TELKOM
NIKA
, Vol.13, No
.3, Septembe
r 2015, pp. 1
105
~11
1
2
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v13i3.1494
1105
Re
cei
v
ed Fe
brua
ry 1, 201
5; Revi
se
d Ju
ly 27, 201
5; Accepted Aug
u
s
t 12, 2015
ERP Selection Using Fuzzy-MOGA Approach: A Food
Enterprise Case Study
Joko Rato
no
*, Kudang
Boro Seminar, Yandra Ark
e
man, Arif Imam Suroso
Bogor Agr
i
cult
ural U
n
ivers
i
t
y
,
Jl. Ra
y
a
P
a
ja
j
a
ran – Bo
gor 1
615
1
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: joko.raton
o@
gmail.c
o
m
A
b
st
r
a
ct
Selection
of E
n
terpris
e
Resour
ce Planning
(ERP) system
is a
c
o
m
p
lex decision-
m
ak
ing proces
s
and
on
e of th
e critica
l
succ
ess factors (C
SF
s) in
ERP
ado
ptio
n life c
ycle. Many E
R
P i
m
pl
e
m
e
n
tati
o
n
failur
e
s are ca
used by i
m
pr
o
per pack
age s
e
lecti
on.
Vario
u
s appr
oach
e
s
have be
en us
ed, but not usi
n
g
opti
m
i
z
at
ion
te
chni
ques. T
h
is
study
deve
l
o
p
ed
a F
u
zz
y
-
M
u
ltio
bjectiv
e
G
enetic
Alg
o
rith
m (F
u
z
z
y
-MO
GA)
appr
oach to o
p
timi
z
e
the q
u
a
lity of ERP selecti
on crit
eri
a
that compli
e
s
w
i
th
ISO250
10 qu
ality stan
dard
and c
o
st. T
he mo
del w
a
s v
a
lid
ated
by th
e exp
e
rts.
A case study w
a
s
cond
ucted o
n
an agr
o-i
ndus
trial
compa
n
y. The
result s
how
s
the
appr
oac
h
of Fu
zz
y
-MOGA w
i
th NSGA-II meth
od fa
cilitate
a c
o
mplex
decisi
on-
makin
g
for ERP sele
ction opti
m
ally.
Ke
y
w
ords
: ER
P selectio
n, F
u
zz
y
-MOGA, NSGA-II, ISO25010, Agro-In
d
u
s
try
Copy
right
©
2015 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
ERP sy
stem i
s
te
chn
o
logy
strategy th
at i
n
tegr
ate
s
a
set of bu
sine
ss fun
c
tion
s,
such
a
s
finance,
HR and purch
asi
ng,
with ope
rational
a
s
p
e
ct
s, su
ch
a
s
manufa
c
turi
n
g
or di
stributi
on,
throug
h tight
linka
ge
s from
ope
ration
al
busi
n
e
ss t
r
an
sa
ction
s
to fi
nan
cial
records [1]. ERP i
s
a
techn
o
logy e
nable
r
of corporate
strate
gy, but t
he fa
ilure rate of
ERP proje
c
ts in 2008 rea
c
hed
51% [2] and Panorama [3] repo
rted in 2013 that 40
% of ERP project
s
wa
s u
n
su
cce
ssful.
One
cau
s
e
of f
a
ilure
was
impro
per pa
ckage
and
ineffec
t
ive s
e
lec
t
ion [2], that affec
t
ed
impleme
n
tation failure [4]. ERP selecti
on is a com
p
l
e
x decisi
o
n
-
makin
g
pro
c
e
ss [5] and on
e of
the CSF
in
ERP a
doption
lifecycle
[6, 7
]. Selectio
n
o
f
ERP i
s
the
activity asso
ciated
with th
e
pro
c
e
s
ses, m
e
thod
s and t
ools u
s
e
d
to determi
ne ERP vendo
rs
and implem
e
n
tation co
nsu
l
tants
(vendo
r). E
R
P sele
ction
must b
e
p
r
o
c
ee
d with
carefully b
e
ca
use
of an
i
m
pact i
n
h
e
l
p
ing
comp
anie
s
to
build
com
pet
itive advanta
ge [8]. Select
ion criteri
a
al
so affe
ct the
su
ccess of E
R
P
impleme
n
tation and e
s
pe
cially in agro
-
indu
stry t
here are sp
eci
a
l
characte
risti
cs that mu
st be
covered by ERP system
like prod
uct
safety,
seasonal, peri
s
h
a
b
le, bulky an
d sho
r
t delivery
cycle
s
an
d expired d
a
te.
