TELKOM
NIKA
, Vol. 13, No. 4, Dece
mb
er 201
5, pp. 1478
~1
485
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v13i4.2805
1478
Re
cei
v
ed
Jul
y
27, 201
5; Revi
sed O
c
tob
e
r 5, 2015; A
c
cepted O
c
to
ber 20, 20
15
Evaluation of the Modernization of Hydraulic Projects
Management Compact-Center-Point Triangular
Whitenization Weight Function
Li Lijie*
1
, Wang Xiao
2
, Zhang Lina
3
Busin
e
ss Scho
ol, Hoh
a
i Un
ive
r
sit
y
, No. 8 F
o
c
hen
g W
e
st Road, Jian
gn
ing
District, Nanji
n
g, Jiangs
u
Provinc
e
, Chin
a, Ph./F
ax. 158
506
54
585
1
, 15
151
97
352
2
2
, 1599
62
822
79
3
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: lilij
ie8
1
@
126.
com
1
, w
x
10
1@
hotmai
l
.com
2
, l
i
n
a
z
ha
ng
v@
163
.co
m
3
A
b
st
r
a
ct
In the actual a
nalysis
of grey
clusterin
g
eva
l
uati
on, it has bee
n foun
d that
the lengt
h o
f
the grey
clusteri
ng inter
v
al is partia
lly l
a
rger, w
h
ich is det
er
mi
ned by
the met
hod of
grey
clusteri
ng
eval
uatio
n bas
e
d
on th
e c
enter-
poi
nt tria
ngu
lar
w
h
iteni
z
a
t
i
o
n
w
e
ight fu
nc
tio
n
. In
order
to
solve
this
pro
b
le
m, a
n
e
w
gr
ey
eval
uatio
n met
hod bas
ed on
refor
m
ative
trian
gul
ar
w
h
ite
n
i
z
a
t
i
on w
e
i
g
h
t
function
is r
e
searc
hed. T
h
e
existin
g
en
d-p
o
int trian
gul
ar
w
h
iteni
z
a
ti
on
w
e
ight func
tio
n
and cent
er-po
i
nt triang
ular w
h
iten
i
z
a
t
io
n w
e
ight
function
are
r
e
vise
d, an
d
a n
e
w
comp
act-center-p
oi
n
t
triang
ular w
h
iten
i
z
a
t
io
n w
e
ig
ht functio
n
is
constructed. T
hen the ru
les for grey categ
o
r
y intervalof
th
e three trian
gul
ar w
h
iteni
z
a
ti
o
n
w
e
ight functi
ons
are co
mp
ared
, and an ex
a
m
p
l
e a
bout t
he eva
l
u
a
tion
of the mod
e
rni
z
at
ion of
hydra
u
lic pr
oj
ects
ma
na
ge
me
nt i
s
prop
ose
d
fo
r ana
ly
z
i
n
g
th
e three
me
th
o
d
s to further
verify that the
improv
ed
gre
y
clusteri
ng eva
l
uatio
n metho
d
base
d
on the c
o
mpact-ce
nter
-
poi
nt triangu
lar
w
h
iteni
z
a
tio
n
w
e
ight function
is
feasib
le a
nd e
ffective. T
he result
s sh
ow
that the co
mp
act-center-p
oin
t
triangu
lar w
h
iteni
z
a
t
i
on w
e
i
g
h
t
function
is sup
e
rior to
both t
he e
nd-p
o
i
n
t trian
gul
ar
wh
i
t
en
i
z
atio
n
we
i
ght fu
n
c
tion a
n
d
the center-
poi
nt
triang
ular w
h
ite
n
i
z
a
t
i
on w
e
ig
ht function.
Ke
y
w
ords
: grey clusterin
g
evalu
a
tion, trian
gul
ar w
h
it
eni
z
a
t
i
on w
e
ig
ht function, h
y
drau
lic proj
e
c
ts,
mo
der
ni
z
a
ti
on
Copy
right
©
2015 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
This pa
pe
r is
based on the
grey system
theor
y whi
c
h
wa
s first put forward by J.L
.
Deng
in 1982 [1]. This theo
ry h
a
s be
en ap
pl
ied in grey
cl
usteri
ng eval
uation an
alysis to solve su
ch
uncertain p
r
o
b
lems
with poor informati
on and sm
all
sample [2]. As one of the most co
nce
r
ned
model
s in re
cent years, th
e grey clu
s
te
ring ev
aluatio
n model ba
sed on triang
u
l
ar white
n
ization
weig
ht functi
on ha
s bee
n widely used in su
ch
fields a
s
economi
cs, ma
nagem
ent a
nd
engin
eeri
ng tech
nolo
g
y. And it has attracted m
any
schola
r
s to d
o
relate
d re
searche
s
, but
the
studies are st
ill limited. The existing researches
mai
n
ly emphasize more on application than
on
theory. At pre
s
ent, the con
s
tru
c
tion met
hod of tria
n
g
u
lar
whiteni
za
tion weig
ht function i
s
still i
n
a
developm
ent stage.
