TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.5, May 2014, pp
. 4101 ~ 41
0
6
DOI: http://dx.doi.org/10.11591/telkomni
ka.v12i5.5121
4101
Re
cei
v
ed
No
vem
ber 5, 20
13; Re
vised
De
cem
ber 2
7
,
2013; Accep
t
ed Jan
uary 2
0
, 2014
Evaluation Methods of Multidimensional Comfort for
High-speed Train Based on FANP
Wang Hai-y
o
ng
1
*
, ZHANG Wei-
y
u
e
1
, Wang Xiao-m
ing
2
, Dang Ji
an-
w
u
1
, G
A
O
Jie-qiong
1
1
School of Elec
tronic an
d Infor
m
ation En
gi
ne
erin
g,
Lanz
hou
Jiaoton
g Un
iversit
y
, L
anzh
o
u
7300
70, Ch
ina
2
Lanzh
ou U
n
iv
ersit
y
of T
e
chn
o
lo
g
y
, L
anz
hou
7300
50, Ch
ina
*Corres
p
o
ndi
n
g
author, e-ma
i
l
:
w
a
ng
h
y
o
n
g
@
mail.lz
jtu.cn
A
b
st
r
a
ct
Aiming
at the
prob
le
ms
of multid
i
m
ens
io
nal
c
o
mfort fo
r hig
h
-spe
ed
train, on
the
base
o
f
system
atic
analysis of m
u
t
ual
effect
s am
ong
evaluation
indexes, the comp
r
ehensive evalua
tion m
e
thods of
mu
ltidi
m
ens
ion
a
l co
mfort for high-s
pee
d train b
a
se
d on
F
u
zz
y
-
ANP h
a
ve be
en est
ablis
he
d by fu
zzy
theori
e
s an
d n
e
tw
ork hierarc
h
y an
alysis. T
h
e w
e
ight
of cal
i
b
rated
eval
uati
on in
dex
es can
be calc
ulat
ed
by
the hi
erarc
h
y
structure an
d j
udg
ment
matri
x
of eval
uati
o
n
ind
e
xes. M
e
a
n
w
h
ile, th
e qu
antitative v
a
l
u
e
of
comfort i
ndex
e
s
can be
deter
mi
ne
d accor
d
i
ng to the e
ffect
on the co
mfor
t for high-sp
ee
d train. T
he fu
zz
y
eval
uatio
n matrix
ca
n
b
e
establis
he
d and
t
he eval
uat
i
o
n
valu
e of
co
mf
ort for h
i
g
h
-sp
eed
train
ca
n
be
achi
eved
to r
e
ali
z
e
the
cal
i
br
ation
of co
mf
ort leve
l.
T
hou
gh
the
exa
m
p
l
e
a
nalysis,
the
effectiven
ess of t
h
e
meth
ods
ca
n b
e
further
prov
e
d
a
nd t
he stro
ng th
eory
su
ppo
rt ca
n b
e
p
r
ovi
d
ed
fo
r the
co
m
f
o
r
t
e
v
al
u
a
t
ion
for high-s
p
e
ed
train.
Ke
y
w
ords
: fu
z
z
y
the
o
ry, ANP, high-sp
ee
d train, eva
l
u
a
tion
of comfort
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
With the
adve
n
t of the e
r
a
of high
-spee
d
rail
,pe
ople
p
a
y more
an
d
more
attentio
n to the
train rid
e
co
mfort durin
g the runni
ng
whe
n
they
pay attention to the high speed trai
n ru
nning
spe
ed.Comp
ared
with the tradition
a
l
trains,hig
h-spe
ed train
equippe
d with full se
aled
stru
cture,the
run
n
ing
pe
rforma
nce of
the
trai
n,ai
r qu
ality an
d the
com
m
on d
e
corat
i
on
environ
ment
of the train
act the
role
s of vari
ou
s i
ndexe
s
that
affect the
ri
de
comfo
r
t
of
passe
nge
rs [1]. So it is ne
ce
ssary to
st
udy a
co
mprehen
sive
eva
l
uation fo
r th
e comfort
of
high-
spe
ed train [2
].
