1 (1 ~
ra~ L ) 1 2 ,1 (1 ~
ra~U ) 1 2 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~ ~
( a ) f ( a~ )
~
~ L 1 2 ~ U 1 2
[( a~ ) , ( a~ ) ], [(~a~ L ) 1 2 , (~a~U ) 1 2 ])
(1 2 )a~
So, we can get the formula (30) is right.
(5) For the left of the formula (31), we have
~ ~ ~
a~ b [ s ,s ],
( a~ )
r~ L ) , (~
([(~
( a~ )
r~ U ) ], [1 (1 ~ L ) ,1 (1 ~U ) ],
a a a a
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~
~ ( a~ ) f ( a~ )
[1 (1 ~a~ L ) ,1 (1 ~a~U ) ])
~ ~
[ s (b~ ) , s (b~ ) ], rb~ L ) , (~
([(~ rb~U ) ], [1 (1 b~ L ) ,1 (1 b~ U ) ],
~ ~ ~ ~
r (b )t (b ),
~ ~ ~ ~
(b )i (b ),
~ ~ ~
~ ( b ) f (b )
[1 (1 ~b~ L ) ,1 (1 ~b~ U ) ])
[ s ( a~ ) (b~ ) , s ( a~ ) (b~ ) ], ra~ L ) (~

([(~ rb~ L ) , (~
ra~U ) (~
rb~U ) ],
~ ~ ~ ~ ~
r ( a~ )t ( a~ ), ~
r (b )t (b ),
~ ~ ~ ~ ~ ~
( a~ )i ( a~ ), (b )i ( b ),
~ ~ ~ ~
~ ( a~ ) f ( a~ ),~ (b ) f (b )
~L ~L ~U ~
[1 (1 a~ ) (1 b~ ) ,1 (1 a~ ) (1 b~ U ) ], [1 (1 ~a~ L ) (1 ~b~ L ) ,1 (1 ~a~ U ) (1 ~b~ U ) ])
[ s( ( a~ ) (b~ )) , s( ( a~ ) (b~ )) ], ([(~ rb~ L ) , (~

ra~ L ~ rb~U ) ],
ra~ U ~
~ ~ ~ ~ ~
r ( a~ )t ( a~ ), ~
r ( b )t (b ),
~ ~ ~ ~ ~ ~
( a~ )i ( a~ ), (b )i (b ),
~ ~ ~ ~
~ ( a~ ) f ( a~ ),~ (b ) f ( b )
~ ~ ~ ~
[1 ((1 a~ L )(1 b~ L )) ,1 ((1 a~ U )(1 b~ U )) ], [1 ((1 ~a~ L )(1 ~b~ L )) ,1 ((1 ~a~ U )(1 ~b~ U )) ])
and for the right of the formula (31), we have
~
(a~ b ) ( [ s ~ ,s ~ ],
( a~ ) (b ) ( a~ ) (b ) ([~
ra~ L ~
rb~ L , ~
ra~U ~
rb~ U ],
~ ~ ~ ~ ~
r ( a~ )t ( a~ ), ~
r ( b )t ( b )
~ ~ ~ ~ ~ ~
( a~ )i ( a~ ), (b )i (b )
~ ~ ~
~ ( a~ ) f ( a~ ),~ (b ) f ( b )
~ L ~ L ~ L~ L ~ U ~ U ~ U ~ U ~ L
[ a~ b~ a~ b~ , a~ b~ a~ b~ ], [ a~ ~b~ L ~a~ L~b~ L ,~a~ U ~b~U ~a~U ~b~U ]) )
8
[ s( ( a~ ) (b~ )) , s( ( a~ ) (b~ )) ], ([(~ rb~ L ) , (~
ra~ L ~ rb~ U ) ],
ra~U ~
~ ~ ~ ~ ~
r ( a~ )t ( a~ ), ~
r (b )t (b ),
~ ~ ~ ~ ~ ~
( a~ )i ( a~ ), (b )i ( b ),
~ ~ ~ ~
~ ( a~ ) f ( a~ ),~ ( b ) f (b )
~L ~L ~U ~
[1 ((1 a~ )(1 b~ )) ,1 ((1 a~ )(1 b~U )) ], [1 ((1 ~a~ L )(1 ~b~ L )) ,1 ((1 ~a~ U )(1 ~b~U )) ])

(6) For formula (32), we have
~ ~
a~ 1 a~ 2 [ s , s ], ( a~ ) 1 ( a~ ) 1 ra~ L ) 1 , (~
([(~ ra~ U ) 1 ], [1 (1 a~ L ) 1 ,1 (1 a~ U ) 1 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~
~ ( a~ ) f ( a~ )
[1 (1 ~a~ L ) 1 ,1 (1 ~a~ U ) 1 ])
~ ~
[ s ( a~ )2 , s ( a~ )2 ], ra~ L ) 2 , (~
([(~ ra~ U ) 2 ], [1 (1 a~ L ) 2 ,1 (1 a~U ) 2 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~
~ ( a~ ) f ( a~ )
[1 (1 ~a~ L ) 2 ,1 (1 ~a~ U ) 2 ])
[ s ( a~ )1 ( a~ ) 2 , s ( a~ )1 ( a~ ) 2 ], ra~ L ) 1 (~
([(~ ra~ L ) 2 , (~
ra~ U ) 1 (~
ra~U ) 2 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~
~ ( a~ ) f ( a~ )
~ ~ ~ ~ U 1
[1 (1 a~ L ) 1 (1 a~ L ) 2 ,1 (1 a~ ) (1 a~ U ) 2 ], [1 (1 ~a~ L ) 1 (1 ~a~ L ) 2 ,1 (1 ~a~ U ) 1 (1 ~a~U ) 2 ])
~ ~
[ s ( a~ ) 12 , s ( a~ )12 ], ra~ L ) 1 2 , (~
([(~ ra~ U ) 1 2 ], [1 (1 a~ L ) 1 2 ,1 (1 a~ U ) 1 2 ],
~ ~
r ( a~ )t ( a~ ),
~ ~
( a~ )i ( a~ ),
~ ~
( a ) f ( a~ )
~
[1 (1 ~a~ L ) 1 2 ,1 (1 ~a~ U ) 1 2 ])
a~ 1 2
~ ~
Definition 9 [9,15]. Let a~ [ s ( a~ ) , s ( a~ ) ], (~
t (a~ ), i (a~ ), f (a~ )) be an interval-valued neutrosophic
hesitant fuzzy uncertain linguistic element (IVNHFULE) , then the score function S (a~ ) of a~ can be
defined as follows:
1 ~
r a~ L ~
r a~ U 1 (1 a~ ) (1 a~ )
~ ~
1
L U
(1 ~a~ ) (1 ~a~ )
L U
[ ~ ~r ( a~ )~t ( a~ ) ( ) ~ ~ ( a~ )~i ( a~ ) ( ) ~ ( a~ ) ~f ( a~ ) ( )]
#t 2 #i 2 #f 2
S (a~ )
3
s ( a~ ) ( a~ )
2
s 1 ~
r a~ L ~
r a~ U 1
~ ~
(1 a~ L ) (1 a~ U ) 1 (1~a~ L ) (1~a~ U )
(33)
( ( a~ ) ( a~ ))*[ ~ r~ ( a~ )~t ( a~ ) ( ) ~ ~ ( a~ )~i ( a~ ) ( ) ~ ( a~ ) ~f ( a~ ) ( )]
#t 2 #i 2 #f 2
6
~ ~ ~ ~ ~ ~
where # t , # i , # f are the number of the interval numbers in t (a~ ) , i (a~ ) , f (a~ ) .
~ ~ ~ ~ ~ ~ ~ ~ ~
Let a~ [ s ( a~ ) , s ( a~ ) ], (~
t (a~ ), i (a~ ), f (a~ )) and b [ s (b~ ) , s (b~ ) ], ( t (b ), i (b ), f (b )) are any
two IVNHFULEs, the comparison method of IVNHFULEs is expressed as follows:

~ ~
(1) If S (a~ ) S (b ) ,then a~ b
9
~ ~
(2) If S (a~ ) S (b ) ,then a~ b
~ ~
(3) If S (a~ ) S (b ) ,then a~ b
2.4 The Bonferroni mean (BM) operator

Definition 10 [40]. Let ai (i 1,2,..., n) be a set of non-negative numbers, and p, q 0 , the Bonferroni
mean operator is defined as follows:
1
pq
1 n p q
BM (a1 , a 2 ,..., a n )
p,q
ai a j (34)
n(n 1) ij,j i1

In real decision making, the weight of input arguments plays an important part in decision process.
But the BM operator doesnt consider the importance of input argument itself. Hence, Zhou [41]
developed a weighted Bonferroni mean (WBM) operator.
Definition 11 [41]. Let a i (i 1, 2 ,..., n ) be a set of non-negative numbers, and p , q 0 . The
weighted Bonferroni mean operator is defined as follows:
1
pq
n wi w j p q
WBM (a1 , a2 ,..., a n )
p,q
ai a j (35)
ii ,j j1 1 wi

where W ( w1 , w2 ,..., wn )T is the weight vector of ai (i 1,2,..., n) , satisfying 0 wi 1 and
w
i 1
i 1.
The WBM operator has four desirable properties as follows:

Theorem 1 [41] (Reducibility).
1 1 1
Let W ( , ,..., )T be the weight vector of ai (i 1,2,..., n) ,then
n n n
1
pq
1 n p q
WBM (a1 , a2 ,..., an )
p,q
ai a j BM p ,q (a1 , a2 ,..., an ) .
n ( n 1) i , j 1
i j
Theorem 2 [41] (Idempotency).
Let a j a( j 1,2,..., n) , then WBM p,q

( a1 , a 2 ,..., a n ) a .
Theorem 3 [41] (Permutation).

Let ( a 1 , a 2 ,..., a n ) be any permutation of ( a 1 ' , a 2 ' ,..., a n ' ) , then
WBM p,q
(a1 ' , a 2 ' ,..., a n ' ) WBM p,q
(a1 , a 2 ,..., a n ) .
Theorem 4 [41] (Monotonicity).
If a j b j ( j 1,2,...,n) , then WBM p,q (a1, a2 ,...,an ) WBM p,q (b1, b2 ,...,bn ) .
Theorem 5 [41] (Boundedness).
10
The WBM p, q operator lies between the max and min operators, i.e.,
min(a1 , a 2 ,..., a n ) WBM p,q

(a1 , a 2 ,..., a n ) max(a1 , a 2 ,..., a n ) .
When the WBM operator select different values to the parameters p and q , some special cases
of WBM can be obtained as follows:
1
n p
(1) If q 0 , then WBM p ,0
(a1 , a 2 ,..., a n ) wi ai p .
i 1
n
(2) If p 1 and q 0 ,then WBM 1,0 (a1 , a2 ,..., a n ) w a .
i 1
i i
1 1 n 1
wi w j

1 1 ,
(3) If p and q , then WBM 2 2 (a1 , a2 ,..., an ) ( ai a j ) 2 .
2 2 i , j 1
1 w i
j i
1
2
n ww
(4) If p 1 and q 1 ,then WBM (a1 , a 2 ,..., an )
1,1 i j
i , j 11 wi
ai a j .

j i

2.5 The geometric Bonferroni mean (GBM) operator
Definition 12 [42]. Let ai (i 1,2,...,n) be a set of non-negative numbers, and p, q 0 , the geometric
Bonferroni mean operator is defined as follows:
n 1

1
GBM p,q (a1, a2 ,...,an ) ( pai qa j ) n(n1)
p q i, j 1 (36)
j i
Similar to the BM operator, the GBN operator also doesnt consider the weights of the input
arguments. The weighted geometric Bonferroni mean (WGBM) operator was proposed by Sun and Liu
[43].
Definition 13 [43]. Let ai (i 1,2,...,n) be a set of non-negative numbers, and p , q 0 . The
weighted geometric Bonferroni mean operator is defined as follows:
wi w j
n

1
WGBM p,q
(a1 , a 2 ,..., a n ) ( pai qa j ) 1 wi
p q i , j 1 (37)
j i
where W ( w1 , w2 ,..., wn )T is the weight vector of ai (i 1,2,..., n) ,satisfying 0 wi 1 and
w
i 1
i 1.
It is easy to demonstrate the WGBM operator has some properties as follows:

Theorem 6 [43] (Reducibility).
1 1 1
Let W ( , ,..., )T be the weight vector of ai (i 1,2,..., n) ,then
n n n
WGBM p,q
(a1 , a 2 ,..., a n ) GBM p,q
(a1 , a 2 ,..., a n ) .
Theorem 7 [43] (Idempotency).
11
Let a j a( j 1,2,..., n) , then WGBM p,q (a1 , a2 ,..., an ) a .
Theorem 8 [43] (Permutation).