Based on the existing
studi
es, there are still
some defi
c
iencies of
the end-point
triangul
ar wh
itenizatio
n
weight fun
c
tio
n
(ET
W
F h
e
reafte
r) and
the cente
r
-point
tria
ngul
ar
whiteni
zation
wei
ght fun
c
ti
on
(CT
W
F
h
e
reafte
r) in
p
r
acti
cal
appli
c
ation, su
ch
as co
mplexity
of
comp
uting a
nd identifying
the endpoi
n
t
s of grey cl
usteri
ng inte
rvals. In orde
r to solve these
probl
em
s an
d make a co
ntribution to t
h
is a
r
ea, thi
s
pape
r propo
se
s an im
pro
v
ed con
s
truct
i
on
model
of tria
ngula
r
whiten
ization
weigh
t
function,
na
mely
the co
mpact
-
center-point
trian
g
u
l
a
r
whiteni
zation
weig
ht functi
on (CCT
WF
here
a
fter).
T
h
en takin
g
the
evaluation in
dex system
o
f
a
reservoi
r in Hubei province
as a ca
se
study, th
is new
model is fu
rth
e
r proved to be feasi
b
le.
This p
ape
r in
clud
es five p
a
rts. Th
e first part i
s
the in
trodu
ction of
the wh
ole p
a
per. Th
e
se
con
d
pa
rt
expoun
ds th
e co
nst
r
u
c
tio
n
of CCT
W
F
.
In the third pa
rt, this
pape
r ma
ke
s a
comp
ari
s
o
n
a
m
ong the
three ki
nd
s of triangul
ar
wh
it
enization
wei
ght functio
n
s
in pra
c
tical u
s
e.
In sectio
n fou
r
, a ca
se stu
d
y is con
d
u
c
ted to fu
rther
verify the feasibility and va
lidity of the grey
clu
s
terin
g
evaluation mo
d
e
l based on
CCT
WF. In t
he last pa
rt, this pa
per
su
mmari
ze
s its
major
finding
s and
con
c
lu
sio
n
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 147
8 – 1485
1479
2. Cons
truc
tion of CCT
W
F
S. F. Liu et al. have co
nstructed th
e gre
y
cluste
ring
e
v
aluation met
hod b
a
sed on
ETWF
and that b
a
se
d on
CT
WF [
3
,4]. In pra
c
tical u
s
e, the
d
i
vision of g
r
e
y
cluste
ring
i
n
tervals in th
ese
existing trian
gular
whiteni
zation weight
functions i
s
short of sci
entificity. In r
e
sp
on
se to this
sho
r
tage, on
the basi
s
of analyzi
ng the
overlappi
ng
prop
ertie
s
, the clu
s
terin
g
coeffici
ents, the
division of g
r
ey cluste
rin
g
intervals
and
the sele
ction
of endpoi
nts i
n
these fu
ncti
ons, this
pap
er
improve
s
the
existing fun
c
tions an
d con
s
tru
c
ts
CCT
W
F. The
cal
c
ulation p
r
o
c
e
dure
of the g
r
ey
clu
s
terin
g
evaluation mo
d
e
l based on
CCT
WF i
s
as follows:
Assu
ming th
e
r
e is
an o
b
je
ct set O =
{Oi|
i = 1,
2, …, n
}, whi
c
h i
s
cl
u
s
tere
d into dif
f
erent
grey clu
s
ters of
s,
an
d s
∈
{
1
,2,3,4}. The
n
g
= {
g
j| j
=1
, 2, …,
m }
is
the evaluatio
n index
set
of a
n
objec
t Oi.
ij
x
, i = 1,
2, …, n
;
j =1,2, …,
mi are the
o
b
se
rvation v
a
lue
s
of a
n
obje
c
t Oi fo
r
clu
s
terin
g
ind
e
x gj. The co
rre
sp
ondi
ng
obje
c
t Oi ca
n
be evaluate
d
according to
the observati
on
value
ij
x
. In order to d
e
scri
be it co
rre
ctl
y
, any object
Oi
∈
O is ta
ken as
an ex
ample. The
followin
g
is th
e pro
c
ed
ure:
Step 1: Dete
rmine the
12
,,
,
s
ij
ij
ij
be
the grey
cent
er poi
nts of th
e clu
s
teri
ng i
ndex gj
of the obj
ect
Oi , and th
e
value range
allowed fo
r
ij
x
is
11
[,
]
s
ij
i
j
aa
, thus we
can get th
e ce
nter
points
01
,
s
ij
ij
by extendin
g
grey clusters
toward different direction
s
.