Evaluation of
compl
e
x sy
stem
s at ho
me and
ab
ro
ad mainly u
s
ing an
alytic hiera
r
chy
pro
c
e
s
s (A
HP) a
s
the m
e
thod fo
r dete
r
mining th
e
in
dex wei
ghts.
AHP is
ba
sed
on a
premise
that
there i
s
n
o
in
teractio
n b
e
twee
n the
system eleme
n
ts in differe
nt l
a
yers an
d th
e sa
me laye
r and
make
s the f
i
nal re
sult d
i
stortion
al. For multip
le
weig
hting, the subj
ective
unce
r
tainty
of
policyma
k
e
r
s will affect th
e obje
c
tivity and a
u
thenti
c
ity of the evaluation
re
su
lts of system
[3
-6].
So we
comb
ine the
fuzzy
theo
ry and
ANP to
b
u
ild
evaluatio
n
model
of hig
h
-spee
d train
s
’s
comfo
r
t ba
se
d on
Fu
zzy
Analytic Network P
r
o
c
e
s
s
(FANP) a
c
cording
to th
e characte
ri
stic of
high-sp
eed
trains’
s
ope
rati
ng e
n
viron
m
e
n
t in o
r
de
r to
evaluate th
e
comfo
r
t of hi
gh-spe
ed t
r
ai
ns
obje
c
tively and accurately.
2. Ev
aluation Model
ANP, develo
ped
on th
e
basi
s
of AHP,is a
de
cisi
on-m
a
ki
ng m
e
thod
whi
c
h
mainl
y
focu
sed
on d
e
ci
sion
-ma
k
in
g pro
b
lem
s
with stru
cture
of feedba
ck a
nd dep
end
en
ce [7]. After the
target of de
cision ma
kin
g
proble
m
is
determi
ned,
ANP element
s ca
n be divi
ded into cont
rol
layer an
d net
work laye
r. T
he co
ntrol la
yer co
ntain
s
the deci
s
io
n
crite
r
ia of de
cisi
on p
r
obl
e
m
s
and the net
work l
a
yer con
t
ains is the
g
r
oup
of
elem
ents do
minat
ed by the co
ntrol layer
which
influen
ce ea
ch other bet
we
en different g
r
oup
s [8].
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 4101 – 41
06
4102
2.1. The De
termination o
f
Index Weig
hts
(1)
Con
s
truct
i
ng judg
ment
matrix
R
. Set
A
as the target o
f
deci
s
ion ma
king p
r
o
b
lem,
set
12
,
,
..
.,
n
P
PP
as el
ement
s of the co
ntrol laye
r,set
12
,
,
..
.,
n
B
BB
as
elem
ents of th
e
netwo
rk
la
ye
r
.,e
le
ments
gr
o
u
p
i
B
co
ntains indi
ca
tors
12
,
,
..
.,
ii
i
n
B
BB
. Set el
ements of t
he
control l
a
yer
(1
,
2
,
.
.
.
,
)
s
P
sn
as the first criterion an
d in
dex
of netwo
rk laye
r elem
ents group
(1
,
2
,
.
.
.
,
)
il
B
ln
as the
se
con
d
criteri
on, we
use
1~
9
quantitative scale metho
d
p
r
ofes
so
r Sa
aty propo
se
d q
uantify the
importa
nce of pairwise co
m
pari
s
on
of mu
ltiple indicators in elem
ent
grou
p
i
B
in orde
r to con
s
tru
c
t
judgme
n
t matrix
()
ij
n
n
Rr
.
(2) Solving
super-mat
rix
W
. We
ca
n get t
he eig
enve
c
tors wh
ich is the weight ve
ctor
of
judgme
n
t matrix
R
and sup
e
r-matrix
W
acco
rding to eige
n
v
alue method
.
(3)
Con
s
tru
c
ting weighte
d
supe
r-mat
r
ix
W
. Set elements of the co
ntrol l
a
yer
(1
,
2
,
.
.
.