Analitycal
Hi
era
r
chy Pro
c
ess (A
HP) a
nd Fu
zzy we
re the
mo
st
popul
ar
anal
ysis to
ols
and wid
e
ly used in the ERP sele
ction
[9]. Data
Envelopme
n
t Analysis (DEA)
approa
ch ca
n be
applie
d to d
e
termin
e the
score
of the sel
e
cti
on
criteria fo
r ERP vendor [1
0]. AHP use
d
to
determi
ne th
e wei
ghting
of the tiered
crite
r
ia a
nd t
he final
sco
re [10, 11]. Analytic Hi
era
r
chy
Process
(ANP) u
s
ed
to
overcome
the
weakne
sses of
AHP to
be
m
o
re
flexible
wi
th feedb
ack [
13,
14].
Fu
zzy was used
to
calcul
ate
the
score
[15,
16
], with fu
zzy-AHP [17,
18]
and
u
s
ed
fu
zzy
ANP [19]. Decisi
on Supp
ort System (DSS) wa
s applie
d to the criteri
a
in the Balance Score
Ca
rd
(BSC) [20]. A
n
hybri
d
of A
N
P an
d An
al
ysis
Ne
ural
Network (ANN) wa
s
used to
determine
the
weig
ht of each criteri
on a
nd to transfo
rm into
the final score. Ozalp et al. [23] applied thre
e
approa
che
s
:
AHP, Fuzzy-AHP and ANP, to select
E
R
P co
nsultan
t
and re
sulte
d
the sam
e
rank.
Fuzzy-G
oal P
r
og
rammi
ng
wa
s u
s
e
d
with optimi
z
atio
n [24]. Pri
n
ci
ple
Comp
one
nt Analysi
s
(PCA
)
wa
s ap
plied
to red
u
ce
crite
r
ia [25]
and
De
cisi
on
Maki
ng T
r
i
a
l and Eval
uation L
abo
ratory
(DEMATEL
) use
d
to find the cau
s
al rel
a
tionship
s
be
tween criteri
o
n [26]. Many
studie
s
rel
a
te
d to
the method
of ERP sel
e
ction wa
s d
e
termini
s
tic
wit
h
previo
us
preferen
ce
s a
n
d
few a
r
e u
s
ing
optimizatio
n tech
niqu
es
without pr
eferen
ce
s. Othe
r to
ols i
s
to si
m
p
lify the criteri
a
, but rem
o
ves
the origi
nal
meanin
g
of
the stan
dard
crite
r
ia. It need
s the
d
e
velopme
n
t of optimizati
on
techni
que
s without
prefere
n
ce
for a co
mplex
sta
nda
rd crite
r
ia wit
hout simplification crite
r
ia so
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
9
30
TELKOM
NIKA
Vol. 13, No. 3, September 20
15 : 1105 – 11
12
1106
that the origin
al meanin
g
of
the
crite
r
ia can be mai
n
tai
ned. Many se
lection
crite
r
i
on develo
ped
in
the previou
s
study but even forget
the IS
O25010 qu
ality crite
r
ia that have bee
n wel
l
stand
ardi
ze
d. ISO2501
0
consi
s
ts
of sta
ndard softwa
r
e
qu
ality
cov
e
ring
8 ch
ara
c
teri
stics with
31
sub
-
cha
r
a
c
teristics and q
u
a
lity in use with 5 ch
ar
a
c
teri
stics and
11 sub
-
cha
r
acteri
stics. While
the co
st ha
s been a
defi
n
ite crite
r
ia i
n
any ER
P selectio
n whi
c
h incl
ude
s th
e total co
st
and
financi
ng with
5 sub characteristics.
This stu
d
y d
e
velope
d a
hybrid
app
ro
ach
Tri
a
ng
ul
ar F
u
zzy
–
Multiobje
c
tive Gen
e
tic
Algorithm (F
u
zzy
-MOGA
) to cond
uct o
p
timizati
on of
ISO25010 q
uality and co
st crite
r
ia. Fu
zzy-
MOGA approach is proposed as
an
alternative solution approach
based
on com
puting
i
n
telligent
system
s
with
optimization
techni
que
s without
p
r
efe
r
en
ce and si
mplification criteria
to
add
ress
the compl
e
x deci
s
io
n-m
a
ki
ng pro
c
e
s
s of ERP system
sele
ction.