Step 2: Let
1
,1
,
2
,
,
2
kk
ij
i
j
k
ij
bk
s
, then we
get the interval
1
,
kk
ij
ij
bb
.
A
ssu
ming
1
max
,
kk
k
k
ki
j
i
j
i
j
i
j
bb
,
1
mi
n
,
k
kk
ki
j
i
j
, then we i
dentify the
grey inte
rval
of the cl
uster k i
s
1
,,
kk
kk
kk
k
k
ij
ij
ij
ij
cc
. Special
note
:
if
1
2
kk
ij
ij
k
ij
bb
, then
k
k
ki
j
ij
cb
, and
1
1
k
k
ki
j
ij
cb
.
Let the
grey i
n
terval of th
e
clu
s
ter
k be
1
,
kk
kk
ij
ij
cc
, c
o
nnec
t
the points
,0
k
k
ij
c
,
,1
k
ij
, and
1
,0
k
k
ij
c
, then
we
can
get t
he tria
ngul
ar
whiteni
zation
weig
ht fun
c
tion of th
e
index j on the
grey clu
s
ter
k is
,
1
,2
,
,
,
1
,2
,
,
,
1
,2
,
,
k
ij
f
in
j
m
k
s
.
For an o
b
servation value
ij
x
of the index j, we ca
n prov
e that its degree of membe
r
sh
i
p
to the grey cl
uster
=1
2
kk
s
,
,
…
,
is
k
ij
i
j
f
x
by the followi
ng formul
as:
1
1
1
1
0,
,
,,
,,
kk
ij
k
k
ij
i
j
k
ij
k
k
ij
kk
ij
i
j
ij
k
i
j
k
ij
k
ij
k
ij
k
ki
j
k
ij
k
ij
i
j
k
k
ij
k
ki
j
ij
xc
c
xc
fx
x
c
c
cx
xc
c
(1)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Evaluatio
n of the Mode
rni
z
ation of Hydrau
lic Proje
c
ts Managem
ent
Com
pact-… (Li Lijie
)
1480
Step 3: The integrate
d
clu
s
terin
g
coefficient
s of the obje
c
t
=1
2
ii
n
,
,
…
,
belon
gin
g
to the grey cl
uster
k can b
e
cal
c
ulate
d
by :
1
m
kk
ii
j
i
j
i
j
j
fx
(2)
Whe
r
e
k
ij
ij
f
x
is the triangula
r
wh
itenizatio
n
we
ight
function
of the index j
belon
ging to
the
g
r
ey clu
s
ter
k, and
ij
is th
e
wei
ght of the
obje
c
t
i
belo
nging
to the
index j
in
comp
re
hen
si
ve cluste
ring.
Step 4: Beca
use of
*
1
max
kk
ii
ks
, we can say that th
e obje
c
t
i
belo
ngs to the g
r
e
y
clu
s
t
e
r
k
.
It
mean
s
that
wh
en
m
o
re
tha
n
one obje
c
ts belon
g to the grey cluste
r
k
, we can
sort th
ese o
b
ject
s a
c
cording to th
e
size of th
e
i
n
tegrate
d
clu
s
terin
g
coeffi
cient
s, an
d then
determi
ne the
precede
nce or quality of each o
b
je
ct wh
ich bel
ong
s to the grey clu
s
ter
k
.
3. Div
i
sion o
f
Gre
y
Clustering Interv
als
There is no p
r
a
g
matic
wa
y of sel
e
cting ET
WF’
s
en
d
points
01
2
1
2
,,
,
,
,
,
s
ss
aa
a
a
a
a
, and the div
i
sion
of grey
clu
s
ters is l
a
ck of scie
ntific eviden
ce.
Als
o
, CTWF lets
k
, which is most likely to belong to the grey cl
ust
e
r
k
, be the end point of
that grey clu
s
ter, so it’s more a
p
t to get
each gre
y
cluster’
s tri
angul
ar whitenization weight
function
s ba
sed on
01
2
1
,,
,
,
,
s
s
. [5,
6] In fac
t, in
ac
cordanc
e
with their think
i
ng
habits,
peo
pl
e have
mo
re
accu
rate
un
derstandi
ng
and j
udgm
en
t of grey cl
ustering
end
po
ints
than
tho
s
e of
grey clu
s
teri
ng
inte
rvals, and CT
WF
i
s
sup
e
ri
or to
ETWF o
n
en
dpoint
s sele
ction.
But the division of grey clu
s
te
r
CTW
F
la
ck
s s
c
ientificity.
