However, sometimes it is somewhat difficult for the DMs to
assign exact values for the membership degrees of certain ele-
ments to a given set, but a range of values belonging to [0,1]
may be assigned. To cope with such a situation, Chen et al. [36]
firstly put forward the concept of interval-valued hesitant fuzzy
set (IVHFS), which permits the memberships of an element to a
given set having a few different interval values:
Definition 4 [36]. Let X be a reference set, and D[0, 1] the set of all
closed subintervals of [0, 1]. An IVHFS on X is
e
A ¼fhx
i
;
~
h
e
A
ðx
i
Þijx
i
2 X; i ¼ 1; 2; ...; ngð3Þ
where
~
h
e
A
ðx
i
Þ: X ? D[0,1] denotes all possible interval membership
degrees of the element x
i
2 X to the set
e
A.
~
h
e
A
ðx
i
Þ is called an
interval-valued hesitant fuzzy element (IVHFE), which satisfies
~
h
~
A
ðx
i
Þ¼f
~
c
j
~
c
2
~
h
~
A
ðx
i
Þg ð4Þ
where
~
c ¼½
~
c
L
;
~
c
U
is an interval number.
~
c
L
¼ inf
~
c and
~
c
U
¼ sup
~
c
express the lower and upper limits of
~
c, respectively. Obviously, if
~
c
L
¼
~
c
U
, then the IVHFEs reduce to the HFEs.
Definition 5 [36]. Let
~
h;
~
h
1
and
~
h
2
be three IVHFEs, then
(1)
~
h
c
¼f½1
~
c
U
; 1
~
c
L
j
~
c
2
~
hg;
(2)
~
h
1
[
~
h
2
¼ max
~
c
L
1
;
~
c
L
2
; max
~
c
U
1
;
~
c
U
2
j
~
c
1
2
~
h
1
;
~
c
2
2
~
h
2
no
;
(3)
~
h
1
\
~
h
2
¼ min
~
c
L
1
;
~
c
L
2
; min
~
c
U
1
;
~
c
U
2
j
~
c
1
2
~
h
1
;
~
c
2
2
~
h
2
no
;
(4)
~
h
k
¼ f½ð
~
c
L
Þ
k
; ð
~
c
U
Þ
k
j
~
c
2
~
hg; k > 0;
(5) k
~
h ¼f½1 ð1
~
c
L
Þ
k
; 1 ð1
~
c
U
Þ
k
j
~
c
2
~
hg; k > 0;
(6)
~
h
1
~
h
2
¼
~
c
L
1
þ
~
c
L
2
~
c
L
1
~
c
L
2
;
~
c
U
1
þ
~
c
U
2
~
c
U
1
~
c
U
2
j
~
c
1
2
~
h
1
;
~
c
2
2
~
h
2
no
;
(7)
~
h
1
~
h
2
¼
~
c
L
1
~
c
L
2
;
~
c
U
1
~
c
U
2
j
~
c
1
2
~
h
1
;
~
c
2
2
~
h
2
no
.
Motivated by the distance measures for HFSs in [38], in what
follows, we propose some basic distance measures between IVH-
FEs, such as the interval-valued hesitant Hamming-Hausdorff dis-
tance, the interval-valued hesitant Euclidean-Hausdorff distance,
the hybrid interval-valued hesitant Hamming distance, and the hy-
brid interval-valued hesitant Euclidean distance:
Definition 6 [36]. For two IVHFEs
~
h
1
and
~
h
2
, the distance measure
between
~
h
1
and
~
h
2
, denoted as dð
~
h
1
;
~
h
2
Þ, should satisfy the
following properties:
(1) 0 6 dð
~
h
1
;
~
h
2
Þ 6 1;
(2) dð
~
h
1
;
~
h
2
Þ¼0 if and only if
~
h
1
¼
~
h
2
;
(3) dð
~
h
1
;
~
h
2
Þ¼dð
~
h
2
;
~
h
1
Þ.