,
)
s
P
sn
as the first criterion an
d in
dex
of netwo
rk laye
r elem
ents group
(1
,
2
,
.
.
.
,
)
il
B
ln
as the
s
e
c
o
nd
c
r
iter
io
n
.
Co
mp
ar
e th
e
e
ffec
t
s be
tw
ee
n el
em
ents
group
s t
o
get j
udg
me
nt matrix
A
. We
can
get the e
i
genve
c
tors o
f
judgment m
a
trix
A
Accordi
n
g eige
nvalue
method a
nd
get wei
ghted
matrix
A
. Then
weig
ht eleme
n
ts of sup
e
r-matrix
W
to get
weig
hted sup
e
r-m
atrix
W
, then
WA
W
.
(4) Determini
ng
wei
ght
()
Wi
.If
the limit of weighted
s
u
per-matrix
lim
k
k
W
W
exists we
can
get weig
hts o
f
multiple indicators by calculating a
c
cording to the formula.
1
(
)
l
i
m
1
/
1
,
2
,
...
,
n
k
k
k
Wi
n
W
i
n
.
2.2. Cons
tru
c
ting Ev
alua
tion Matrix
Determine
th
e de
gre
e
of
multiple
in
dicators’
s
i
n
flue
ncin
g
comfo
r
t of hig
h
-spe
e
d
train
s
by
c
o
ns
truc
ting f
u
zz
y evaluation matrix [9]. Firs
t, c
o
nst
r
uct comment
set of comfo
r
t of high-sp
e
ed
train [10], a
s
sho
w
n
in T
a
b
l
e 1. Th
en,
we evalu
a
te
the multiple indicators
in
or
de
r
to
g
e
t
fuz
zy
evaluation m
a
trix of multiple
indi
cato
rs of each ele
m
ent group
(
)
(
1
,
2
,
...,
ki
j
m
n
Ee
i
;1
,
2
,
.
.
.
,
)
nj
n
according t
o
the co
mment set
by usin
g
the expert scorin
g
method, in the
(
)
(
1
,
2
,
.
..,
;
1
,
2
,
.
..,
)
ki
j
m
n
E
ei
n
j
n
:
k
repre
s
e
n
ts the numbe
r of element
grou
ps of the de
cisi
on
makin
g
p
r
obl
em,
m
rep
r
e
s
e
n
t
s the n
u
mbe
r
of indi
cato
rs within th
e el
ements grou
p,
n
r
e
pr
es
e
n
t
s
the num
be
r o
f
levels
of co
mment
set,
ij
e
re
pre
s
ent
s m
e
mbershi
p
d
e
g
r
ee
of evalu
a
t
ing indi
cato
rs
in elem
ents
grou
p a
s
j
V
,
it
also
ca
n b
e
said t
h
at
ij
e
represe
n
ts exp
e
rts propo
rtion
numbe
r of
evaluating th
e ith indicato
r as the jth gra
de.
Table 1. Co
m
m
ent Set of Comfort of Hig
h
-spee
d Trai
n
Q
uant
iz
ed v
a
lu
e
j
V
Ev
aluati
o
n
fe
at
ure
0.1
S
light
ly
affec
t
e
d
0.3
Les
s
aff
e
c
t
ed
0.5
G
ene
ral
l
y
affec
t
e
d
0.7
G
r
eat
ly
affec
t
e
d
0.9
V
e
ry
great
ly
affec
t
ed
2.3. Compre
hensiv
e Ev
aluation Proc
e
s
s
Acco
rdi
ng to
multiple in
di
cators
wei
ght
s an
d fu
zzy
evaluation m
a
trix, comp
re
hen
sive
evaluation p
r
oce
dure of high-spe
ed trai
n comfo
r
t as follows:
(1) Solving
evaluatio
n
matrix of
ea
ch
ele
m
ents grou
p
(
1
,
2
,
...
,
)
i
B
in
,
kk
m
k
BW
E
,
k
rep
r
e
s
ent
s the numb
e
r of
element g
r
o
ups of the d
e
ci
sion m
a
ki
ng problem,
m
repre
s
e
n
ts th
e
numbe
r of ind
i
cators withi
n
the element
s grou
p.
km
W
rep
r
e
s
ents weight
s of multiple indicato
rs.