Let
12
,,
,
s
ij
ij
ij
be the g
r
ey ce
nter p
o
i
nts of the cl
usteri
ng in
de
x gj of the object Oi,
and the
ran
g
e
of value all
o
we
d for
ij
x
is
11
[,
]
s
ij
i
j
aa
. Acco
rdi
ng to
the co
nstruction method
s of
C
T
W
F
an
d C
C
T
W
F
,
th
e
gr
e
y
in
t
e
rvals of th
e grey clu
s
ter
k
are
11
,
kk
ij
ij
and
1
,
kk
kk
ij
i
j
cc
, resp
ectively
, and it is cle
a
r that
1
11
,,
kk
kk
kk
i
j
i
j
ij
ij
cc
. Thus
for the
same
g
r
ey in
terval, the
grey interval
le
ngth divi
d
ed
by CCT
W
F i
s
sm
aller, th
e
cro
s
sing
a
r
ea
of
grey
ha
ze
set is dimini
shed, the
cal
c
ulatio
n
effici
ency i
s
increased, m
ean
while
it furth
e
r
differentiate
s index observation value’
s memb
ersh
i
p
grad
e for
each grey
cl
uster, a
nd th
us
ensures the concl
u
si
on’s reliability.
4. Case Stud
y
Based
on
the
co
nnotation,
feature
s
a
nd
func
tion
of th
e evaluatio
n
of the mo
dernizatio
n
of hydra
u
lic proj
ect
s
m
anag
ement,
and
on a
c
count of
a
nalyzin
g water
con
s
e
r
va
ncy
con
s
tru
c
tion’
s macro
s
copi
c backg
rou
nd
and u
nde
rsta
nding th
e ev
aluation
of the mode
rni
z
ati
on
of hydrauli
c
proje
c
ts m
a
n
ageme
n
t, this pap
er
, u
s
i
ng the meth
ods li
ke literature readi
n
g
,
freque
ncy an
alysis, attribu
t
e
red
u
ctio
n and refe
ren
c
e metho
d
, co
nfirms the i
n
d
e
x syste
m
of
the
evaluation
of the mode
rni
z
ation
of hydraulic
proj
ect
s
manag
eme
n
t. The examp
l
e of a reserv
oir
in Hubei
is p
r
opo
se
d, an
d
ea
ch
evalua
tion ind
e
x’s cha
r
a
c
teri
stic
value
s
a
nd weig
hts are
as
sho
w
n in Ta
b
l
e 1 [7].
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 147
8 – 1485
1481
T
abl
e 1.
The
Evaluation Index Sy
stem of the Moderni
zation of Hyd
r
auli
c
Proje
c
t
s
Mana
gem
e
n
t
criterion la
y
e
r
index level
subg
oal
w
e
ight
index
w
e
ight
E
v
al
uati
on
va
l
u
e
%
the modernizatio
n
level of
organizational
management
0.2
the improvement
rate of man
age
ment s
y
stem and
operational mech
anism
0.2 90
the improvement
rate of man
age
ment regulations
0.2
97
the adaptive rate
of personnel nu
mber on gu
ard
and post setting r
equirements
0.15 90
the target
rate of
talent propo
rtion
0.10
75
the target
rate of
staf
f annual training propo
rtion
0.15
80
the target
rate of
archives management s
y
stem
0.2
95
the modernizatio
n
level of safety
management
0.15
the target
rate of
w
a
t
e
r administrat
ion management
0.40
93
the
target rate
of flood-prevention ability
0.30
96
the target
rate of
accident anticipa
t
ed plan
formulation and i
m
plementation
0.30 98
the modernizatio
n
level of project
management
0.5
the intact r
a
te of
pr
ojects facilit
ies
0.15
75
the intact rate of
observation facilities
0.10
90
the target
rate of
projects inspection
standardization
0.10 93
the target
rate of
projects maintenance
implementation
0.10 80
the target
rate of
projects safet
y
m
onitoring
automation deg
r
ee
0.15 83
the target
rate of
projects automatic control degree
0.15
81
the target
rate of
of
fice automation degree
0.15
86
the target
rate of
w
a
t
e
rs functional areas'
w
a
ter
quality
0.05 100
the control rate
o
f
water an
d soil loss
0.05
97
the modernizatio
n
level of economic
management
0.15
the profit and los
s
rate of
w
a
te
r m
anagement u
n
its
0.25
100
the implement ra
te of repair
and
maintenance fun
d
s
0.25
100
the tax
rate of
re
asonable
w
a
ter c
harges and ot
her
fees
0.25 89
the utilization rati
o of developable
w
a
t
e
r an
d soil
r
e
sour
ces
0.25 85
(1)
Co
nfirm the evaluatio
n
of
grey cl
ust
e
rs. A
c
cordi
n
g to
the eval
uation requi
rements,
we divid
e
the
evaluation
of
the mod
e
rni
z
ation
of
hyd
r
auli
c
p
r
oje
c
t
s
ma
nag
eme
n
t into four g
r
ey
clu
s
ters: “po
o
r
type”, “ge
n
e
r
al type”
, “go
od type” and
“excell
ent type”.