In most cases, the number of intervals for different IVHFEs could
be different, and the interval values are usually out of order. For
convenience, let l ¼ maxfl
~
a
; l
~
b
g, where l
~
a
and l
~
b
are the numbers
of intervals in IVHFEs
~
a
and
~
b, respectively, and we arrange the
interval values in any order based on a possibility degree formula
[39]. In order to more accurately calculate the distance between
two IVHFEs, it is necessary that two IVHFEs have the same number
of intervals. Similar to Definition 2, we can define a new extension
rule for IVHFEs as follows:
Definition 7. Assume an IVHFE
~
h ¼f
~
h
rðiÞ
ji ¼ 1; 2; ...; l
~
h
g, and stip-
ulate that
~
h
þ
and
~
h
are the maximum and the minimum interval
values in the IVHFE
~
h, respectively; then we call
~
h ¼
g
~
h
þ
þ
ð1
g
Þ
~
h
an extension interval value, where
g
(0 6
g
6 1) is the
parameter determined by the DM according his/her risk
preference.
Therefore, if l
~
a
– l
~
b
, then we should add the different interval
values to the IVHFE using the parameter
g
according the DM’s risk
preference. In other word, when the DM’s risk preference is risk-
neutral, we can add the extension interval value
~
h ¼
1
2
ð
~
h
þ
þ
~
h
Þ,
i.e.,
g
¼
1
2
; when the DM’s risk preference is risk-seeking, we can
add the extension interval value
~
h ¼
~
h
þ
, i.e.,
g
= 1; when the
DM’s risk preference is risk-averse, we can add the extension inter-
val value
~
h ¼
~
h
, i.e.,
g
= 0. For instance, let
~
a
¼f½0:1; 0:2; ½0:2; 0:3; ½0:1; 0:3g,
~
b ¼f½0:3; 0:4; ½0:4; 0:5g, and
l
~
a
> l
~
b
. According to the rules mentioned above, we should extend
~
b until it has the same length with
~
a
, the risk-seeking DM may ex-
tend
~
b as
~
b ¼f½0:3; 0:4; ½0:4; 0:5; ½0:4; 0:5g, the risk-averse DM
may extend it as
~
b ¼f½0:3; 0:4; ½0:3; 0:4; ½0:4; 0:5g, and the risk-
neutral DM may extend it as
~
b ¼f½0:3; 0:4; ½0:35; 0:45; ½0:4; 0:5g.
Apparently, the parameter
g
provided by the DM reflects his/her
risk preference which can affect the final decision results. In this
paper, we assume that the decision makers are all risk-averse.
Based on the well-known Hamming distance and the Euclidean
distance as well as the above operational principles, Chen et al. [36]
gave the distance measures for IVHFEs:
d
2
ð
~
a
;
~
bÞ¼
1
2l
X
l
i¼1
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
ð5Þ
d
3
ð
~
a
;
~
bÞ¼
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
2l
X
l
i¼1
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
2
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
2
s
ð6Þ
where
~
a and
~
b are two IVHFEs;
~
a
L
r
ðiÞ
;
~
a
U
r
ðiÞ
;
~
b
L
r
ðiÞ
and
~
b
U
r
ðiÞ
are the ith
largest values in
~
a
L
;
~
a
U
;
~
b
L
and
~
b
L
, respectively, which will be used
thereafter.
Analogous to the distance measures for HFEs in [38], we further
propose some other basic distance measures between IVHFEs:
The interval-valued hesitant Hamming–Hausdorff distance:
d
4
ð
~
a
;
~
bÞ¼
1
2
max
i
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
no
ð7Þ
The interval-valued hesitant Euclidean–Hausdorff distance:
d
5
ð
~
a
;
~
bÞ¼
1
2
max
i
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
2
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
2
ð8Þ
The hybrid interval-valued hesitant Hamming distance:
d
6
ð
~
a
;
~
bÞ¼
1
2
1
2l
X
l
i¼1
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
þ
1
2
max
i
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
no
ð9Þ
The hybrid interval-valued hesitant Euclidean distance:
d
7
ð
~
a
;
~
bÞ¼
1
2
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
2l
X
l
i¼1
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
2
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
2
s
þ
1
2
max
i
~
a
L
r
ðiÞ
~
b
L
r
ðiÞ
2
þ
~
a
U
r
ðiÞ
~
b
U
r
ðiÞ
2
ð10Þ
3. MADM based on TOPSIS and the maximizing deviation
method
This section puts forward a framework for determining attri-
bute weights and the ranking orders for all the alternatives with
incomplete weight information under hesitant fuzzy environment.
Z. Xu, X. Zhang / Knowledge-Based Systems 52 (2013) 53–64
55