Let ( a 1 , a 2 ,..., a n ) be any permutation of ( a 1 ' , a 2 ' ,..., a n ' ) , then
WGBM p,q
( a1 ' , a 2 ' ,..., a n ' ) WGBM p,q
(a1 , a 2 ,..., a n ) .
Theorem 9 [43] (Monotonicity).
If a j bj ( j 1,2,...,n) , then WGBMp,q (a1, a2 ,...,an ) WGBMp,q (b1, b2 ,...,bn ) .
Theorem 10 [43] (Boundedness).

The WGBM p ,q operator lies between the max and min operators, i.e. ,
min(a1 , a 2 ,..., a n ) WGBM p,q

(a1 , a 2 ,..., a n ) max(a1 , a 2 ,..., a n )
When the WGBM operator select different parameters p and q , some special cases of WGBM
operator can be obtained as follows:
n
p
(1) If q 0 , then WGBM p,q (a1, a2 ,...,an ) 1 ( pa ) i
wi
i 1
n
(2) If q 0, p 1 then WGBM p,q (a1, a2 ,...,an ) a
i 1
i
wi
wi w j
1 1 n
1 wi

1 1 , 1 1
(3) If p and q ,then WGBM 2 2 (a1 , a 2 ,..., an ) ( ai a j )
2 2 i , j 1
2 2
j i
wi w j
n 1 wi

1
(4) If p 1 and q 1 ,then WGBM 1,1 (a1 , a2 ,..., a n ) ( ai a j )
2 i , j 1
j i
3. Interval-valued neutrosophic hesitant fuzzy uncertain linguistic
Bonferroni mean aggregation operators

Interval-valued neutrosophic hesitant fuzzy uncertain linguistic set (IVNHFULS) is an
important tool to describe the decision makers preference and to handle the hesitant, uncertain,
incomplete information. Bonferroni mean (BM) can replace the simple averaging of other mean type
operators because of the excellent modeling capability.
In this section, we combine IVNHFULS with BM operator and propose some interval-valued
neutrosophic hesitant fuzzy uncertain linguistic Bonferroni mean aggregation operators.
3.1 The interval-valued neutrosophic hesitant fuzzy uncertain linguistic
Bonferroni mean operator
~ ~ ~
Definition 14. Let a~i [ s ( a~i ) , s ( a~i ) ], ( t (a~i ), i (a~i ), f (a~i )) (i 1,2,..., n) be the set of all
IVNHFULEs, and p , q 0 , then the interval-valued neutrosophic hesitant fuzzy uncertain
12
linguistic Bonferroni mean operator can be defined as follows:
1
pq
p,q ~ ~ ~ 1 n
~ ~ q
IVNHFULBM (a1 , a 2 ,..., an ) ai a j
p
n(n 1) ij,j i1 (38)

~ ~
Theorem 11. Let a~i [ s ( a~i ) , s ( a~i ) ], (~
t (a~i ), i (a~i ), f (a~i )) (i 1,2,..., n) be the set of IVNHFULEs,
and p , q 0 , then, the result aggregated from Definition 14 is still an IVNHFULE, and even
1
1 n n pq
IVNHFULBM p,q
(a~1 , a~2 ,..., a~n ) a~i p a~ j q
n(n 1) i 1 j 1, j i

s 1 ,s 1 ,
n n 1 pq
. ( a~i ) p . ( a~ j ) q
n n

1 pq
. ( a~i ) p . ( a~ j ) q

i 1 j 1, j i n ( n 1)

i 1 j 1, j i n ( n 1)
~ ~ ~
~ r ( a~1 ) t ( a~1 ), ~ r ( a~2 ) t ( a~2 ),..., ~r ( a~n ) t ( a~n )
1 1
n n 1 pq n n 1 pq
1 (1 ( ra~i ) p .( ra~ j ) q ) n ( n 1) , 1 (1 ( ra~i ) p .( ra~ j ) q ) n ( n 1)
L L
U U

,

i 1 j 1 , j i i 1 j 1, j i
(39)
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
( a1 ) i ( a1 ), (a 2 ) i (a 2 )..., ( a n ) i ( a n )
1 1
1 pq 1 pq
n n ( n 1) n n ( n 1)

n n
1 1 1 (1 a~i L ) p (1 a~ j L ) q ,1 1 1 (1 a~i u ) p (1 a~ j u ) q ,

i 1 j 1 , j i i 1 j 1 , j i

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
(a1 ) f (a1 ), (a 2 ) f (a 2 ),..., (a n ) f (a n )
1 1
1 pq 1 pq
n n ( n 1) n n ( n 1)

n n
1 1 1 (1 a~i L ) p (1 a~ j L ) q ,1 1 1 (1 a~i u ) p (1 a~ j u ) q
i 1 j 1, j i
i 1 j 1, j i

Proof .
By the operational rules of the interval-valued neutrosopic hesitant fuzzy uncertain linguistic
variables, we have
a~i p [ s ( a~i ) p , s ( a~i ) p ], {(ra~i ) p , (ra~i ) p }

L U
~
~
r ( a~i )t ( a~i )
a~ j q [ s ( a~ j ) q , s ( a~ j ) q ], {( ra~ j ) q , (ra~ j ) q }
L U
~
~
r ( a~ j )t ( a~ j )
a~i p a~ j q [ s ( a~i ) p . ( a~ j ) q , s ( a~i ) p . ( a~ j ) q ],

((ra~i ) p .(ra~ j L ) q , (ra~i ) p .(ra~ j ) q ),
L U U
~
~
~
r ( a~i )t ( a~i ), ~
r ( a~ j )t ( a~ j )
~ ~ ~ ~
~ ~
~ ~
(1 (1 a~i L ) P (1 a~ j L ) P ,1 (1 a~i U ) P (1 a~ j U ) P )
( a~i )i ( a~i ), ( a~ j )i ( a~ j )
~
~
(1 (1 ~a~i L ) P (1 ~a~ j L ) P ,1 (1 ~a~i U ) P (1 ~a~ j U ) P )
~ ( a~i ) f ( a~i ),~ ( a~ j ) f ( a~ j )
then
13
1 ~p ~q
ai a j s 1 , s 1
,
n(n 1) n ( n 1) . ( a~i ) p . ( a~ j ) q n ( n 1) . ( a~i ) p . ( a~ j ) q

1 1
1 (1 r ( ~ L ) p .(r~ L ) q ) n ( n 1) ,1 (1 (r~ U ) p .(r~ U ) q ) n ( n 1) ,

~r ( a~ )~t ( a~ ),~r ( a~ )~t ( a~ ) ai aj ai aj
i

i j j

1 1
~ (1 (1 ~ L ) p (1 ~ L ) q ) n ( n 1) , (1 (1 ~ U ) p (1 ~ U ) q ) n ( n 1) ,
( a~i )i ( a~i ), ( a~ j )i ( a~ j )
~ ~ ~ ai aj ai aj

1 1
(1 (1 ~ L ) p (1 ~ L ) q ) n ( n 1) , (1 (1 ~ U ) p (1 ~ U ) q ) n ( n 1)

~ ( a~i ) f ( a~i ),~ ( a~ j ) f ( a~ j )
~ ~ a i a j ai a j

and
1 n n n n 1 ~p ~q
a~i p a~ j q ai a j
n(n 1) i 1 j i i 1 j i n(n 1)

s n n , s ,
q
n n

1 ~ ~ 1 ~ ~
( ai ) ( a j )
p q
( ai ) ( a j )
p
i1 j1, j i n ( n 1) n ( n 1)
i 1 j 1, j i
~ ~ ~
~r (a~1 ) t (a~1 ), ~ r (a~2 ) t (a~2 ),..., ~r (a~n ) t (a~n )
n n 1 n n 1
1 U q n ( n 1)
( 1 ( r~
a i
L p
) .( r~
a j
L q n ( n 1)
) ) ,1
( 1 ( r~
a i
U p
) .( r~
a j
) ) ,
i 1 j 1, j i i 1 j 1, j i
~ ~ ~
~ (a~1 ) i (a~1 ), ~ (a~2 ) i (a~2 )..., ~ (a~n ) i (a~n )
n

n 1 n n 1
1 (1 a~i L ) p (1 a~ j L ) q n ( n 1) , 1 (1 a~i U ) p (1 a~ j U ) q n ( n 1) ,

i 1 j 1, j i i 1 j 1, j i
~ ~ ~
~ (a~1 ) f (a~1 ),~ (a~2 ) f (a~2 ),...,~ (a~n ) f (a~n )
n

n 1 n n 1
1 (1 a~i L ) p (1 a~ j L ) q n ( n 1) , 1 (1 a~i U ) p (1 a~ j U ) q n ( n 1)

i 1 j 1, j i i 1 j 1, j i
Then
1
1 n n pq
a~i p a~ j q s 1 ,s 1 ,
n(n 1) i 1 j i n n 1 pq n n
. ( a~i ) p . ( a~ j ) q
1 pq
. ( a~i ) p . ( a~ j ) q

i1 j 1, j i n ( n 1) i 1 j 1, j i n ( n 1)
~ ~ ~
~ r ( a~1 ) t ( a~1 ), ~r ( a~2 ) t ( a~2 ),..., ~
r ( a~n ) t ( a~n )
1 1
n n 1 pq n n 1 pq
, 1
1
(1 ( ra~i ) .( ra~ j ) )
L p L q n ( n 1)
(1 (r a~i
U p U
) .( ra~ j ) ) q n ( n 1)

,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~
~ ( a~ ) i ( a~ ), ~ ( a~ ) i (a~ )..., ~ (a~ ) i ( a~ )
1 1 2 2 n n
1 1
pq pq

n n 1 n n 1
1 1 1 (1 a~i L ) p (1 a~ j L ) q n ( n 1) ,1 1 1 (1 a~i U ) p (1 a~ j U ) q n ( n 1) ,

i 1 j 1, j i i 1 j 1, j i

14
~ ~ ~
~ (a~1 ) f (a~1 ),~ (a~2 ) f (a~2 ),...,~ (a~n ) f (a~n )
1 1
pq pq
1 (1
n n 1 n n 1

1 1 1 (1 a~i ) (1 a~ j ) n ( n 1)
L p L q ,1 1 u p
) (1 a~ j u ) q n ( n 1)
a~i

i 1 j 1, j i i 1 j 1, j i

which completes the proof of theorem 11.
Moreover, the IVNHFULBM operator also has the following properties:
Theorem 12 (Idempotency).
Let a~ j [ S ( a ) , s ( a ) ], ([ ra L , ra U ], [ a L , a U ], [ a L , a U ]) a~ ( j 1,2,..., n) , then
IVNHFULBM p,q
(a~1 , a~2 ,..., a~n ) a~
Proof .
Since a~ j [ s ( a ) , s ( a ) ], {[ ra L , ra U ]} a~ ( j 1, 2 ,..., n ) , then according to the theorem 11,
we can get
1
1 n n pq
IVNHFULBM p,q
(a~1 , a~2 ,..., a~n ) a~i p a~ j q
n(n 1) i 1 j 1, j i

s 1 ,s 1 ,
n n 1 pq n n
. ( a~ ) p . ( a~ ) q
1 pq
. ( a~ ) p . ( a~ ) q
i1 j 1, j i n ( n 1) i 1 j 1, j i n ( n 1)
~ ~ ~
~ r ( a~1 ) t ( a~1 ), ~r ( a~2 ) t ( a~2 ),..., ~
r ( a~n ) t ( a~n )
1 1
n n 1 pq n n 1 pq
1 L q n ( n 1)
, 1
( 1 ( r
L p
~ ) .( ra~ ) )
a (1 (r a~
U p
) .( ra~ ) )
U q n ( n 1)

,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~
~ ( a~ ) i (a~ ), ~ ( a~ ) i ( a~ )..., ~ ( a~ ) i ( a~ )
1 1 2 2 n n
1 1
pq pq
1 (1
n n 1 n n 1
1 1 1 (1 a~ ) p (1 a~ ) q n ( n 1)
L L
,1 1 a~
U
) p (1 a~ U ) q n ( n 1) ,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~
~ (a~ ) f (a~ ),~ (a~ ) f (a~ ),...,~ (a~ ) f (a~ )
1 1 2 2 n n
1 1
pq pq
1 (1 1 (1
n n 1 n n 1
1 1 a~
L p
) (1 a~ L ) q n ( n 1) ,1 1 a~
U
) p
(1 a~ U ) q n ( n 1)

i 1 j 1, j i i 1 j 1, j i

[ s ( a ) , s ( a ) ], ([ ra L , ra U ], [ a L , a U ], [ a L , a U ]) a~
So, we complete the proof of the theorem 12.