(2) Combini
n
g the
pro
p
o
s
als
offered
b
y
the
expe
rts of hydrauli
c
proje
c
ts ma
n
ageme
n
t
institution
s
, we ca
n confirm
each
g
r
ey cl
uster’
s
end
p
o
ints, an
d the
n
cal
c
ul
ate its grey
clu
s
teri
ng
intervals
accordin
g to the
calculation f
o
rmul
as
of
CTWF a
nd CCT
W
F, whi
c
h are d
epi
cte
d
in
Table 2.
T
abl
e 2.
The
Divisio
n
of Grey Clus
te
ring
Intervals Ba
sed on CT
WF
and CCT
W
F
w
h
itenization w
e
ight
fu
ncti
on
poor
general
goo
d
excellen
t
CTWF
11
55
8
5
x
11
70
93
x
11
85
9
8
x
11
93
103
x
CCTWF
12
65
7
5
.
5
x
12
75.5
92.5
x
12
89
97
x
12
95
.5
10
0
.
5
x
1) Fo
r the
gre
y
cluste
ring
model b
a
sed
on CT
WF, th
e proce
d
u
r
e
of the evaluat
ion of the
mode
rni
z
atio
n of hydrauli
c
proje
c
ts ma
n
ageme
n
t is a
s
follows:
Step 1: Accordin
g to th
e co
nstructio
n
method
of CTWF, we
can
cal
c
ula
t
e the
whiteni
zation
weight fun
c
tion clu
s
teri
n
g
coe
ffici
ent
s of ea
ch in
dex, and it mean
s we can
cal
c
ulate the
membe
r
shi
p
gra
de
()
kk
ji
j
f
x
of each index
whi
c
h bel
o
ngs to g
r
ey
clu
s
ter
(
1
,2
,
3
,4
)
kk
.
If
11
90
x
, then
12
3
4
1
1
11
1
1
1
1
11
1
1
,
,
,
0
,0
.
3
7
5
,0
.
6
2
5
,0
f
x
fx
f
x
fx
. Similarly, we
can
get the
whiteni
zatio
n
wei
ght fun
c
tion
clu
s
teri
ng coefficie
n
t
s
of
the evaluation of
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Evaluatio
n of the Mode
rni
z
ation of Hydrau
lic Proje
c
ts Managem
ent
Com
pact-… (Li Lijie
)
1482
mode
rni
z
atio
n of hydrauli
c
proje
c
ts ma
n
ageme
n
t, whi
c
h are depi
ct
ed in Table 3.
T
abl
e 3.
The
T
r
ian
gula
r
Whitenization
W
e
ig
ht
Fun
c
tion of the Evaluation Indexe
s
ba
sed o
n
CT
WF
code
1
ji
j
f
x
2
ji
j
f
x
3
ji
j
f
x
4
ji
j
f
x
code
1
ji
j
f
x
2
ji
j
f
x
3
ji
j
f
x
4
ji
j
f
x
11
x
0 0.375
0.625
0
33
x
0 0
1 0
12
x
0 0
0.2
0.8
34
x
0.333
0.667
0
0
13
x
0 0.375
0.625
0
35
x
0.133
0.867
0
0
14
x
0.667
0.333
0
0
36
x
0.267
0.733
0
0
15
x
0.333
0.667
0
0
37
x
0 0.875
0.125
0
16
x
0 0
0.6
0.4
38
x
0 0
0
0.6
21
x
0 0
1 0
39
x
0 0
0.2
0.8
22
x
0 0
0.4
0.6
41
x
0 0
0
0.6
23
x
0 0
0 1
42
x
0 0
0
0.6
31
x
0.667
0.333
0
0
43
x
0 0.5
0.5
0
32
x
0 0.375
0.625
0
44
x
0 1
0 0
Step 2: Acco
rding to th
e integrate
d
cl
u
s
teri
n
g
coefficient formula
s
in the met
hod
s of
CT
WF clu
s
te
ring evaluati
on,
we
can cal
c
ulate
ea
ch criterio
n la
yer’s i
ndex
as
well
as t
he
integrate
d
cl
usteri
ng
co
efficients of th
e evalu
a
ti
on
of the m
ode
rnizatio
n of
h
y
drauli
c
p
r
oj
e
c
ts
manag
eme
n
t, which are d
e
p
icted in Ta
bl
e 4.
T
abl
e 4.
The Integrate
d
Wh
itenizatio
n
W
e
ight
Fun
c
tio
n
of the Evaluation Indexe
s
base
d
on
CT
WF
gre
y
clus
ter
1
x
2
x
3
x
4
x
x
poor
0.1
1
7
0
0.193
0
0.12
general
0.265
0 0.525
0.375
0.372
good
0.379
0.52
0.191
0.125
0.268
excellent 0.24
0.48
0.07
0.3
0.2
Step 3: By a
nalyzin
g the
clu
s
terin
g
re
sults in
Ta
ble
4, and
by
2
14
m
a
x
=
0.372
k
k
, we
kno
w
that the re
sult of the ev
aluation
of the modernizatio
n
of
hydrauli
c
proje
c
ts ma
nag
em
ent
belon
gs to “g
eneral type”.