(2) Solving
e
v
aluation m
a
trix of ea
ch
crite
r
ia
(1
,
2
,
.
.
.
,
)
s
P
sn
,
kk
n
k
n
P
WD
,
k
r
e
p
r
e
s
en
ts
the
numbe
r of cri
t
eria of the d
e
ci
sion m
a
ki
n
g
pro
b
lem,
n
re
pre
s
ent
s the
numbe
r of el
ements
group
s
of each
crite
r
i
a
,
kn
W
rep
r
e
s
ent
s weig
hts of multiple indicators of ele
m
en
ts grou
ps.
12
(
,
,
...
,
)
kn
n
D
BB
B
.
(3) Solving e
v
aluation mat
r
ix of the target
A
,
kk
A
WQ
,
k
re
pre
s
e
n
ts the number
of criteri
a
of the deci
s
io
n makin
g
pro
b
lem,
k
W
re
pre
s
e
n
ts wei
ght of crite
r
ia layer,
12
(,
,
.
.
.
,
)
kk
QP
P
P
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Evaluatio
n M
e
thod
s of Multidim
ensional
Com
f
ort for High-spe
ed Train… (Wan
g Hai-yo
ng
)
4103
(4) Evaluatio
n of
comfo
r
t of high
-spee
d
train.
Evalua
tion value
of
comfo
r
t of hi
gh-spe
e
d
train Ca
n be
rep
r
e
s
ente
d
as:
f
AV
,
V
rep
r
e
s
en
ts evaluation
grad
e ro
w vector in evalu
a
t
ion set,
(
0
.
1
,
0
.
3
,0
.
5
,0
.
7
,0
.
9
)
V
.
Grad
e of
co
mfort of hig
h
-
sp
eed
train
can
be d
e
termined
as
Ta
ble 2 a
c
co
rdi
ng to the
size of evalua
tion value
f
of high-sp
eed trains
comfo
r
t.
Table 2. Grad
e of Comfort
of High-sp
ee
d Train
Ev
aluati
o
n
V
a
lu
e
f
Comfort
lev
e
ls
[0.
8
,1
.0)
V
e
ry
uncomf
o
rt
ab
l
e
[0.
6
,0
.8)
Les
s
comf
or
tab
l
e
[0.
4
,0
.6)
G
ene
ral
l
y
comfor
t
abl
e
[0.
2
,0
.4)
More
com
f
or
ta
ble
[0,
0
.2
)
V
e
ry
comfor
tab
l
e
3. The Example Analy
s
is
Xi'an to
Zhe
n
g
zh
ou
se
ctio
n of
high
spe
ed trai
n i
s
ta
ken
a
s
an
example
in th
e
literature
[11], the typical multiple in
dicato
rs a
r
e
cho
s
e
n
to
be
analyze
d
in this articl
e to build evaluat
ion
system of mu
ltiple index of high-sp
eed
train comfo
r
t, as sh
own in Table 3.
Table 3. Multi
p
le Index Evaluation Sy
ste
m
of Comfort
of High-sp
ee
d Train
Targe
t
s
Criter
io
n l
e
v
e
l
Multi
p
le
In
di
cat
o
r
s
Multi
p
le
ind
e
x
ev
aluati
o
n
sy
stem
O
per
ati
ng pe
rfo
r
m
anc
e
B
1
O
per
ati
ng os
cil
l
at
i
on
B
11
O
per
ati
ng smo
o
th
nes
s
B
12
O
per
ati
ng pe
rfo
r
m
anc
e
B
2
T
he t
e
mpe
r
a
t
ur
e i
n
si
de
th
e c
a
r
B
21
Th
e
hu
mi
di
t
y
i
n
s
i
de
th
e
c
a
r
B
22
A
i
r cl
ea
nli
n
e
ss
B
23
O
per
ati
ng pe
rfo
r
m
anc
e
B
3
S
e
ati
ng comf
or
t
B
31
T
he li
gh
tin
g
i
n
s
i
de
th
e c
a
r B
32
3.1. Dete
rmining Index Weigh
t
s
Con
s
tru
c
ting
ANP netwo
rk structu
r
e a
s
Figur
e 1 accordin
g relatio
n
shi
p
betwe
e
n
high-
spe
ed train
comfort multipl
e
indicators.