Theorem 13 (Commutativity).
Let (a~1' , a~2 ' ,...,a~n ' ) be any permutation of (a~1 , a~2 ,..., a~n ) , then we can get
IVNHFULBM p,q
(a~1 ' , a~2 ' ,..., a~n ' ) IVNHFULBM p,q
(a~1 , a~2 ,..., a~n ) .
Proof.
15
1
1 n n pq
IVNHFULBM p ,q (a~1' , a~2 ' ,..., a~n ' ) a~i ' p a~ j 'q
n(n 1) i 1 j 1, j i

s 1 ,s 1 ,
n n 1 pq
. ( a~i ' ) p . ( a~ j ' ) q
n n

1 p q
. ( a~i ' ) p . ( a~ j ' ) q

i1 j 1, j i n ( n 1) i 1 j 1, j i n ( n 1)
~ ~ ~
~ r ( a~ '1 ) t ( a~1' ), ~r ( a~2 ' ) t ( a~2 ' ),..., ~
r ( a~n ' ) t ( a~n ' )
1 1
n n 1 pq n n 1 pq
1 L q n ( n 1)
, 1
(1 ( r~
ai
L p
' ) .( r~ ' ) )
aj
(1 ( ra~ ' ) p .( ra~
i
U
j
'
U q n ( n 1)
) )

,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~
~ ( a~ ' ) i ( a~ ' ), ~ ( a~ ' ) i ( a~ ' )..., ~ ( a~ ' ) i ( a~ ' )
1 1 2 2 n n
1 1
n n 1 pq n n 1 pq
1 1
1 (1 ' L ) p (1 ' )
a~i a~ j
L q n ( n 1)

,1 1

1 (1 ' U ) p (1 ' )
a~i a~ j
U q n ( n 1)

,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~
~ (a~1' ) f (a~1' ),~ (a~2 ' ) f (a~2 ' ),...,~ (a~n ' ) f (a~n ' )
1 1
n n 1 pq n n 1 pq

1 1
1 (1 ' L ) p (1 ' L ) q n ( n 1)

~
ai ~
aj

,1 1

1 (1 a~i '
U U q n ( n 1)
) p (1 a~ ' )
j

i 1 j 1, j i i 1 j 1, j i

Since a , a ,..., a
'
1
'
2
'
n
is any permutation of a1 , a2 ,..., an , then we have
n n n n
(a~i ) (a~i ) , (a~ j ) (a~ j ) ,
' p p ' q q
i 1 i 1 j 1, j i j 1, j i
n
(a~ )
n n n
(a~i ) , (a~ j ) (a~ j ) .
p ' p ' q q
i
i 1 i 1 j 1, j i j 1, j i
so, we can get

1
1 pq
n n 1 pq 1 n n ' q
(a~i ' ) p (a~ j ' ) q (a~i ) (a~ j )
' p
i 1 j 1, j i n(n 1) n(n 1) i 1 j 1, j i
1
pq
1 n n
q
(a~i ) (a~ j )
p
n ( n 1) i 1 j 1, j i
1
1 p q
n n 1 pq 1 n n ' q
(a~i ' ) p (a~ j ' ) q (a~i ) (a~ j )
' p
i 1 j 1, j i n(n 1) n(n 1) i 1 j 1, j i 1
p q
1 n n
q
(a~i ) (a~ j )
p
n(n 1) i 1 j 1, j i
and
16
1
n n 1
n n n ( n 1)
(1 ( ra~ ' ) p .( ra~
i
L
j
'
L q n ( n 1)
) )
(1 ( ra~ ' ) p .( ra~ ' ) q )
i
L
j
L

i 1 j 1, j i i 1 j 1, j i
1
n n n ( n 1) n n 1

(1 ( ra~i ) p .( ra~ j ) q )
L L

(1 ( ra~i ) p .( ra~ j ) q ) n ( n 1)
L L
i 1 j 1, j i i 1 j 1, j i
Similarly,
n n 1 n n 1
i 1 j 1, j i
(1 (ra~ ' U ) p (ra~ ' U ) q ) n ( n 1)
i j

i 1 j 1, j i
(1 (ra~i U ) p (ra~ j U ) q ) n ( n 1)
1 (1
n n n n
i 1 j 1, j i
1 (1 ' L ) p (1 ' L ) q
a~i a~ j i 1 j 1, j i
a~i
L p
) (1 a~ j L ) q
1 (1
n n n n

i 1 j 1, j i
1 (1 ' U ) p (1 ' U ) q
a~i a~ j i 1 j 1, j i
a~i
U
) p (1 a~ j U ) q
1 (1
n n n n
1 (1
i 1 j 1, j i
a~i '
L p
) (1 a~ ' L ) q
j i 1 j 1, j i
a~i
L p
) (1 a~ j L ) q
1 (1
n n n n
i 1 j 1, j i
1 (1 ' U ) p (1 ' U ) q
a~i a~ j i 1 j 1, j i
a~i
U
) p (1 a~ j U ) q
So, according to the theorem 13, we can get
IVNHFULBM p,q
(a~1 , a~2 ,..., a~n ) IVNHFULBM p,q
(a~1' , a~2 ' ,..., a~n ' ) .
Theorem 14 (Monotonicity).
~ ~ ~
Suppose (a~1 , a~2 ,..., a~n ) and (b1 , b2 ,..., bn ) are any two sets of IVNHFULEs. If s ( a~i ) s (b~ ) ,
i
s ( a~i ) s (b~ ) , ra~i L rb~ L , ra~i U rb~ U , a~i L b~ L , a~i U b~ U and a~i L b~ L , a~i U b~ U for all
i i i i i i i
i 1,2,..., n ,then
~ ~ ~
IVNHFULBM p,q
(b1 , b2 ,..., bn )
Proof.
~
(1) Since s ( a~i ) s (b~ ) , then (a~i ) (bi ) , according to theorem 11 , we have
i
1 1
n n pq n n
~ pq
(a~i ) p (a~ j ) q (bi ) p (b j ) q

1 1 ~
,
n ( n 1) n(n 1)
i 1 j 1, j i i 1 j 1, j i
(2) Similar to (1),we have
1 1
n n pq n n
~ pq
(a~i ) p (a~ j ) q (bi ) p (b j ) q

1 1 ~
n ( n 1) n(n 1)
i 1 j 1, j i i 1 j 1, j i
(3) Since ra~i L rb~ L for all i and p, q 0 , then (ra~ i L ) p (rb~ L ) p ,
i i
17
(ra~ i L ) p (ra~ j L ) q (rb~ L ) p (rb~ ) , 1 (ra~ i L ) p (ra~ j L ) q 1 (rb~ L ) p (rb~
L q L q
) ,
i j i j
1 1
1 (r
n n n ( n 1) n n n ( n 1)
i 1 j 1, j i
a~ i
L p
) (ra~ j ) L q
1 (r
i 1 j 1, j i
~
bi
L p
) (rb~
j
)
L q

and
1 1
1 (r
n n n ( n 1) n n n ( n 1)
1
i 1 j 1, j i
a~ i
L p
) (ra~ j L ) q 1 1 (r
i 1 j 1, j i
~
bi
L p
) (rb~
j
L q
)

So
1 1
1 pq 1 pq
1 (r
n n n ( n 1) n n n ( n 1)
1
i 1 j 1, j i
a~ i
L p
) (ra~ j L ) q

1

1 (r
i 1 j 1, j i
~
bi
L p
) (rb~
j
L q
)

Similar to (3), we have
1 1
1 pq 1 pq
1 (r
n n n ( n 1) n n n ( n 1)
1
i 1 j 1, j i
a~ i
U
) p (ra~ j U ) q

1

1 (r
i 1 j 1, j i
~
bi
U
) p (rb~
j
U q
)

(4) since a~i L b~ L for all i and p, q 0 , then (1 a~i L ) p (1 b~ L ) p ,

i i
(1 a~i L ) p (1 a~ j L ) q (1 b~ L ) p (1 b~ L ) q , 1 (1 a~i L ) p (1 a~ j L ) q 1 (1 b~ L ) p (1 b~ L ) q
i j i j

1 1
1 (1 a~i L p
) (1 a~ j L q n ( n 1)
) 1 (1 b~ L ) p (1 b~ L ) q n ( n 1)
i j
And
1 (1
n n 1 n n 1
i 1 j 1, j i
a~i
L p
) (1 a~ j L ) q n ( n 1) i 1 j 1, j i
1 (1 ~ L ) p (1 ~ L ) q n ( n 1)
bi bj
Then
1
pq

n n 1
1 1 (1 a~i L ) p (1 a~ j L ) q n( n1)

i 1 j 1, j i
1
n n 1 pq

1

1 (1 ~
bi
L p
) (1 b~ )
j
L q n ( n 1)

i 1 j 1, j i
So
1
pq

n n 1
1 1 1 (1 a~i L ) p (1 a~ j L ) q n( n1)

i 1 j 1, j i
1
n n 1 pq

1 1

1 (1 ~
bi
L p
) (1 b~ )
j
L q n ( n 1)

i 1 j 1, j i
18
Similarly,
1
pq

n n 1
1 1 1 (1 a~i U ) p (1 a~ j U ) q n( n 1)

i 1 j 1, j i
1
n n 1 pq

1 1

1 (1 ~
bi
) (1 b~ )
U p
j
U q n ( n 1)

i 1 j 1, j i
(5) Similar to (4), we can know
1
pq

n n 1
1 1 1 (1 a~i L ) p (1 a~ j L ) q n( n1)

i 1 j 1, j i
1
n n 1 pq

1 1

1 (1 ~
bi
L p
) (1 b~ )
j
L q n ( n 1)

i 1 j 1, j i
1
pq

n n 1
1 1 1 (1 a~i U ) p (1 a~ j U ) q n( n1)

i 1 j 1, j i
1
n n 1 pq

1 1

1 (1 ~
bi
U p
) (1 b~ )
j
U q n ( n 1)

i 1 j 1, j i
~ ~ ~
So, we can get IVNHFULBM p,q
(b1 , b2 ,..., bn ) ,
which completes the proof of the theorem 14.