2. Approac
h
Dev
e
lopme
n
t
2.1. Criterion
of
ERP Selection
ISO25010
qu
ality standa
rd
is o
ne
of the
most
i
m
po
rtance
crite
r
ia f
o
r E
R
P sele
ction [9].
ISO25010
software
qu
al
ity defines
system
and
software
q
u
ality model
s for the
e
i
ght
cha
r
a
c
t
e
ri
st
ic
s an
d 31
s
u
b
-
ch
ar
act
e
rist
i
cs.
T
he m
o
d
e
l incl
ude
s
th
e quality of
software
prod
uct
and the
quali
t
y of use ISO/IEC 2501
0
[27]. The
ch
ara
c
teri
stics can be see
n
in
Figu
re
1 and
Figure 2. The measurem
ent on su
rve
y
was appli
e
d the SQua
RE method
of ISO25023
with
some a
daptat
ions. Anothe
r criteri
on that
is oft
en use
d
is the co
st of ERP adoptio
n [14], [18-22
],
[28] whi
c
h
co
nsi
s
t of impl
e
m
entati
on co
st
s (lic
en
se
s,
con
s
ult
i
n
g
,
inf
r
ast
r
u
c
t
u
re
),
sup
port
i
n
g
co
st
s
and oth
e
r
co
sts (hi
dden
co
sts) an
d
characteri
stics
of financi
ng fro
m
vendo
r
o
r
finan
cial firm,
se
e
Figure 3. T
h
e weighting
of crite
r
ia
wa
s d
e
te
rmi
ned
from E
R
P e
x
perts
su
rve
y
on
con
s
um
er
prod
uct go
od
s indu
stry.
Figure 1. Cha
r
acte
ri
stics a
nd su
b-
cha
r
a
c
teri
stics of software p
r
o
d
u
c
t quality crite
r
ia (ISO2
501
0)
Figure 2. Cha
r
acte
ri
stics a
nd su
b-cha
r
a
c
teri
st
ics of q
uality of use criteria (ISO 2
5010
)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
9
30
ERP Selec
t
ion Us
ing Fu
zz
y-M
OGA
A
p
p
r
oa
ch:
A
Foo
d
E
n
t
e
rpri
se
Ca
se S
t
udy
(
J
o
k
o R
a
t
o
no
)
1107
Figure 3. Cha
r
acte
ri
stics a
nd su
b-cha
r
a
c
t
e
ri
st
ic
s of
t
h
e co
st
crit
e
r
ia
2.2. Scoring w
i
th Fu
zzy
-
M
OG
A
2.2.1. Triangular Fuzz
y
a
nd MOG
A
The usefulne
ss of fuzzy set theory is to quantify the con
c
ept of
fuzzine
ss in
human
thought. Tri
a
ngula
r
fu
zzy
widely u
s
e
d
becau
se it i
s
easy in
the
calcul
ation [16
]. Linguisti
c
t
e
rm
s
use
d
in this study includ
e the wei
ght an
d score
with the symbol
an
d the membe
r
shi
p
functio
n
as
see
n
in Table
1 [16, 29].
Table 1. Te
rms and m
e
m
bership fun
c
ti
ons
Score
Ver
y
P
oor
Poor
Fair
Go
od
Ver
y
Go
od
S
y
mbol
VP
P
F
G
VG
Members
(0;0;0.2
)
(0;0.2;0.4
)
(0.3;0.5;0.
7)
(0.6;0.8;1
)
(0.8;1;1
)
Cost
Ver
y
C
h
ea
p
Cheap
Fair
Expensi
v
e
Ver
y
Ex
pensi
v
e
S
y
mbol
VC
C
F
E
VE
Members
(0;0;0.2
)
(0;0.2;0.4
)
(0.3;0.5;0.
7)
(0.6;0.8;1
)
(0.8;1;1
)
Trian
gula
r
fuzzy wa
s a
pplied fo
r
weig
hting a
nd sco
r
ing
of quality a
nd cost.