Figure 1. ANP Netwo
r
k St
ructu
r
e of Mul
t
iple
Indicato
rs of Ccomfort
of High-spe
e
d
Train
(1)
Con
s
tru
c
ti
ng judgm
ent matrix
As an exa
m
ple of the
operating
perfo
rman
ce
1
B
,
1~
9
scale meth
o
d
pro
p
o
s
ed
by
Professo
r Sa
aty is adopte
d
to con
s
tru
c
t
judgment ma
trix as Table
4-Ta
ble 7.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 4101 – 41
06
4104
Table 4. Ju
dg
ment Matrix1
Table 5. Ju
dg
ment Matrix2
B
11
B
21
B
22
B
23
Weights
B
21
1
2
1/2
0.286
B
22
1/2
1
2
0.143
B
23
2
1/2
1
0.571
B
12
B
21
B
22
B
23
Weights
B
21
1
2
2
0.5
B
22
1/2
1
2
0.25
B
23
1/2
1/2
1
0.25
Table 6. Ju
dg
ment Matrix3
Table 7. Ju
dg
ment Matrix4
B
11
B
31
B
32
Weights
B
31
1
3
0.75
B
32
1/3
1
0.25
B
12
B
31
B
32
Weights
B
31
1
4
0.8
B
32
1/4
1
0.2
(2) Con
s
tru
c
ti
ng
wei
ghted sup
e
r-matrix
,sup
er-matrix
and limiting super-mat
rix
The
softwa
r
e
Supe
r
De
cisi
on i
s
u
s
e
d
in
this
articl
e to
obtain
weigh
t
ed matrix
W
, super-
matrix
W
, limiting su
per-mat
ri
x
W
of multiple index evalu
a
tion sy
stem of
comfo
r
t of hi
gh-spe
ed
train. So as
follows
:
0
1
0.
5
0
.
6
6
7
0
.
75
0.
83
3
0
.
6
6
7
1
0
0.
5
0
.
3
33
0
.
25
0.
1
6
7
0
.
3
3
3
0.
33
3
0
.
4
93
0
0
.
6
6
7
0.
3
3
3
0
.
5
94
0.
6
1
2
W=
0.
33
3
0
.
3
1
1
0
.
66
7
0
0
.
6
6
7
0
.
2
49
0.
2
0
9
0.
33
4
0
.
1
96
0.
3
3
3
0
.
3
3
3
0
0
.
1
5
7
0
.
17
9
0.
75
0.
8
0
.
1
67
0.
6
6
7
0
.
2
0
1
0.
25
0
.
2
0
.
8
33
0.
33
3
0
.
8
1
0
0
0
.
6
37
0.
1
6
7
0
.
2
2
2
0.
2
5
0
.
27
8
0
.
2
22
0
.
63
7
0
0.
167
0.
111
0
.
08
3
0
.
0
56
0.
11
1
0.
086
0
.
1
2
7
0
0
.
22
2
0
.
1
11
0
.
19
8
0
.
2
04
W=
0.
086
0
.
08
0,
2
2
2
0
0.
22
2
0
.
0
8
3
0.
07
0.
086
0.
05
1
0
.
1
1
1
0.
11
1
0
0
.
0
5
2
0
.
0
6
0.
079
0
.
08
4
0
.
0
5
5
0
.
22
2
0
.
0
6
7
0
0
.
3
3
3
0.
026
0.
02
1
0
.
2
7
8
0.
11
1
0
.
2
6
7
0.
33
3
0
0.
257
0.
25
7
0
.
257
0.
257
0.
257
0.
257
0.
25
7
0.
221
0.
221
0.
221
0.
2
2
1
0
.
221
0.
221
0.
221
0.
126
0.
126
0.
126
0.
126
0.