Theorem 15 (Boundedness).
Let a~i (i 1,2,..., n) be a set of IVNHFULEs. If s min { s ( a~i ) } , s min{s ( a~i ) } ,
1 i n 1i n
s max{s ( a~i ) } , s max{s ( a~i ) } , r L min{ra~i L | [ra~i L , ra~i U ] t (a~i )} ,

1i n 1i n 1i n
r U min { ra~ i U | [ ra~ i L , ra~ i U ] t ( a~ i )} , r L max { ra~ i L | [ ra~ i L , ra~i U ] t ( a~ i )}

1 i n 1 i n
r U max{ra~i U | [ra~i L , ra~i U ] t (a~i )} , L min{ a~i L | [ a~i L , a~i U ] i (a~i )} ,

1 i n 1i n
U min { a~i U | [ a~i L , a~i U ] i ( a~ i )} , L

max { a~ i L | [ a~ i L , a~ i U ] i ( a~ i )}
1 i n 1 i n
U max{ a~i U | [ a~i L , a~i U ] i (a~i )} , L min{ a~i L | [ a~i L , a~i U ] f (a~i )} ,
1 i n 1i n
U min { a~i U | [ a~i L , a~i U ] f ( a~ i )} , L

max { a~ i L | [ a~ i L , a~i U ] f ( a~ i )}
1 i n 1 i n
U max{ a~i U | [ a~i L , a~i U ] f (a~i )} , for all i 1, 2 ,..., n

1 i n
a~ [ s , s ], ([r L , r U ], [ L , U ], [ L , L ]) and
a~ [s , s ], ([r L , r U ],[ L , U ],[ L , U ]) then
19
a~ IVNHFULBM p,q
(a~1 , a~2 ,..., a~n ) a~
Proof.
Based on above terms, we can get
(1) p q (a~i ) p (a~ j ) q p q
1 1 1
n n 1 pq n n 1 pq n n 1 pq
p q . (a~i ) p . (a~ j ) q p q
i 1 j 1, j i n(n 1) i 1 j 1, j i n(n 1) i 1 j 1, j i n(n 1)
1
n n 1 pq
. (a~i ) p . (a~ j ) q
i 1 j 1, j i n(n 1)
Similarly, we can know
1
n n 1 pq
(a~i ) p (a~ j ) q
i 1 j 1, j i n(n 1)
(2) (r L ) p q (ra~i ) p .(ra~ j ) q (r L ) p q

L L
1 1 1
(1 (r L ) p q ) n ( n 1) (1 (ra~i ) p .(ra~ j ) q ) n ( n 1) (1 (r L ) p q ) n ( n 1)
L L
n n 1 n n 1 n n 1
(1 (r L ) p q ) n ( n 1)
i 1 j 1, j i

i 1 j 1, j i
(1 (ra~i ) p .(ra~ j ) q ) n ( n 1)
L L
(1 (r L ) p q ) n ( n 1)
i 1 j 1, j i
1 1
n n 1 pq n n 1 pq
1 1
(1 (r L p q n ( n 1)
) ) (1 (r a~i
L
) p .(ra~ j ) )
L q n ( n 1)

i 1 j 1, j i i 1 j 1, j i
1
n n 1 pq
1 (1 (r L p q n ( n 1)
) )

i 1 j 1, j i
1
n n 1 pq
r L 1 (1 (r ) p .(ra~ j ) ) q n ( n 1) r L
L L

a~i
i 1 j 1, j i
Likewise, we also get

1
n n 1 pq
rU 1 (1 (r U q n ( n 1)
) p .(ra~ j ) ) rU
U

a~i
i 1 j 1, j i
(3) (1 L ) p q (1 a~i L ) p (1 a~ j L ) q (1 L ) p q
1 (1 1 (1 1 (1
n n n n n n
L pq L pq
) a~i
L p
) (1 a~ j L ) q )
i 1 j 1, j i i 1 j 1, j i i 1 j 1, j i
20
1 1
pq pq
1 (1
n n 1 n n 1
1 1 L p q n ( n 1)
) 1 1 1 (1 a~i L ) p (1 a~ j L ) q n ( n 1)

i 1 j 1, j i i 1 j 1, j i
1
pq
1 (1
n n 1
1 1 L p q n ( n 1)
)

i 1 j 1, j i
1
pq

n n 1
L 1 1 1 (1 a~i L ) p (1 a~ j L ) q n ( n 1) L

i 1 j 1, j i
Similarly,
1
pq

n n 1
U 1 1 1 (1 a~i U ) p (1 a~ j U ) q n ( n 1) U

i 1 j 1, j i
(4)Similar to the proof of the (3), the results can be got.
1
pq

n n 1
L 1 1 1 (1 a~i L ) p (1 a~ j L ) q n ( n 1) L

i 1 j 1, j i
1
pq

n n 1
U 1 1 1 (1 a~i U ) p (1 a~ j U ) q n ( n 1) U

i 1 j 1, j i
So, we can know a~ IVNHFULBM p,q

(a~1 , a~2 ,..., a~n ) a~

In the following, we will discuss some special cases of the IVNHFULBM operator with respect to
the different parameters p and q .
(1) If q 0 , then
1 1
1 n n p 1 n p
IVNHFULBM p ,0 (a~1 , a~2 ,..., a~n ) a~i p a~i p
n(n 1) i 1 j 1, j i n i 1
~ ~ ~
~ r ( a~1 ) t ( a~1 ), ~
r ( a~2 ) t ( a~2 ),..., ~r ( a~n ) t ( a~n )
1 1
n 1 p n 1 p

s n 1 ,s 1 ,
1
(1 ( ra~i ) p ) n , 1
L
(1 ( ra~i ) p ) n
U

,

1 ( a~i ) p p 1 ( a~i ) p p
n
i 1 i 1
i 1 n i 1 n
~ ~ ~
~ ( a~1 ) i ( a~1 ), ~ ( a~2 ) i ( a~2 )..., ~ ( a~n ) i ( a~n )
1 1
1 p 1p
n n n
n
1 1 1 (1 a~i ) ,1 1 1 (1 a~i U ) p
L
,
p

i 1
i 1

21
~ ~ ~
~ (a~1 ) f (a~1 ),~ (a~2 ) f (a~2 ),...,~ (a~n ) f (a~n )
1 1
1 p 1p
n n n
n
1 1 1 (1 a~i ) ,1 1 1 (1 a~i U ) p
L p

i 1
i 1

(2) If p 1 and q 0 ,then
1 n n 1 n
IVNHFULBM 1,0 (a~1 , a~2 ,..., a~n ) a~i a~i
n(n 1) i 1 j 1, j i n i 1

n 1 n 1
s n n 1 ,s
n n 1
( a~i ) ( a~i )

, ~ 1
r ( a~i )t ( a~i )
~ (1 r a~i L ) n , 1
(1 r a~i U ) n ,

i 1 i 1
i1 j 1, j i n i 1 j 1, j i
n

1 1 1 1
n n n n n n n n

( a~i )i ( a~i )
~ ~ a~i
L ,
a~i
U , ~
~ ( a~i ) f ( a~i ) a~i
L ,
a~i
U

i 1 i 1 i 1 i 1

1 1
(3) If p and q ,then
2 2
1 1
, 1 n n 1 1
IVNHFULBM 2 2 (a~1 , a~2 ,..., a~n ) a~i 2 a~ j 2
n(n 1) i 1 j 1, j i

s n n 1 1 , s n n 1 1 ,
1
( a~i ) 2 ( a~ j ) 2

1
( a~i ) 2 ( a~ j ) 2

i1 j 1, ji n ( n 1) i 1 j 1, j i
n ( n 1)

1 1
n n 1 n ( n 1) n n 1 n ( n 1)
~ ~
1
r ( a~ j )t ( a~ j )
r ( a~i )t ( a~i ), ~
~ (1 (r a~i L r a~ j L ) 2 ) , 1

(1 (r a~i U r a~ j U ) 2 ) ,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~ ~
(a~i ) i (a~i ), (a~ j ) i (a~ j )
1 1
n n 1 1 n( n 1) n n 1 1 n( n 1)

1 (1 ~ L ) 2 (1 ~ L ) 2
ai aj

,
1 (1 ~ U ) 2 (1 ~ U ) 2
ai aj

,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~ ~
(a~i ) f (a~i ), (a~ j ) f (a~ j )
1 1
n n 1 1 n( n1) n n 1 1 n( n 1)

1 (1 ~ L ) 2 (1 ~ L ) 2
ai aj

,
1 (1 ~ U ) 2 (1 ~ U ) 2
ai aj

i 1 j 1, j i i 1 j 1, j i

1
1 n n 2
IVNHFULBM 1,1 (a~1 , a~2 ,..., a~n ) a~i a~ j
n(n 1) i 1 j 1, j i
22

s 1 s

1 ,
n n 2 n n 2
1 ( a~i ) ( a~ j ) , 1 ( a~i ) ( a~ j )
i1 j 1, j i n ( n 1) i 1 j 1, j i
n ( n 1)

~ ~
~ r (a~i ) t (a~i ), ~ r (a~ j ) t (a~ j )
1 1
1 2 1 2
n n n ( n 1) n n n ( n 1)
1

(1 r ai r a j )
i 1 j 1, j i
~ L ~ L
, 1

(1 r
i 1 j 1, j i
a~i
U
r a~ j U ) ,

~ ~ ~ ~
(a~i ) i (a~i ), (a~ j ) i (a~ j )
1 1
1 2 1 2
n n(n1) n(n1)
1 (1
n n n
1 1 1 (1 a~i L )(1 a~ j L ) ,1 1 a~i
U
)(1 a~ j U ) ,
i 1 j 1, j i
i 1 j 1, j i

~ ~
~(a~ ) f (a~ ),~(a~ ) f (a~ )
i i j j
1 1
1 2 1 2
n(n1) n(n1)
1 (1 1 (1
n n n n
1 1 a~i
L
)(1 a~ j L ) ,1 1 a~i
U
)(1 a~ j U )

i 1 j 1, j i i 1 j 1, j i

3.2 The interval-valued neutrosophic hesitant fuzzy uncertain linguistic weighted
Bonferroni mean operator
~ ~ ~
IVNHFULEs, and p , q 0 , then an interval-valued neutrosophic hesitant fuzzy uncertain

linguistic weighted Bonferroni mean operator can be expressed as follows:
1
n n wi w j ~ p ~ q p q
IVNHFULWBM p ,q (a~1 , a~2 ,..., a~n ) ai a j (40)
i 1 j 1, j i 1 wi
~ ~
t (a~i ), i (a~i ), f (a~i )) (i 1,2,..., n) be the set of all IVNHFULEs,
1
IVNHFULWBM p,q
(a~1 , a~2 ,..., a~n ) ai a j
i 1 j 1, j i 1 wi

s 1 ,s
,
n n ww
1
pq n n wi w j ~ p ~ q p q
i j ( a~i ) p ( a~ j ) q
( ai ) ( a j )

1
i1 j 1, j i i w 1 w
i 1 j 1, j i i
23
~ ~
~ r (a~i ) t (a~i ), ~ r (a~ j ) t (a~ j )
1 1
pq pq
1 (r
n n wi w j n n wi w j
1 (41)
1 (ra~i L ) p (ra~ j L ) q 1 wi , 1 a~i
U
) p (ra~ j U ) q 1 wi

i 1 j 1, j i i 1 j 1, j i

~ ~ ~ ~
(a~i ) i (a~i ), (a~ j ) i (a~ j )
1 1
pq pq
1 (1
1 1 1 (1 a~i L ) p (1 a~ j L ) q 1 wi ,1 1 a~i
U
) p (1 a~ j U ) q 1 wi

i 1 j 1, j i i 1 j 1, j i

~ ~
~ (a~i ) f (a~i ),~ (a~ j ) f (a~ j )
1 1
pq pq
1 (1
1 1 1 (1 a~i L ) p (1 a~ j L ) q 1 wi ,1 1 a~i
U
) p (1 a~ j U ) q 1 wi

i 1 j 1, j i i 1 j 1, j i

The proof of Theorem 16 can be easily completed similar to Theorem 11.

The IVNHFULWBM operator satisfies the following properties:
Theorem 17 (Reducibility).
1 1 1
Let W ( , ,..., ) T be the weight vector of a~i (i 1,2,..., n ) , then
n n n
IVNHFULWBM p,q
(a~1 , a~2 ,..., a~n )
Proof.
1 1 1
Since W ( , ,..., )T is the weight vector of a~i (i 1,2,..., n) ,then according to definition
n n n
15,we have
1
IVNHFULWBM p,q
(a~1 , a~2 ,..., a~n ) ai a j
i 1 j 1, j i 1 wi
1
1 1 pq 1
n n n n q
1 p
n n a~i p a~ j q a~i p a~ j q
i 1 j 1, j i 1 1

n(n 1) i 1 j 1, j i
n
IVNHFULBM p,q
(a~1 , a~2 ,..., a~n )

Let a~ j [ S ( a ) , s ( a ) ], ([ ra L , ra U ], [ a L , a U ], [ a L , a U ]) a~ ( j 1,2,..., n) ,then
IVNHFULWBM p,q
(a~1 , a~2 ,..., a~n ) a~ .

If (a~1' , a~2 ' ,...,a~n ' ) be any permutation of (a~1 , a~2 ,..., a~n ) , then we can get
24
IVNHFULWBM p ,q
(a~1 ' , a~2 ' ,..., a~n ' ) IVNHFULWBM p,q
(a~1 , a~2 ,..., a~n ) .