2) Fo
r the grey cluste
ring
model ba
se
d
on
CCT
WF,
the pro
c
ed
ure of the evaluation of
the mode
rni
z
ation of hydra
u
lic pr
oje
c
ts
manag
eme
n
t is as follo
ws:
Step 1: Accordin
g to th
e co
nstructio
n
method
of CCT
WF,
we ca
n calcul
ate the
whiteni
zation
weight fun
c
tion clu
s
teri
n
g
coe
ffici
ent
s of ea
ch in
dex, and it mean
s we can
cal
c
ulate the
membe
r
shi
p
gra
de
()
kk
ji
j
f
x
of each index
whi
c
h bel
o
ngs to g
r
ey
clu
s
ter
(1
,
2
,
3
,
4
)
kk
.
If
11
90
x
, then
1
234
1
1
1
1
11
1
1
1
1
11
,
,
,
0
,0
.
3
3
3
,0
.
2
5
,
0
f
x
fx
f
x
fx
. Si
milar
l
y, w
e
can
get the whi
t
enizatio
n
weight fun
c
tio
n
clu
s
te
ring
coeffici
ents of the ev
aluation
of the
mode
rni
z
atio
n of hydrauli
c
proje
c
ts ma
n
ageme
n
t, whi
c
h are depi
ct
ed in Table 5.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 147
8 – 1485
1483
T
abl
e 5.
The
T
r
ian
gula
r
Whitenization
W
e
ig
ht
Fun
c
tion of the Evaluation Indexe
s
ba
sed o
n
CCT
WF
code
1
ji
j
f
x
2
ji
j
f
x
3
ji
j
f
x
4
ji
j
f
x
code
1
ji
j
f
x
2
ji
j
f
x
3
ji
j
f
x
4
ji
j
f
x
11
x
0 0.333
0.25
0
33
x
0 0
1 0
12
x
0 0
0
0.714
34
x
0 0.474
0
0
13
x
0 0.333
0.25
0
35
x
0 0.789
0
0
14
x
0.091
0
0
0
36
x
0 0.579
0
0
15
x
0 0.474
0
0
37
x
0 0.867
0
0
16
x
0 0
0.5
0
38
x
0 0
0
0.2
21
x
0 0
1 0
39
x
0 0
0
0.714
22
x
0 0
0.25
0.2
41
x
0 0
0
0.2
23
x
0 0
0 1
42
x
0 0
0
0.2
31
x
0.091
0
0
0
43
x
0 0.467
0
0
32
x
0 0.333
0.25
0
44
x
0 1
0 0
Step 2: Ba
sed o
n
the
in
tegrated
cl
ustering
co
efficient form
ula
s
in the
meth
ods of
CCT
WF
clustering
evalua
tion, we
can
cal
c
ul
ate e
a
c
h
criterio
n l
a
yer’s ind
e
x
as
well
a
s
t
h
e
integrate
d
cl
usteri
ng
coef
ficient of the
evaluati
on
of the mode
rnizatio
n of hydrauli
c
proj
ects
manag
eme
n
t, which is de
pi
cted in Tabl
e 6.
T
abl
e 6.
The Integrate
d
Wh
itenizatio
n
W
e
ight
Fun
c
tio
n
of the Evaluation Indexe
s
base
d
on
CCT
WF
gre
y
clus
ter
1
x
2
x
3
x
4
x
x
poor
0.009
0 0.014
0 0.009
general
0.188
0 0.416
0.367
0.301
good
0.188
0.475
0.125
0
0.171
excellent 0.143
0.36
0.046
0.1
0.121
Step 3: By a
nalyzin
g the
clu
s
terin
g
re
sults in
Ta
ble
6, and
by
2
14
m
a
x
=
0.301
k
k
, we
kno
w
that the re
sult of the ev
aluation
of the modernizatio
n
of
hydrauli
c
proje
c
ts ma
nag
em
ent
belon
gs to “g
eneral type”.
(3) Analy
s
is o
f
the evaluation re
sults
From Ta
ble
3, we ca
n se
e that there i
s
overl
ap bet
wee
n
the adj
ace
n
t grey cl
usteri
ng
intervals fo
r CT
WF, while
there is n
o
overlap b
e
twe
e
n
1
()
k
ji
j
f
x
and
2
()
k
ji
j
f
x
for CCTWF.
Comp
ari
ng T
able 4 an
d T
able 6, we
can se
e
that the clu
s
te
ring
coefficie
n
t b
a
se
d on
CCT
WF
sim
p
lifies cro
s
si
ng functio
n
cal
c
ulatio
n
a
nd thus
ma
kes it ea
sy a
nd co
nvenie
n
t to
operate.