126
0.
1
2
6
0
.
126
W=
0.
101
0.
101
0.
101
0.
1
0
1
0
.
101
0.
101
0.
101
0.
071
0.
071
0.
071
0.
0
7
1
0
.
071
0.
071
0.
071
0.
111
0.
111
0.
111
0.
1
1
1
0
.
111
0.
111
0.
111
0.
113
0.
113
0.
1
1
3
0
.
113
0.
113
0.
113
0.
113
Acco
rdi
ng to
the limiting
super-mat
rix
W
, weig
ht value
of multiple in
dicato
rs of co
mfort
of high-spe
e
d
train is a
s
sh
own in Ta
ble
8.
Table 8. Weight Value of Multiple Indi
cators of
Comf
ort of High
-sp
eed Train
Targe
t
s
Criter
io
n l
e
v
e
l
W
e
ig
hts
Multi
p
le
In
di
cat
o
r
s
W
e
ig
hts
Multi
p
le
ind
e
x
ev
aluati
on
sy
stem
B
1
0
.
47
8
B
11
0
.
25
7
B
12
0
.
22
1
B
2
0
.
29
8
B
21
0
.
12
6
B
22
0
.
10
1
B
23
0
.
07
1
B
2
0
.
22
4
B
31
0
.
11
1
B
32
0
.
11
3
3.2. Cons
tru
c
ting Ev
alua
tion Matrix
Acco
rdi
ng to
the de
gree
of high
-spe
ed t
r
ain
s
comfort
influenced by multiple i
n
dicators
and
com
m
en
t set, expe
rt
scorin
g meth
od i
s
ad
opte
d
to obtai
n th
e fuzzy evalu
a
tion of m
u
ltiple
indicators wit
h
in ea
ch ele
m
ent gro
up,a
s
sh
own in Table 9.
Tab.9 Fu
zzy
evaluation ta
ble of multiple i
ndicators o
f
comfort of high-spe
ed trai
n
Indi
ca
to
r
Slight
ly
A
ffect
ed
Les
s
A
ffect
ed
G
ene
ral
l
y
A
ffect
ed
G
r
eat
ly
A
ffect
ed
V
e
ry
great
ly
A
ffect
ed
B
11
0
.
01
0
.
03
0
.
06
0
.
13
0
.
77
B
12
0
.
03
0
.
08
0
.
12
0
.
26
0
.
51
B
21
0
.
03
0
.
11
0
.
15
0
.
34
0
.
37
B
22
0
.
05
0
.
09
0
.
03
0
.
21
0
.
62
B
23
0
.
03
0
.
04
0
.
11
0
.
37
0
.
45
B
31
0
.
01
0
.
04
0
.
07
0
.
22
0
.
66
B
32
0
.
02
0
.
11
0
.
17
0
.
21
0
.
49
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Evaluatio
n M
e
thod
s of Multidim
ensional
Com
f
ort for High-spe
ed Train… (Wan
g Hai-yo
ng
)
4105
Fuzzy evalua
tion matrix
k
E
of multiple indicators
is
as
follows
:
1
0
.
01
0.
03
0.
06
0.
13
0.
77
0.0
3
0.
08
0.
12
0.
26
0.
51
E
2
0.03
0
.
1
1
0.15
0
.
3
4
0.37
0.05
0.09
0.03
0.21
0.62
0.03
0.04
0.11
0
.
3
7
0.45
E
3
0.01
0.04
0.0
7
0.22
0.66
0.02
0.11
0.17
0.21
0
.
49
E
3.3. Compre
hensiv
e Ev
aluation
Acco
rdi
ng to
the formul
a
kk
m
k
BW
E
, the evaluat
ion matrix of
each elem
e
n
t grou
p is
obtaine
d, so
as follo
ws
：
11
2
1
0.0
1
0
.
03
0.0
6
0
.
13
0.7
7
0.
25
7,
0.
221
0.
009
2,
0.
02
54,
0.
0419,
0
.
0909
,
0
.
310
6
0.
03
0.
0
8
0.
12
0
.
26
0.