~ ~ ~
Suppose (a~1 , a~2 ,..., a~n ) and (b1 , b2 ,..., bn ) are two sets of IVNHFULEs. If s ( a~i ) s (b~ ) ,
i
i i i i i i i
~ ~ ~
i 1,2,..., n ,then IVNHFULWBM p,q
(a~1 , a~2 ,..., a~n ) IVNHFULWBM p,q
(b1 , b2 ,..., bn )

1 i n 1i n

1i n 1i n 1i n
r U min { ra~i U | [ ra~i L , ra~i U ] t ( a~ i )} , r L max { ra~i L | [ ra~i L , ra~i U ] t ( a~ i )}

1 i n 1 i n

1 i n 1i n

max { a~i L | [ a~i L , a~i U ] i ( a~ i )}
1 i n 1 i n
1 i n 1i n

max { a~i L | [ a~i L , a~i U ] f ( a~ i )}
1 i n 1 i n

1 i n
a~ [ s , s ], ([r L , r U ], [ L , U ], [ L , L ]) and
a~ IVNHFULWBM p,q
(a~1 , a~2 ,..., a~n ) a~ .

In the following, we will discuss some specials of the IVNHFULWBM operator with respect to
(1) If q 0 , then
1 1
ww n n p n p
IVNHFULWBM p ,0
(a~1 , a~2 ,..., a~n ) i i a~i p wi a~i p
1 wi i 1 j 1, j i i 1

s n 1 ,s 1 ,
wi ( a~i ) p p wi ( a~i ) p p
n
i 1 i 1
25
~ ~ ~
~ r ( a~1 ) t ( a~1 ), ~
r ( a~2 ) t ( a~2 ),..., ~r ( a~n ) t ( a~n )
1 1
n p n p
1

(1 ( ra~i ) p ) wi , 1
L

(1 ( ra~i ) p ) wi
U

,

i 1 i 1

~ ~ ~
~ ( a~1 ) i (a~1 ), ~ ( a~2 ) i ( a~2 )..., ~ ( a~n ) i ( a~n )
1 1
p

n p n
1 1 1 (1 a~i L ) p ,1 1 1 (1 a~i U ) p ,
wi w i

i 1 i 1

~ ~ ~
~ (a~1 ) f (a~1 ),~ (a~2 ) f (a~2 ),...,~ (a~n ) f (a~n )
1 1
wi p

wi p
n n
1 1 1 (1 a~i ) L p ,1 1 1 (1 a~i )
U p

i 1 i 1

n
IVNHFULWBM 1,0 (a~1 , a~2 ,..., a~n ) wi a~i
i 1

n n
s
n , s
n
, ~ 1
r ( ai )t ( ai )
~ ~ ~ (1 r a~i L ) wi , 1
(1 r U wi
a~i ) ,

i1 wi ( ~
ai )

wi ( ~
a i )
i 1 i 1
i 1
n n n n
~
( ai )i ( a~i )
~ ~

i 1
( a~i L ) wi ,
i 1
( a~i U ) wi , ~
~ ( a~i ) f ( a~i )

i 1
( a~i L ) wi , (
i 1
a~i )
U wi

1 1
(3) If p and q , then
2 2
n n wi w j ~ 1 ~ 1
1 1
,
IVNHFULWBM 2 2 (a~1 , a~2 ,..., a~n ) ai 2 a j 2
i 1 j 1, j i 1 wi

s n n 1 1 , s n n 1 1 ,
ww ww
i j ( a~i ) 2 ( a~ j ) 2 i j ( a~i ) 2 ( a~ j ) 2
i1 j 1, j i 1 wi i 1 j 1, j i
1 wi

~ ~
~ r (a~i ) t (a~i ), ~ r (a~ j ) t (a~ j )
wi w j wi w j
n n 1 1 w n n 1 1 w
1
, 1
,
i i
(1 (r a~i L r a~ j L ) 2 ) (1 (r a~i U r a~ j U ) 2 )

i 1 j 1, j i i 1 j 1, j i

26
~ ~ ~ ~
(a~i ) i (a~i ), (a~ j ) i (a~ j )
wi w j wi w j
n n 1 1 1 w n n 1 1 1 w
1 (1 ~ L ) 2 (1 ~ L ) 2 ,
1 (1 ~ U ) 2 (1 ~ U ) 2 ,
i i
ai a j ai aj

i 1 j 1, j i i 1 j 1, j i

~ ~
~(a~ ) f (a~ ),~(a~ ) f (a~ )
i i j j
wi w j wi w j
n n 1 1 1 w n n 1 1 1 w
1 (1 ~ L ) 2 (1 ~ L ) 2 ,
1 (1 ~ U ) 2 (1 ~ U ) 2
i i
ai aj
ai aj

j 1, j i j 1, j i
i 1
i 1

(4) If p 1 and q 1 , then

1
n n wi w j ~ ~ 2
IVNHFULWBM 1,1 (a~1 , a~2 ,..., a~n ) ai a j
i 1 j 1, j i 1 wi

s s
1 ,
n n ww 2
1
2 n n wi w j
i j
( a~i ) ( a~ j ) ,

( a~i ) ( a~ j )

i1 j 1, j i 1 wi i 1 j 1, j i
1 wi

~ ~
~ r (a~i ) t (a~i ), ~ r (a~ j ) t (a~ j )
1 1
wi w j 2 wi w j 2
n n 1 wi n n 1 wi
1

(1 r a~i r a~ j )
L L
, 1
(1 r a~i
U
r a~ j U ) ,

i 1 j 1, j i i 1 j 1, j i

~ ~ ~ ~
(a~i ) i (a~i ), (a~ j ) i (a~ j )
1 1
2 2
1 (1
1 1 1 (1 a~i )(1 a~ j ) i
L L 1 w ,1 1
a~i
U
)(1 a~ j U ) 1wi ,

i 1 j 1, j i i 1 j 1, j i

~ ~
~(a~ ) f (a~ ),~(a~ ) f (a~ )
i i j j
1 1
2 2
1 (1 1 (1
1 1 a~i
L
)(1 a~ j L ) 1wi ,1 1
a~i
U
)(1 a~ j U ) 1wi

i 1 j 1, j i i 1 j 1, j i

3.3 The interval-valued neutrosophic hesitant fuzzy uncertain linguistic
geometric Bonferroni mean operator
~ ~ ~
IVNHFULEs, and p , q 0 , then an interval-valued neutrosophic hesitant fuzzy uncertain

linguistic geometric Bonferroni mean operator can be defined as follows:
1
1 n n
IVNHFULGBM p , q (a~1 , a~2 ,..., a~n ) ( pa~i qa~ j ) n ( n 1) (42)
p q i 1 j 1, j i
~ ~
27
1
1 n n
IVNHFULGBM p , q (a~1 , a~2 ,..., a~n ) ( pa~i qa~ j ) n ( n 1)
p q i 1 j 1, j i

s 1 n n , s ,
p ( a~i ) q ( a~ j ) n ( n1) p ( a~i ) q ( a~ j ) n ( n1)
1 1 n n 1
p q i1 j 1, j i p q i 1 j 1, j i
1
pq

n n 1
~ 1 1 1 (1 ra~i L ) p (1 ra~ j L ) q n ( n 1) ,
r ( a~i )t ( a~i )
~
r ( a~ j )t ( a~ j )
~ ~ i 1 j 1, j i

1
pq

n n 1
1 1 1 (1 ra~i U ) p (1 ra~ j U ) q n ( n 1) , (43)

i 1 j 1, j i

1 1
pq pq

n n 1 n n 1
1 1 ( a~i L ) p ( a~ j L ) q n ( n 1) , 1 1 ( a~i U ) p ( a~ j U ) q n ( n 1) ,
( a~i )i ( a~i )
~ ~
( a~ j )i ( a~ j )
~ ~ i 1 j 1, j i i 1 j 1, j i

1 1
pq pq

n n 1 n n 1
~ 1 1 ( a~i L ) p ( a~ j L ) q n ( n 1) , 1 1 ( a~i U ) p ( a~ j U ) q n ( n 1)
~ ( a~i ) f ( a~i )
~ ( a~ j ) f ( a~ j )
~ i 1 j 1, j i i 1 j 1, j i

The IVNHFULGBM operator satisfies the following properties:
Let a~ j [ S ( a ) , s ( a ) ], ([ ra L , ra U ], [ a L , a U ], [ a L , a U ]) a~ ( j 1,2,..., n) ,then
IVNHFULGBM p,q
(a~1 , a~2 ,..., a~n ) a~ .

IVNHFULGBM p,q
(a~1 ' , a~2 ' ,..., a~n ' ) IVNHFULGBM p,q
(a~1 , a~2 ,..., a~n ) .

~ ~ ~
i
i i i i i i i
i 1,2,..., n ,then
~ ~ ~
IVNHFULGBM p,q
(a~1 , a~2 ,..., a~n ) IVNHFULGBM p,q
(b1 , b2 ,..., bn )
28
1 i n 1i n

1i n 1i n 1i n

1 i n 1 i n

1 i n 1i n

1 i n 1 i n
1 i n 1i n

1 i n 1 i n

1 i n
a~ [ s , s ], ([r L , r U ], [ L , U ], [ L , L ]) and
a~ IVNHFULGBM p ,q
(a~1 , a~2 ,..., a~n ) a~ .

In the following, we will discuss some specials of the IVNHFULGBM operator with respect to the
different parameters p and q .
(1) If q 0 , then
1
1 n
IVNHFULGBM p ,0
(a~1 , a~2 ,..., a~n ) ( pa~i ) n
p i 1
1 1
p

p
s n
n 1 n 1
1 ,s 1 ,
( a~ ) n n ( a~ ) n ~ 1 1 1 (1 ra~i L ) p n ,1 1 1 (1 ra~i U ) p n ,
r ( a~i )t ( a~i )
~
i1 i i 1
i
r ( a~ j )t ( a~ j )
~ ~ i 1 i 1

1 1
p

n 1 p n 1
1 1 ( a~i L ) p n , 1 1 ( a~i U ) p n ,
( ai )i ( ai )
~ ~ ~ ~
( a~ j )i ( a~ j )
~ ~ i 1 i 1

1 1
p

n 1 p n 1
~ 1 1 ( a~i L ) p n , 1 1 ( a~i U ) p n
~ ( a~i ) f ( a~i )

~ ( a~ j ) f ( a~ j )
~ i 1 i 1

29
n 1
IVNHFULGBM 1,0 (a~1 , a~2 ,..., a~n ) a~i n
i 1

s n
r , r
1 ,s 1 ,
n 1 n 1
( a~ ) n n ( a~ ) n ~ L n U n ,
a~i a~i
i1 i r ( a~i )t ( a~i )
~
r ( a~ )t ( a~ )
i
i 1
~ ~ i 1 i 1
j j
1 ,1 1 1 ,1 1
n 1 n 1 n 1 n 1
~ 1 L n U n , ~ 1 L n U n
a~i a~i a~i a~i
( a~i )i ( a~i ) ~ ( a~i ) f ( a~i )
~
~ ~
( a~ )i ( a~ )
i 1 i 1 ~ ~ ~ ~ i 1 i 1
j j ( a j ) f ( a j )
1 1
2 2
1 1 1
, n n1 1
IVNHFULGBM ~ , a~ ,..., a~ )
2 2 (a
1 2 n ( a~i a~ j ) n ( n 1)
i 1 j 1, j i 2 2

s 1 ,s 1 ,
( a~i ) 1 ( a~ j ) n ( n1) 1 ( a~i ) 1 ( a~ j ) n ( n1)
n n 1 n n
i1 j 1, j i 2 2 i 1 j 1, j i 2 2
1 1
n n 1 1 n ( n 1) n n 1 1 n ( n 1)
~
r ( a~i )t ( a~i )
~ 1 (1 r~ L ) 2 (1 r~ L ) 2

j 1, j i
ai aj

, 1 (1 r~ U ) 2 (1 r~ U ) 2
ai aj

,

r ( a~ j )t ( a~ j )
~ ~ i 1 i 1 j 1, j i

1 1
n n 1 n ( n 1) n n 1 n ( n 1)