A
cco
rdi
ng t
o
*
14
max
kk
k
,
we ca
n see
from
T
able
5 and Table
7
that
the re
sult of
the
evaluation
of
the mo
dernization of
hydra
u
lic
project
s
manag
eme
n
t belo
n
g
s
to
“good
type” in
the
mode
rni
z
atio
n level of
saf
e
ty manag
em
ent, “ge
n
e
r
al
type” in
proj
e
c
t man
age
me
nt and
“g
ene
ral
type” in
eco
nomic ma
na
gement.
Ho
wever, th
ere
exist different evalu
a
tio
n
re
sult
s in
the
mode
rni
z
atio
n level of organi
zatio
nal
manage
me
nt, among
whi
c
h the result of the
grey
clu
s
terin
g
evaluation
ba
sed o
n
CT
WF bel
ong
s to
“g
ood
type”, but the
re
sult of
the
g
r
ey
clu
s
terin
g
ev
aluation b
a
se
d on CCT
W
F
sho
w
s that
the seconda
ry
index of this proje
c
t ha
s
a
simila
r mem
bership
gra
de between
“goo
d ty
pe” an
d “g
e
neral type
”. Observing
the
comp
re
hen
si
ve evaluatio
n re
sults,
we
can
se
e that the disti
n
ction b
e
twe
en
2
an
d o
t
her
clu
s
terin
g
co
efficients i
s
much
clea
re
r when
the g
r
ey evaluation
method is b
a
se
d on CCTWF
rathe
r
than
CT
WF. The
hydrauli
c
p
r
o
j
ect bel
ong
s to “gen
eral
type” in the
comp
re
hen
sive
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Evaluatio
n of the Mode
rni
z
ation of Hydrau
lic Proje
c
ts Managem
ent
Com
pact-… (Li Lijie
)
1484
evaluation re
sults, ha
s go
od tech
nical indexe
s
,
attaches imp
o
rta
n
c
e to enviro
n
m
ent prote
c
ti
on
and soci
al influen
ce, and it works well in
gene
ral.
5. Conclusio
n
On the basi
s
of researching the pro
b
lems
li
ke g
r
ey clu
s
terin
g
overlap, cl
usteri
n
g
indexe
s
and grey clu
s
teri
n
g
intervals co
nfir
mation in
ETWF and
CTWF, we con
s
tru
c
t CCTWF
and the con
c
l
u
sio
n
s a
r
e a
s
follows:
(1) From the
com
p
a
r
ative
re
sea
r
ch
of
grey
clu
s
teri
n
g
overl
appi
n
g
features,
we kno
w
that CCT
WF i
s
su
peri
o
r to
ETWF in the
ov
erlap
p
ing f
eature
s
of grey cluste
rs.
(2) From th
e
comp
arative rese
arch
of
cl
usteri
ng
coefficient
s, we
g
e
t that ET
WF
fails to
meet the
no
rmalizatio
n, CTWF
meet
s t
he
wea
k
no
rmalizatio
n,
a
nd CCTWF,
whi
c
h simplifi
e
s
cro
s
sing fun
c
tion cal
c
ulatio
n and ma
ke
s it easy
and convenie
n
t to operate, mee
t
s norm
a
lity.
(3) From
the comp
arative
rese
arch
of th
e divisio
n
of
grey cl
uste
r,
we
can
s
see
that the
confirmation
of grey clu
s
te
ring inte
rvals
based on
CCTWF fits in wi
th end point
s’
conn
otation to
a highe
r de
g
r
ee
,
dimini
sh
es the
crossi
ng ar
ea
s of grey ha
ze
set, and furth
e
r differentiates
index ob
serv
ation value’
s membe
r
ship
grad
e for ea
ch grey clu
s
te
r.
(4)
The
eval
uation of th
e
mode
rni
z
ati
on of
hyd
r
a
u
lic p
r
oje
c
t
s
mana
geme
n
t further
verifies the t
h
ree a
bove m
e
thod
s’ validit
y as we
ll a
s
the fea
s
ibility and effe
ctiveness of the
g
r
ey
clu
s
terin
g
evaluation meth
od. In con
c
lu
sion, CCT
W
F
is sup
e
rio
r
to
ETWF and
CTWF.
Referen
ces
[1]
JL
Deng.
Grey system theor
y (V
ersion 2)
, Huaz
hon
g Uni
v
ersit
y
of scie
n
ce an
d techn
o
lo
g
y
pr
ess
.
200
4: 1-12.
[2]
YH W
ang a
n
d
YG Dang.
T
he post-eva
l
u
a
ti
on meth
o
d
of gre
y
fi
xed
w
e
i
ght cluster ba
sed on D-
S
evid
ence th
eor
y
.
Systems En
gin
eeri
ng-th
eor
y & Practice
. 2009; 29(
5): 123
-128.