51
BW
E
22
3
2
0.
03
0.
1
1
0.
15
0.
34
0.
37
B
=
W
E
=
(
0.
126,
0.
101,
0.
07
1)
0.
05
0.
09
0.
03
0.
21
0.
62
(
0
.
011,
0.
02
58,
0
.
029
7,
0.
0903,
0.
1412)
0.
03
0.
04
0
.
11
0
.
37
0.
45
33
2
3
0.
01
0.
04
0.
07
0
.
2
2
0
.
6
6
B
=
W
E
=
(
0
.
1
11,
0
.
1
13)
(
0
.0
034
,0
.
016
9,
0.
02
7,
0.
04
81,
0
.
1
286
)
0.
0
2
0
.
11
0
.
1
7
0.
21
0.
49
Acco
rdi
ng to
the form
ula
kk
A
WQ
, the eval
uati
on mat
r
ix of t
a
rget
A
is obta
i
ned,
so
as
follows
:
33
0.
0
092
0
.
025
4
0
.
0
419
0.
09
09
0.
3
106
(
0
.
4
78
,
0
.
2
98
,
0
.
2
24)
0.
0
1
1
0
.
0
2
5
8
0
.
0
297
0.
0
903
0.
1
412
(
0
.
0
0
84,
0
.
033
7,
0.
0
3345
,
0
.
0
811,
0.
219
4)
0.
0
034
0
.
016
9
0
.
0
27
0
.
048
1
0
.
1
286
AW
Q
Acco
rdi
ng to
the form
ula
f
AV
, the value
of e
v
aluation
of the hi
gh-sp
ee
d train
comfo
r
t
is obtain
ed, so as follo
ws:
0.
1
0.
3
(
0.
0084,
0.
033
7,
0.
03345,
0.
0
811,
0.
2194)
0.
2826
0.
5
0.
7
0.
9
fA
V
Acco
rdi
ng to
the value
0
.
2826
f
of evaluation
of the hi
g
h
-spee
d trai
n comfort,
comp
re
hen
si
ve evaluation
of high-sp
ee
d train co
mfo
r
t is "More comfortabl
e". While the Hi
gh-
spe
ed trai
n runnin
g
pe
rformance w
ill
b
e
improved in
the future, th
e accu
ra
cy of
valuation of t
he
high-sp
eed train comfo
r
t should b
e
improved in many
aspe
cts.
4.Conclu
sion
A comp
re
he
nsive eval
ua
tion and
re
search
of the
high
-speed
train
comfo
r
t is a
system
atic project. In this article,throu
gh the
an
alysis of fu
zzy
netwo
rk l
e
vel
,
comprehe
n
s
ive
multiple eval
uation mod
e
l
of comfort o
f
high-spee
d train is e
s
tab
lishe
d. This
method not o
n
ly
solve
s
the p
r
oblem of d
e
v
iating from
the actu
al sit
uation
whe
r
e
AHP is u
s
e
d
to obtain t
he
evaluation
re
sults, but also
ove
r
come
s
the subje
c
tive un
ce
rtainty
sin
c
e th
e int
r
odu
ction
of fu
zzy
theory, which
provid
es
a st
rong
theo
reti
cal
sup
p
o
r
t for the
re
sea
r
ch of hig
h
-spe
ed trai
n comf
ort
evaluation.
Ackn
o
w
l
e
dg
ements
This wo
rk wa
s
supp
orted by
the
Natu
ral
Scien
c
e F
ound
ation
of Gan
s
u
(No:1
212
RJZA
055).
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 5, May 2014: 4101 – 41
06
4106
Referen
ces
[1]
WANG Hai-
yo
ng, WANG Xia
o
-ming, DA
NG Jian-
w
u
. C
o
m
p
reh
ensiv
e Ev
alu
a
tion Mo
de
l
of Comfort for
Intercit
y
Hi
gh-s
pee
d Rai
l
w
a
y.
Co
mp
uter Engi
neer
ing
. 2
011;
(01): 263-
26
7.