( a~i )i ( a~i )
~ ~
1
1 ( ~ L ~ L ) 2

j 1, j i
ai aj

,1 1 ( ~ ~ ) 2
ai
U
aj
U

,

( a~ j )i ( a~ j )
~ ~ i 1 i 1 j 1, j i

1 1
n n 1 n ( n 1) n n 1 n ( n 1)
~ 1
~ ( a~i ) f ( a~i )
1 ( ~ L ~ L ) 2

j 1, j i
ai aj

,1 1 ( ~ U ~ U ) 2
ai aj

~ ( a~ j ) f ( a~ j )
~ i 1 i 1 j 1, j i

1
1 n n
IVNHFULGBM 1,1
(a~1 , a~2 ,..., a~n ) (a~i a~ j ) n ( n 1)
2 i 1 j 1, j i

s 1 n n , s ,
( a~i ) ( a~ j ) n ( n1) ( a~i ) ( a~ j ) n ( n1)
1 1 n n 1
2 i1 j 1, ji 2 i 1 j 1, j i
1 1
2 2
1 (1 r 1 (1 r
n n 1 n n 1
~ 1 1 a~i
L
)(1 ra~ j L ) n( n 1) ,1 1
a~i
U
)(1 ra~ j U ) n ( n 1) ,
r ( a~i )t ( a~i )
~
r ( a~ j )t ( a~ j )
~ ~ i 1 j 1, j i i 1 j 1, j i

30
1 1
2 2

n n 1 n n 1
1 1 a~i L a~ j L n ( n 1) , 1 1 a~i U a~ j U n ( n 1) ,
( a~i )i ( a~i )
~ ~
( a~ j )i ( a~ j )
~ ~ i 1 j 1, j i i 1 j 1, j i

1 1
2 2

n n 1 n n 1
~ 1 1 a~i L a~ j L n ( n 1) , 1 1 a~i U a~ j U n ( n 1)
~ ( a~i ) f ( a~i )
~ ( a~ j ) f ( a~ j )
~ i 1 j 1, j i i 1 j 1, j i

3.4 The interval-valued neutrosophic hesitant fuzzy uncertain linguistic weighted
geometric Bonferroni mean operator
~ ~ ~
IVNHFULEs, and p , q 0 , then the interval-valued neutrosophic hesitant fuzzy uncertain

linguistic weighted geometric Bonferroni mean operator can be defined as follows:
wi w j
1 n n
IVNHFULWGBM p,q
(a~1 , a~2 ,..., a~n ) ( pa~i qa~ j ) 1 wi (44)
p q i 1 j 1, j i
~ ~
wi w j
1 n n
IVNHFULWGBM p,q
(a~1 , a~2 ,..., a~n ) ( pa~i qa~ j ) 1 wi
p q 1 j 1, j i
i

s 1 n n wi w j , s
,
p ( a~i ) q ( a~ j ) 1wi p ( a~i ) q ( a~ j ) 1wi
wi w j
1 n n
p q i1 j 1, ji p q i 1 j 1, j i
1
pq
1 (1 r
n n wi w j
~ 1 1 a~i
L p
) (1 ra~ j L ) q 1 wi ,
r ( a~i )t ( a~i )
~
r ( a~ j )t ( a~ j )
~ ~ i 1 j 1, j i

1
pq
1 (1 r
n n wi w j
1 1 a~i
U
) p (1 ra~ j U ) q 1 wi , (45)

i 1 j 1, j i

1 1
pq pq
1 ( 1 (
1 a~i ) ( a~ j L ) q
L p 1 wi , 1 a~i
U
) p ( a~ j U ) q 1 wi ,
( a~i )i ( a~i )
~ ~
( a~ j )i ( a~ j )
~ ~ i 1 j 1, j i i 1 j 1, j i

1 1
pq pq
1 ( 1 (
~ 1 a~i ) ( a~ j L ) q
L p 1 wi , 1 a~i
U
) p ( a~ j U ) q 1 wi
~ ( a~i ) f ( a~i )
~ ( a~ j ) f ( a~ j )
~ i 1 j 1, j i i 1 j 1, j i

The IVNHFULWGBM operator satisfies the following properties:
Theorem 28 (Reducibility).
31
1 1 1
Let W ( , ,..., ) T be the weight vector of a~i (i 1,2,..., n ) , then
n n n
IVNHFULWGB M p,q
( a~1 , a~2 ,..., a~n ) IVNHFULGBM p,q
(a~1 , a~2 ,..., a~n )

Let a~ j [ S ( a ) , s ( a ) ], ([ ra L , ra U ], [ a L , a U ], [ a L , a U ]) a~ ( j 1,2,..., n) ,
then
IVNHFULWGB M p,q
(a~1 , a~2 ,..., a~n ) a~ .

IVNHFULWGB M p,q
(a~1 ' , a~2 ' ,..., a~n ' ) IVNHFULWGB M p,q
(a~1 , a~2 ,..., a~n ) .

~ ~ ~
i
i i i i i i i
i 1,2,..., n ,then
~ ~ ~
IVNHFULWGB M p,q
(a~1 , a~2 ,..., a~n ) IVNHFULWGB M p,q
(b1 , b2 ,..., bn )

Let a~i (i 1,2,..., n) be a collection of IVNHFULEs. If s min { s ( a~i ) } , s min{s ( a~i ) } ,
1 i n 1i n

1i n 1i n 1i n

1 i n 1 i n

1 i n 1i n

1 i n 1 i n
1 i n 1i n

1 i n 1 i n

1 i n
a~ [ s , s ], ([r L , r U ], [ L , U ], [ L , L ]) and
32
a~ IVNHFULWGB M p,q
(a~1 , a~2 ,..., a~n ) a~ .

In the following, we will discuss some specials of the IVNHFULWGBM operator with respect to
(1) If q 0 , then
1 n
IVNHFULWGBM p ,0
(a~1 , a~2 ,..., a~n ) ( pa~i ) wi
p i 1
1 1
p
1 (1 r 1 (1 r
n p n
s n ~ wi , s n ~ wi , ~ 1 1 a~i
L p wi
) ,1 1
a~i
U
) p wi ,
i1 ( ai ) ( ai ) r ( a~i )t ( a~i )
~
i 1 r ( a~ j )t ( a~ j )
~ ~ i 1 i 1

1 1
p
1 ( 1 (
n p n
1 a~i
L p wi
) , 1
a~i
U
) p wi ,
( a~i )i ( a~i )
~ ~
( a~ j )i ( a~ j )
~ ~ i 1 i 1

1 1
p
1 ( 1 (
n p n
~ 1 a~i
L p wi
) , 1
a~i
U
) p wi
~ ( a~i ) f ( a~i )

~ ( a~ j ) f ( a~ j )
~ i 1 i 1

n
IVNHFULWGBM 1,0 (a~1 , a~2 ,..., a~n ) a~i wi
i 1

r , r
n n
s n ~ wi , s n ~ wi , ~ L wi U wi ,
a~i a~i
i1 ( ai ) r ( a~i )t ( a~i )
( ai )

~
r ( a~ )t ( a~ )
i 1
~ ~ i 1 i 1
j j
1 1

1 1
n n n n
~ 1 L wi
,1 U wi , ~ 1 L wi
,1 U wi
a~i a~i a~i a~i
( ai )i ( a~i )
~ ~ ( ai ) f ( ai )
~ ~ ~
~ ~ ~ ~ i 1 i 1 ~ ~ ~ ~ i 1 i 1
( a )i ( a )
j j
( a j ) f ( a j )
1 1
2 2
1 1 wi w j
, n 1n1
IVNHFULWGBM ~ , a~ ,..., a~ )
2 2 (a
1 2 n ( a~i a~ j ) 1 wi
i 1 j 1, j i 2 2

s wi w j , s wi w j ,
n n 1 ( a~i ) 1 ( a~ j ) 1wi n n 1 ( a~i ) 1 ( a~ j ) 1wi
i1 j 1, ji 2 2 i 1 j 1, j i 2 2
33
wi w j wi w j
n n 1 1 1 w n n 1 1 1 w
~ 1 (1 r~ L ) 2 (1 r~ L ) 2
1 (1 r~ U ) 2 (1 r~ U ) 2 ,
i i
,
r ( a~i )t ( a~i )
~ ai aj
ai aj

i 1 j 1, j i i 1 j 1, j i
r ( a~ j )t ( a~ j )
~ ~

wi w j wi w j
n n 1 1 w n n 1 1 w
~ 1 1 ( ~ L ~ L ) 2
1 ( ~ U ~ U ) 2 ,
i i
,1
( a~i )i ( a~i )
~ ai aj
ai aj

i 1 j 1, j i i 1 j 1, j i
( a~ j )i ( a~ j )
~ ~

wi w j wi w j
n n 1 1 w n n 1 1 w
~ 1 1 ( ~ L ~ L ) 2
1 ( ~ U ~ U ) 2
i i
,1
~ ( a~i ) f ( a~i ) ai aj
ai aj

i 1 j 1, j i i 1 j 1, j i
~ ( a~ j ) f ( a~ j )
~

wi w j
1 n n
IVNHFULWGBM 1,1
(a~1 , a~2 ,..., a~n ) (a~i a~ j ) 1 wi
2 i 1 j 1, j i

s 1 n n ww ,s
,
( a~i ) ( a~ j ) 1i wij 1 n n ( a~i ) ( a~ j ) 1i wij
ww
2 i1 j1, ji 2 i 1 j 1, j i
1 1
2 2
1 (1 r 1 (1 r
~ 1 1 a~i
L
)(1 ra~ j L ) 1 wi ,1 1
a~i
U
)(1 ra~ j U ) 1 wi ,
r ( a~i )t ( a~i )
~
r ( a~ j )t ( a~ j )
~ ~ i 1 j 1, j i i 1 j 1, j i

1 1
2 2
1 1
1 a~i
L
a~ j L 1 wi , 1
a~i
U
a~ j U 1 wi ,
( a~i )i ( a~i )
~ ~
( a~ j )i ( a~ j )
~ ~ i 1 j 1, j i i 1 j 1, j i

1 1
2 2
1 1
~ 1 a~i
L
a~ j L 1 wi , 1
a~i
U
a~ j U 1 wi
~ ( a~i ) f ( a~i )
~ ( a~ j ) f ( a~ j )
~ i 1 j 1, j i i 1 j 1, j i

4. Two multiple attribute decision making methods based on
IVNHFULWBM operator and IVNHFULWGBM operator

In a multiple attribute decision making problem with interval-valued neutrosophic hesitant
uncertain linguistic information, let A A1 , A2 ,..., Am be a discrete set of alternatives,
C C1 , C 2 ,..., C n be a set of attributes. The evaluation values can be represented by the form of
interval-valued neutrosophic hesitant uncertain linguistic
elements
nij s ( nij ) , s ( nij ) ~ ~ ~
, t (nij ), i (nij ), f (nij ) ,
~
t (nij ) ~
r (nij ) | ~
~
r (nij ) t (nij ) ,
~ ~
~ ~ ~
~

i (nij ) (nij ) | (nij ) i (nij ) , f (nij ) ~ (nij ) | ~ (nij ) f (nij ) . Therefore, we can get the decision
matrix N (nij ) mn . The weighting vector of attributes is given by W ( w1 , w2 ,..., wn )T with w j 0 ,
34
n
w j 1 . Then, the steps of the decision making method are described as follows:
j 1
Step1: Utilized the IVNHFULWBM operator and IVNHFULWGBM operator to derive the
overall assessment values N i (i 1,2,..., n) of the alternatives Ai (i 1,2,..., n) .
Step2: Calculate the score values of the overall assessment values N i (i 1,2,..., m) by the Eq.(33).
Step3: Rank all the alternatives A1 , A2 ,..., An by score values S ( N i ) , and select the best
alternative.
Step4: End.
5. A numerical example
In this section, we give a numerical example to demonstrate the MADM methods based on
IVNHFULWBM and IVNHFULWGBM operators.
Suppose that there is a company, which wants to invest a sum of money to an industry. There are
four companies as alternatives to be chosen, including: (1) A1 is a car company; (2) A2 is a food
company; (3) A3 is a computer company; (4) A4 is an arms company. There are three evaluation
attributes, including: (1) C1 is the risk; (2) C 2 is the growth; (3) C3 is the environment. We can
know the attributes C1 and C 2 are benefit criteria, and the type of C3 is cost criteria. The weight
vector of the attribute is w {0.25,0.35,0.4} .The final evaluation values are expressed by the
IVNHFULEs, and shown in Table 1.
Table 1 Interval-valued neutrosophic hesitant fuzzy uncertain linguistic decision matrix
C1 C2 C3
A1 <[S4,S5],([0.3,0.4],[0.4,0.4]),( <[S3,S5],([0.4,0.5],[0.5,0.6]),([0 <[S4,S5],([0.2,0.3]),([0.1,0.2]
[0.1,0.2]),([0.3,0.4])> .2,0.3],[0.3,0.3]),([0.3,0.4])> ,[0.4,0.5]),([0.5,0.6])>
A2 <[S3,S5],([0.6,0.7]),([0.1,0.2] <[S2,S4],([0.6,0.7]),([0.1,0.1]),([ <[S3,S5],([0.6,0.7]),([0.1,0.2]
),([0.1,0.2],[0.2,0.3])> 0.2,0.3])> ),([0.1,0.2])>
A3 <[S2,S3],([0.3,0.4],[0.5,0.6]), <[S2,S4],([0.5,0.6]),([0.2,0.3]),([ <[S3,S5],([0.5,0.6]),([0.1,0.2]
([0.2,0.4]),([0.2,0.3])> 0.3,0.4])> ,[0.2,0.3]),([0.2,0.3])>
A4 <[S1,S3],([0.7,0.8]),([0.2,0.3] <[S2,S5],([0.6,0.7]),([0.3,0.3]), <[S2,S5],([0.3,0.5],[0.2,0.3]),
),([0.1,0.2])> ([0.2,0.2])> ([0.1,0.2]),([0.3,0.3])>
5.1 Procedure of decision making method based on the IVNHFULWBM operator