[3]
SF
Liu,
YG Dang, Z
G
F
ang and NM Xi
e.
T
he gre
y
s
y
stem t
heor
y
and its app
licati
on.
Sci
ence Press
.
201
0: 1
18-
129.
[4]
SF
Liu and N
M
Xie. Ne
w
g
r
e
y
ev
alu
a
tio
n
method bas
e
d
on reformati
ve triang
ular
w
h
ite
n
izati
o
n
w
e
ig
ht functio
n
.
Journal of Sy
stem En
gin
eeri
n
g
. 201
1; 26(2)
: 244-25
0.
[5]
JF
Z
hang, Z
J
W
u
, PF
F
eng, DW
Y
u
. Evaluati
on s
y
stems and meth
ods o
f
enterprise inf
o
rmatio
n
and
app
licati
on.
Ex
pert Syst.
Appl.
201
1; 38: 8
9
3
8
-89
48.
[6]
W
Meng, SF
Liu, B Z
eng and NM
Xie. Stand
ard trian
g
u
l
ar
w
h
it
e
n
izati
on
w
e
i
ght function and it
s
app
licati
on i
n
g
r
e
y
cl
usteri
ng e
v
alu
a
tion.
J. Grey Syst
. 2012; 24: 39-4
8
.
[7]
YK
T
an, S Gu
an a
nd
N Z
h
a
o
.
T
he ind
e
x
s
y
stem
of
the
e
v
alu
a
tion
of th
e mod
e
rniz
atio
n of h
y
d
r
au
li
c
proj
ects mana
geme
n
t and its method resea
r
ch.
Journal of
China T
h
ree
Gorges Univ
er
sity (Natural
Scienc
es)
. 201
3; 35(3): 36-3
9
.
[8]
C Li, Sustain
abl
e post-eval
uatio
n s
y
stem
research of h
y
dra
u
lic pro
j
ects.
Southw
est Petroleu
m
Univers
i
ty
. 201
1: 1-5.
[9]
YQ Gao, GH
F
ang, CH Ha
n and
H W
a
n
g
.
Anal
ys
is of
connot
ation,
purp
o
se a
nd c
ontent of th
e
moder
nizati
on of h
y
dra
u
l
i
c projects mana
ge
ment.
Journal of China T
h
ree
Gorges Unive
r
sity (Natura
l
Scienc
es)
. 200
9; 31(4): 45-4
8
,
60.
[10]
HX Ou
ya
n
g
, X L
i
an
d GH
F
ang.
T
he moder
ni
zati
on
o
f
h
y
dr
au
lic pr
ojects man
a
g
e
ment an
d its
eval
uatio
n in
d
e
x s
y
stem.
So
uth to North W
a
ter T
r
ansfers and W
a
ter Scienc
e & T
e
chnol
ogy
. 201
2;
10(1): 15
0-1
5
2
,
157.
[11]
GH
F
ang, YQ
Gao,
W
X
T
an,
Z
Z
Z
heng and
N Guo.
T
he evalu
a
tion in
de
x s
y
stem
’
s
construction of th
e
moder
nizati
on
of h
y
dr
aul
ic projects mana
ge
ment.
Advanc
ed in Scienc
e
and T
e
chno
lo
gy of W
a
ter
Resources
. 20
13; 33(3): 3
9
-4
4.
[12]
HH Ni, D
Y
Ni
and
Y
X
Ma. Di
scuss on th
e
moder
ni
zati
on
of h
y
dra
u
lic
pr
ojects ma
nag
e
m
ent an
d the
refinem
ent con
s
truction.
Hai
h
e W
a
ter Reso
u
r
ces
. 2012; 1: 24-2
6
.
[13]
SF
Liu. On index s
y
stem and
mathematical
mode
l for evaluatio
n
of scien
tific and techni
cal strength.
Kyben
etes
. 20
06; 35: 12
56-1
264.
[14]
LN Z
h
a
ng, F
P
W
u
and P
Jia
.
Gre
y
ev
alu
a
ti
on
mo
del
bas
e
d
on r
e
formati
ve trian
gul
ar
w
h
iten
izatio
n
w
e
ig
ht function
and its appl
ica
t
ion in
w
a
ter ri
g
h
ts alloc
a
tion s
y
stem.
Open C
y
bern
e
tics and
Systemic
s
Journ
a
l
. 20
13; 7(1) 1-1
0
.
[15]
SF
Liu a
nd
YD Z
hu. Stud
y on tria
ngu
lar
mode
l
an
d i
n
de
xes
in s
y
nt
hetic ev
alu
a
tio
n
of reg
i
on
a
l
econ
om
y
.
T
r
an
sactions of the
Chin
ese So
c
i
et
y of
Agricultur
a
l
Engi
neer
in
g
. 199
3; 9(2): 8-1
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.