[2]
WANG Hai-
yo
ng, WANG Xia
o
-ming, DA
NG Jian-
w
u
. C
o
m
p
reh
ensiv
e Ev
alu
a
tion
of Co
mfort of High-
spee
d T
r
ains Based o
n
F
u
zz
y
Red
u
ction.
Jo
u
r
nal of the Ch
in
a Railw
ay Soc
i
ety.
2010; 32(
0
5
): 98-10
2.
[3]
Z
H
AO Yan-l
o
n
g
,
XIONG La
n, XU Mi
n-ji
e,
XI
Z
hao-
hui, W
A
NG Hai-
yi
n
g
. S
y
nt
hetic Ev
al
u
a
tion
for Dr
y-
T
y
pe
T
r
ansformers’ R
unn
in
g Co
nd
ition
Based
on
F
A
HP.
Journ
a
l
of Cho
n
g
q
in
g
Univ
ersity of
T
e
chno
logy (N
atural Sci
ence)
. 2013; 27(
04): 60-6
7
.
[4]
Liu
Xue-c
hen
g
,
Yang J
i
-pe
n
g
, Li Y
un. On
Re
si
denti
a
l
Li
vabil
i
t
y
Com
p
r
ehe
nsive
Eval
uatio
n Mo
del
Based o
n
F
u
zz
y Hi
erarch
ica
l
Anal
ys
is.
Co
mputer App
lic
ations an
d Softw
are
. 201
3; 30(0
1
): 129-1
31.
[5]
LIAO He-pi
ng,
DENG Qi-zho
ng, Z
H
OU Z
h
i
-
qi
a
ng, CAO
Xi
a. Secur
i
t
y
-c
risis assessm
e
n
t and
earl
y
w
a
rni
n
g
for co
al mi
nin
g
enter
prise
bas
ed
on
F
u
zz
y AHP.
J
ourn
a
l
of Hu
na
n Un
iversity
of
Scienc
e
&
T
e
chno
logy (N
atural Sci
ence
Editio
n)
. 201
3; 28(0
1
): 17-2
2
.
[6]
T
I
AN Lin-gang, JIN Cong-con
g
, BA Chao. Applicati
on of improved fuzzy
A
H
P to safe
ty
e
v
aluation of
sea
w
all engine
ering.
Engineeri
ng Journal
of
Wuhan Universi
ty
. 2013; 46(03): 3
17-320.
[7]
W
A
NG Lian-fe
n. T
he
T
heor
y
and A
l
g
o
rithm of
Anal
ytic Net
w
o
r
k
Proc
ess.
Systems E
ngi
n
eeri
ng T
h
e
o
ry
and Practic
e
. 2
001; (3): 45-5
0
.
[8]
YANG Hon
g
-
w
ei, YUE Yo
ng,
YANG Xu
e-q
i
a
ng. Group E
x
te
nsio
n Ana
l
ysis
Net
w
ork Proc
e
ss Decisi
on
Method Bas
ed
on Interval Sc
a
l
es.
Co
mp
uter Scienc
e
. 201
2; 39(06): 21-
24.
[9]
F
U
Li-hu
a
, W
A
NG Dan. Res
e
arch o
n
fuzz
y
l
ogi
c-
base
d
tru
s
t evalu
a
tion
o
f
trusted comp
uting m
ode
ls
[J]. Computer Engi
neer
in
g an
d Appl
icatio
ns. 201
0; 46(2
2
): 4-5.
[10]
DU Jian, CHEN
Hong-bin, Z
H
AO
F
eng
. Digital
Home Service
Schedu
ling Algorithm Ba
sed on F
u
zzy
Logic.
Computer Scien
c
e
. 2013; 40(06): 63-66.
[11]
W
A
N
G
Hai-y
ong
, W
A
N
G
Xiao-mi
ng, DANG Jian-w
u
.
Evaluation of the Comfo
r
t o
f
High-speed
T
r
a
i
ns Based on
Multi-di
mensiona
l
In
complete
Information Sy
stems.
Automation and
Instrumenta
t
ion.
2
012; (1
): 163-164
.
Evaluation Warning : The document was created with Spire.PDF for Python.