Step1: Utilized the IVNHFULWBM operator to derive the overall assessment values
N i (i 1,2,..., n) of the alternatives Ai (i 1,2,..., n) . We can get
N1=<[S2.90,S4.61],([0.2853,0.3869],[0.3088,0.4118],[0.3175,0.3869],[0.3429,0.4118]),([0.1255
,0.2268],[0.2393,0.3431],[0.1482,0.2268],[0.2689,0.3431]),([0.3768,0.4783])>
N2=<[S2.71,S4.72],([0.6000,0.7000]),([0.1000,0.1714]),([0.1255,0.2268],[0.1623,0.2629])>
N3=<[S2.34,S3.97],([0.4258,0.5274],[0.5000,0.6000]),([0.1623,0.2988],[0.2000,0.3352]),([0.22
68,0.3273])>
35
N4=<[S1.61,S4.24],([0.5107,0.6569],[0.4644,0.5743]),([0.1882,0.2692]),([0.2021,0.2373])>
Step2: Calculate the score values of the overall assessment values N i (i 1,2,..., m) , which are
shown as follows (for convenience, we can use the subscript):
S(N1)=2.1135, S(N2)=2.8716, S(N3)=2.0908, S(N4)= 2.0554
Step3: Rank all the alternatives A1 , A2 ,..., An by score values S ( N i ) , and select the best
alternative.
So, A2 A1 A3 A4 , and the best alternative is A2 .
Step4: End.
5.2 Procedure of decision making method based on the IVNHFULWGBM
operator
Step1: Utilized the IVNHFULWGBM operator to derive the overall evaluation values
N i (i 1,2,..., n) of the alternatives Ai (i 1,2,..., n) . We can get
N1=<[S2.95,S4.62],([0.2900,0.3911],[0.3164,0.4189],[0.3276,0.3911],[0.3557,0.4189]),([0.1246
,0.2260],[0.2174,0.3249],[0.1452,0.2260],[0.2496,0.3249]),([0.3678,0.4697])>
N2=<[S2.71,S4.72],([0.6000,0.7000]),([0.1000,0.1704]),([0.1246,0.2260],[0.1583,0.2601])>
N3=<[S2.36,S4.01],([0.4330,0.5343],[0.5000,0.6000]),([0.1583,0.2904],[0.2000,0.3333]),([0.22
60,0.3266])>
N4=<[S1.64,S4.28],([0.5411,0.6751],[0.5128,0.6223]),([0.1822,0.2601]),([0.1914,0.2346])>
Step2: Calculate the score values of the overall assessment values N i (i 1,2,..., m) , which are
shown as follows (for convenience, we can use the subscript):
S(N1)=2.1648, S(N2)=2.8975, S(N3)=2.1183, S(N4)=2.1222
Step3: Rank all the alternatives A1 , A2 ,..., An by score values S ( N i ) ,and select the best
alternative.
So, A2 A1 A4 A3 , and the best alternative is A2 .
Step4: End.
In this section, we propose two methods based on IVNHFULWBM operator and
IVNHFULWGBM operator to solve the multiple attribute decision making problems in interval-valued
neutrosophic hesitant fuzzy uncertain linguistic environment. Finally, we get the same best alternative.
At the same time, we can find that the main advantages of the developed methods in this paper are that
not only accommodating the interval-valued hesitant uncertain linguistic environment but also
considering the interrelationship of the input arguments.
5.3 Analysis of the influence of the parameters p and q

In order to demonstrate the influence of the parameters p and q on decision making results, we
36
use the different values p and q in IVNHFULWBM operator and IVNHFULWGBM operator to rank
the alternatives. The ranking results are shown in Table 2 and Table 3.
Table 2 Ordering of the alternatives by utilizing the different p, q in IVNHFULWBM operator

p, q Score function S ( N i ) Ranking
1 S(N1)=2.0983,S(N2)=2.8630, A2 A1 A3 A4
pq
2 S(N3)=2.0805,S(N4)=2.0286
p q 1 S(N1)=2.1135, S(N2)=2.8716, A2 A1 A3 A4
S(N3)=2.0908, S(N4)=2.0554
p 1, q 0 S(N1)=2.1706, S(N2)=2.9115, A2 A1 A4 A3
S(N3)=2.1564, S(N4)=2.1570
p 0, q 1 S(N1)=2.2300, S(N2)=2.8692, A2 A1 A4 A3
S(N3)=2.1141, S(N4)=2.1670
p 1, q 2 S(N1)=2.1601, S(N2)=2.8801, A2 A1 A4 A3
S(N3)=2.1084, S(N4)=2.1095
p 2, q 1 S(N1)=2.1435, S(N2)=2.8922, A2 A1 A3 A4
S(N3)=2.1239, S(N4)=2.1081
p 2, q 2 S(N1)=2.1473, S(N2)=2.8886, A2 A1 A4 A3
S(N3)=2.1115, S(N4)=2.1149
p 5, q 5 S(N1)=2.2531, S(N2)=2.9360, A2 A4 A1 A3
S(N3)=2.1694, S(N4)=2.2846
p 10, q 0 S(N1)=2.6175, S(N2)=3.0393, A2 A1 A4 A3
S(N3)=2.4994, S(N4)=2.5667
p 0, q 10 S(N1)=2.6442, S(N2)=3.0184, A2 A1 A4 A3
S(N3)=2.4619, S(N4)=2.5649
p 15, q 15 S(N1)=2.4632, S(N2)=3.0276, A2 A4 A1 A3
S(N3)=2.2789, S(N4)=2.5187
Table 3 Ordering of the alternatives by utilizing the different p, q in IVNHFULWGBM operator

p, q Score function S ( N i ) Ranking
1 S(N1)=2.1727,S(N2)=2.9003, A2 A1 A4 A3
pq
2 S(N3)=2.1208,S(N4)=2.1310
p q 1 S(N1)=2.1648, S(N2)=2.8975, A2 A1 A4 A3
S(N3)=2.1183, S(N4)=2.1222
p 1, q 0 S(N1)=2.0379, S(N2)=2.8962, A2 A3 A1 A4
S(N3)=2.0850, S(N4)=1.1963
p 0, q 1 S(N1)=2.0995, S(N2)=2.8501, A2 A1 A3 A4
S(N3)=2.0462, S(N4)=1.9921
p 1, q 2 S(N1)=2.1492, S(N2)=2.8819, A2 A1 A3 A4
S(N3)=2.1009, S(N4)=2.0921
p 2, q 1 S(N1)=2.1283, S(N2)=2.8979, A2 A1 A3 A4
S(N3)=2.1139, S(N4)=2.0855
p 2, q 2 S(N1)=2.1477, S(N2)=2.8915, A2 A1 A3 A4
S(N3)=2.1125, S(N4)=2.1028
37
p 5, q 5 S(N1)=2.1072, S(N2)=2.8783, A2 A1 A3 A4
S(N3)=2.0961, S(N4)=2.0574
p 10, q 0 S(N1)=1.8177, S(N2)=2.8326, A2 A3 A4 A1
S(N3)=1.9907, S(N4)=1.8845
p 0, q 10 S(N1)=1.8639, S(N2)=2.7869, A2 A3 A4 A1
S(N3)=1.9555, S(N4)=1.8948
p 15, q 15 S(N1)=2.2248, S(N2)=3.4678, A2 A4 A3 A1
S(N3)=2.3038, S(N4)=2.4967
As we can see from Table 2 and Table 3, the ranking of the alternatives may be different for the
different values p, q in IVNHFULWBM operator and IVNHFULWGBM operator. But the best
alternative is the same although the different values p, q . Thus, the organization can properly select
the desirable alternative according to his interest and the actual needs. In general, we can get p q 1
1
or p q .
2
6. Conclusions
Interval-valued neutrosophic hesitant fuzzy uncertain linguistic set (IVNHFULS) can be as an
important tool of describing the decision makers preference and handling the hesitant, uncertain,
incomplete information, which plays a vital role in decision making. Bonferroni mean (BM) can
replace the simple averaging of other mean type operators because of the excellent modeling capability.
In real decision making, how to flexibly and rationally deal with input arguments is a vital problem. In
order to have the advantages of both IVNHFULS and BM, we extend BM operator to interval-valued
neutrosophic hesitant fuzzy uncertain linguistic environment. Hence, we finally give some
interval-valued neutrosophic hesitant fuzzy uncertain linguistic BM aggregation operators, such as
operator, interval-valued neutrosophic hesitant fuzzy uncertain linguistic weighted Bonferroni mean
(IVNHFULWBM) operator, interval-valued neutrosophic hesitant fuzzy uncertain linguistic geometric
Bonferroni mean (IVNHFULGBM) operator, interval-valued neutrosophic hesitant fuzzy uncertain
linguistic weighted geometric Bonferroni mean (IVNHFULWGBM) operator. It is essential to verify
the effectiveness of IVNHFULWBM operator and IVNHFULWGBM operator, so we apply the two
operators to deal with a numerical example. To get the desirable alternative decision maker wanted, we
change the values of p, q according to their interest and actual need.
In the further research, we should dug deeper into the applications of the proposed methods and
constantly improve the multiple attribute decision making method.
Acknowledgment
This paper is supported by the National Natural Science Foundation of China (No. 71271124), the
Humanities and Social Sciences Research Project of Ministry of Education of China
(No.10YJA630073 and No.09YJA630088), the Natural Science Foundation of Shandong Province
(No. ZR2011FM036), and Graduate education innovation projects in Shandong Province
(SDYY12065).
38
Compliance with Ethical Standards
(1) Disclosure of potential conflicts of interest
We declare that we do have no commercial or associative interests that represent a conflict of
interests in connection with this manuscript. There are no professional or other personal interests that
can inappropriately influence our submitted work.
(2) Research involving human participants and/or animals
This article does not contain any studies with human participants or animals performed by any of
the authors.
References
[1] L. A. Zadeh, Fuzzy sets, information and control 8(1965) 338353.
[2] K.T. Atanassov, Intuitionistic fuzzy sets, Fuzzy Sets and Systems 20(1986) 87-96.
[3] F. Smarandache, A unifying field in logics. neutrosophy: Neutrosophic probability, set and logic,
American Research Press, Rehoboth, 1999.
[4] H. Wang, F. Smarandache, Y.Q. Zhang and R. Sunderraman, Interval neutrosophic sets and logic,
Theory and applications in computing, Hexis, Phoenix, AZ, 2005.
[5] H. Wang, F. Smarandache, Y.Q. Zhang and R. Sunderraman, Single valued neutrosophic sets,
Multispace and Multistructure 4 (2010), 410413.
[6] V. Torra, Hesitant fuzzy sets, International Journal of Intelligent System 25(6)(2010) 529539.
[7] V. Torra and Y. Narukawa, On hesitant fuzzy sets and decision, In Proceedings of the IEEE
International Conference on Fuzzy Systems, pp. 13781382, August 2009.
[8] N. Chen, Z.S. Xu and M.M. Xia, Interval-valued hesitant preference relations and their applications
to group decision making, Knowledge-Based Systems 37 (2013) 528540.
[9] J. Ye, Multiple-attribute decision-making method under a single-valued neutrosophic hesitant fuzzy
environment, J Intell Syst (2015), DOI:10.1515/jisys-2014-0001.
[10] P.D. Liu, Some Generalized Dependent Aggregation Operators with Intuitionistic Linguistic
Numbers and Their Application to Group Decision Making, Journal of Computer and System
Sciences 79(1) (2013) 131143.
[11] P.D. Liu, F. Teng, An extended TODIM method for multiple attribute group decision-making based
on 2-dimension uncertain linguistic variable, Complexity (2014), DOI: 10.1002/cplx.21625.
[12] P.D. Liu, F. Jin, Methods for Aggregating Intuitionistic Uncertain Linguistic variables and Their
Application to Group Decision Making, Information Sciences205 (2012) 5871.
[13] L. A. Zadeh, The concept of a linguistic variable and its application to approximate reasoning. I,
Information Sciences 8(1975) 199249.
[14] Z. S. Xu, Uncertain linguistic aggregation operators based approach to multiple attribute group
decision making under uncertain linguistic environment, Information Sciences 168(14)(2004)
171184.
[15] R. M. Rodriguez, L. Martinez, and F. Herrera, Hesitant fuzzy linguistic term sets for decision
making, IEEE Transactions on Fuzzy Systems 20(1)(2012) 109119.
[16] J. Wang and Z. Zhang, Aggregation operators on intuitionistic trapezoidal fuzzy number and its
application to multi-criteria decision making problems, Journal of Systems Engineering and
Electronics 20(2)(2009) 321326.
[17] R. Lin, X.F. Zhao and G.X. Wei, Models for selecting an ERP system with hesitant fuzzy linguistic
information, Journal of Intelligent & Fuzzy Systems 26(5)(2014) 2155-2165.
[18] Z.S.Xu, Uncertain linguistic aggregation operators based approach to multiple attribute group
39
decision making under uncertain linguistic environment, Information Science 168(2004)171-184.
[19] Z.S.Xu, Induced uncertain linguistic OWA operators applied to group decision making,
Information fusion, 7(2006) 231-238.
[20] G. W. Wei, Uncertain linguistic hybrid geometric mean operator and its Application to group
decision making under uncertain linguistic environment, International Journal of Uncertainty,
Fuzziness and Knowledge-Based Systems, 17(2) (2009) 251-267.
[21] Y.B. Ju, X.Y. Liu, Some aggregation operators with hesitant uncertain linguistic numbers and their
application to multiple attribute group decision making, Tech Report, 2013 .
[22] X.Y Liu, Y.B. Ju, S.H. Yang, Some generalized interval-valued hesitant uncertain linguistic
aggregation operators and their applications to multiple attribute group decision making, Soft
Computing (2014) DOI:10.1007/s00500-014-1518-z.
[23] C. Bonferroni, Sulle medie multiple di potenze, Bolletino Matematica Italiana, 5 (1950) 267-270.
[24] R. R. Yager, On generalized Bonferroni mean operators for multi-criteria aggregation,
International Journal of Approximate Reasoning 50(2009) 1279-1286.
[25] R. R. Yager, G. Beliakov, S. James, On generalized Bonferroni means, In: Proceedings of the
Eurofuse Workshop Preference Modelling and Decision Analysis, 16-18, September, 2009
[26] G. Beliakov, S. James, J. Mordelova, T. Ruckschlossova, R. R. Yager, Generalized Bonferroni
mean operators in multi-criteria aggregation, Fuzzy Sets and Systems 161(2010) 2227-2242.
[27] B. Zhu, Z. S. Xu, M.M. Xia, Hesitant fuzzy geometric Bonferroni means, Information Sciences
205(1)(2010) 72-85.
[28] F. Herrera, E. Herrera Viedma, J.L. Verdegay, A model of consensus in group decision making
under linguistic assessments, Fuzzy Sets and Systems 79(1996) 73 -87.
[29] F. Herrera, E. Herrera Viedma, Linguistic Decision Analysis: Steps for Solving Decision Problems
Under Linguistic Information, Fuzzy Sets and Systems 115(2000) 67-82.
[30] H. Brokl, M.J. Menou, Index of information Utilization Potential, Final Report of Phase 2 of the
IUP Pilot Porject. Los Angeles: GSLIS/UCLA,1982.
[31] Z.S. Xu, Induced uncertain linguistic OWA operators applied to group decision making,
Information Fusion 7 (2006) 231238.
[32] Z.S. Xu, Uncertain linguistic aggregation operators based approach to multiple attribute group
decision making under uncertain linguistic environment, Information Science 168 (2004)
171184.
[33] Z.S. Xu, Induced uncertain linguistic OWA operators applied to group decision making,
Information Fusion 7 (2006) 231238.
[34] F. Smarandache, A unifying field in logics. neutrosophy: Neutrosophic probability, set and logic,
American Research Press, Rehoboth, 1999.
[35] H. Wang, F. Smarandache, Y.Q. Zhang, et al., Interval neutrosophic sets and logic: Theory and
applications in computing, Hexis, Phoenix, AZ, 2005.
[36] V. Torra, Hesitant fuzzy sets, International Journal of Intelligent Systems 25 (6) (2010) 529-539.
[37] V. Torra Y. Narukawa, On hesitant fuzzy sets and decision, IEEE International Conference on
Fuzzy Systems, 18 (2009) 1378-1382.
[38] N. Chen, Z.S. Xu, M.M. Xia, Correlation coefficients of hesitant fuzzy sets and their applications
to clustering analysis, Applied Mathematical Modeling 37 (4) (2013) 2197-2211.
[39] T.Y. Chen, H.P. Wang, Y.Y. Lu, A multicriteria group decision making approach based on
interval-valued intuitionistic fuzzy sets: A comparative perspective, Expert Systems with
40
Applications 38 (6) (2011) 7647-7658.
[40] C. Bonferroni, Sulle medie multiple di potenze, Bolletino Matematica Italiana 5 (1950) 267-270.
[41] W. Zhou, J.M. He, Intuitionistic Fuzzy Normalized Weighted Bonferroni Mean and Its Application
in Multicriteria Decision Making, Journal of Applied Mathematics 2012(2012) 1-22.
[42] M.M. Xia, Z.S. Xu, B. Zhu, Geometric Bonferroni means with their application in multi-criteria
decision making, Knowledge-Based Systems 40 (2013) 88-100.
[43] M. Sun, J. Liu, Normalized Geometric Bonferroni Operators of Hesitant Fuzzy Sets and Their
Application in Multiple Attribute Decision Making, Journal of Information and Computational
Science 10 (9) (2013) 2815-2822.
41

Some Interval-Valued Neutrosophic Hesitant Fuzzy Uncertain Linguistic Bonferroni Mean Aggregation Operators and Their Application in Multiple Attribute Decision Making

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Some Interval-Valued Neutrosophic Hesitant Fuzzy Uncertain Linguistic Bonferroni Mean Aggregation Operators and Their Application in Multiple Attribute Decision Making

Uploaded by

Copyright:

Available Formats

Some interval-valued neutrosophic hesitant fuzzy uncertain linguistic

Bonferroni mean aggregation operators and their application in multiple

(1) If i j , then si s j (i.e., si is better than s j );

(2) Negative operator: neg ( si ) sl 1i ;

(3) If si s j , then max( si , s j ) si ;

(4) If si s j , then min( si , s j ) si .

linguistic set S ( s0 , s1 , , sl 1 ) is extended to a continuous linguistic set S {s | [0, q ]} , q is

sufficiently large number. The operational laws are defined as follows:[28][29]

(1) s1 s2 [ sa1 , sb1 ] [ sa 2 , sb 2 ] [ sa1 a 2 , sb1 b 2 ] (5)

(2) s1 s2 [ sa1 , sb1 ] [ sa 2 , sb 2 ] [ sa1a 2 , sb1b 2 ] (6)

(3) s1 / s2 [ sa1 , sb1 ] / [ sa 2 , sb 2 ] [ sa1/ b 2 , sb1/ a 2 ]if a 2 0, b2 0 (7)

(4) s1 [ sa1 , sb1 ] [ s *a1 , s *b1 ] 0 (8)

2.2 The neutrosophic set

A( x ) x | (T A ( x), I A ( x), FA ( x)) (9)

where, TA ( x) is the truth-membership function, I A ( x) is the indeterminacy-membership function,

and FA ( x) is the falsity-membership function. TA x , I A x and FA x are real standard or

There is no restriction on the sum of TA ( x) , I A ( x) and FA ( x) , so 0 TA ( x) I A ( x) FA ( x) 3 .

A( x ) x | (T A ( x), I A ( x), FA ( x)) (10)

where TA ( x) is the truth-membership function, I A ( x) is the indeterminacy-membership function,

Definition 4 [35]. Let X be a universe of discourse, with a generic element in X denoted by x. A

interval neutrosophic set A in X is

A( x ) x | (T A ( x), I A ( x), FA ( x)) (11)

and FA ( x) is the falsity-membership function. For each point x in X, we have

can call an interval neutrosophic number (INN).

2.3 Interval-valued hesitant fuzzy set

(2) h1 h2 max 1 , 2 , (14)

(3) h1 h2 min 1 , 2 . (15)

on, it will be convenient to call a [ s ( a ) , s ( a ) ], h( a ) a hesitant fuzzy uncertain linguistic element

(HFULE) and A the set of all HFULEs.

uncertain linguistic variable [ s ( x ) , s ( x ) ] , and satisfies these limits :

where ~ ~~t ( x ) max~ , ~ ~~i ( x ) max~, and ~ ~ ~f ( x ) max~ for x X .

two IVNHFULEs, the operation rules are defined as follows:

[~a~ L ~b~ L ~a~ L~b~ L ,~a~U ~b~U ~a~U~b~U ]) (24)

[1 (1 ~a~ L ) ,1 (1 ~a~ U ) ]) (26)

IVNHFULEs, and , 1 , 2 0 , then we have:

(4) 1a~ 2 a~ (1 2 )a~ (30)

[ s ( ( a~ ) (b~ )) , s ( ( a~ ) (b~ )) ], ra~ L ) (1 ~

[ s (b~ ) , s (b~ ) ], rb~ L ) ,1 (1 ~

[ s ( ( a~ ) (b~ )) , s ( ( a~ ) (b~ )) ], ra~ L ) (1 ~

(4) For formula (30), we have

[ s ( a~ ) (b~ ) , s ( a~ ) (b~ ) ], ra~ L ) (~

[ s( ( a~ ) (b~ )) , s( ( a~ ) (b~ )) ], ([(~ rb~ L ) , (~

So, we can get the formula (31) is right.

two IVNHFULEs, the comparison method of IVNHFULEs is expressed as follows:

2.4 The Bonferroni mean (BM) operator

where W ( w1 , w2 ,..., wn )T is the weight vector of ai (i 1,2,..., n) , satisfying 0 wi 1 and

The WBM operator has four desirable properties as follows:

Let a j a( j 1,2,..., n) , then WBM p,q

Theorem 3 [41] (Permutation).

Theorem 4 [41] (Monotonicity).

Theorem 5 [41] (Boundedness).

min(a1 , a 2 ,..., a n ) WBM p,q

where W ( w1 , w2 ,..., wn )T is the weight vector of ai (i 1,2,..., n) ,satisfying 0 wi 1 and

It is easy to demonstrate the WGBM operator has some properties as follows:

Theorem 7 [43] (Idempotency).

Theorem 8 [43] (Permutation).

Theorem 9 [43] (Monotonicity).

If a j bj ( j 1,2,...,n) , then WGBMp,q (a1, a2 ,...,an ) WGBMp,q (b1, b2 ,...,bn ) .

Theorem 10 [43] (Boundedness).

min(a1 , a 2 ,..., a n ) WGBM p,q

3. Interval-valued neutrosophic hesitant fuzzy uncertain linguistic

Bonferroni mean aggregation operators

IVNHFULEs, and p , q 0 , then the interval-valued neutrosophic hesitant fuzzy uncertain

(4) s1 [ sa1 , sb1 ] [ s a1 , s b1 ] 0 (8)