Defu
zzifi
catio
n
of the wei
g
ht and value
of quality to
be cal
c
ulate
d
by the
Cente
r
of Gravity (CoG)
techni
que.
T
he o
p
timizat
i
on p
r
o
c
e
s
s wa
s
appl
i
ed M
OGA
evolutiona
ry app
r
oa
ch
with
Non
domin
ate
d
Sorting Ge
netic Algo
rith
m II (NSG
A-II) method. NS
GA-II was int
r
odu
c
e
d
by Deb
et al
. [30] i
s
a ge
netic
alg
o
rithm fo
r m
u
lti-obje
c
tive
functio
n
which
is
one
of the
be
st metho
d
s
to
gene
rate p
a
reto optimum
solutio
n
[31] and to
be th
e
base of MO
GA optimization develo
p
m
ent
[32, 33]. Fuzzy
-MOGA
algorith
m
is sho
w
n in
Figure 4. Pareto optim
a
l solution i
s
a
nond
ominate
d
sol
u
tion in t
he criteri
on
space or
an ef
ficient or
an o
p
timal sol
u
tio
n
in the d
e
ci
sion
spa
c
e. It i
s
t
o
a fe
asi
b
le
solutio
n
a
r
ou
nd
whi
c
h the
r
e i
s
n
o
way
of imp
r
oving
any o
b
jectiv
e
without d
e
g
r
ading
at lea
s
t one oth
e
r
obje
c
tive. Th
e fitness val
ue is the val
ue of the fitn
ess
function. Pa
reto optimal
solution p
r
ovi
de all th
e be
st fitness val
ue with
certa
i
n ch
rom
o
so
me
values
. In this c
a
se, c
h
romos
o
me value
is
def
ine
d
as
partici
pation
coeffici
ent of vendor.
Figure 4. Fuzzy-MO
GA Algorithm
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
9
30
TELKOM
NIKA
Vol. 13, No. 3, September 20
15 : 1105 – 11
12
1108
Quality and cost obje
c
tive functio
n
s utili
ze the expone
ntial function
as follo
ws:
(
1
)
(
2
)
X
j
is pa
rticip
ation coeffici
ent of vendo
r
j
, w
h
e
r
e 0
≤
X
j
≤
1 a
nd con
s
traint
.
Optimizatio
n
with
con
s
trai
nt
function
s can be
solve
d
in several
ways, by a
d
d
ing a
pen
al
ty
function
on t
he fitness fu
nction [3
4], [35] or by
setting the
co
nst
r
aint a
s
the
o
b
jective fun
c
t
i
on,
but co
nstraint
violation wa
s very hig
h
[34], so
thi
s
rese
arch a
ppli
ed the pe
nalt
y
function. T
he
equatio
ns of
fitness fun
c
tion
f1 = - Q +
Penalty
a
nd
f2
=
C +
Penalty
,
with the aim to get t
h
e
value of
X
j
and notation a
s
follow:
X
j
: participatio
n
coefficie
n
t of vendor
j
W
i
: quality weight
i
V
ijk
: quality score
i
, vendor
j
and expert
k
M
i
:
cost
wei
ght
i
C
ijk
:
cost
s
c
o
r
e
i
, vendor
j
and
expert
k
K
: total expert numbe
r
Penalty function is form
ulat
ed:
|
2
∑
|
(
3
)
Chromo
som
e
codin
g
is very importa
nt
in genetic
a
l
gorithm
s [36
], which is a
ll of possi
bili
ty
solutio
n
s to the pro
b
lem, can be seen in
Figure 5.
Figure 5. Chromosome
with
j
gene
Chromo
som
e
length of
j
g
ene indi
cate
s the numb
e
r
of ERP vend
ors, the
gre
a
t
er of
X
j
value
sho
w
s that E
R
P vend
or
h
a
s
gre
a
ter
pa
rticipatio
n
in the optimiz
a
tion to obtain t
h
e bes
t
fitnes
s
value.
2.2.2. Parameter o
f
NSG
A
-II
No
paramete
r
s insta
n
t on
all the fu
nctio
n
s
an
d
ci
rc
u
m
st
an
ce
s [
3
7
,
38]
,
but
t
h
e
res
u
lt
of
studie
s
[39]
defined th
e
size of the p
opulatio
n (
PopSiz
e
) =
10
n
whe
r
e
n
i
s
the numb
e
r
of
deci
s
io
n vari
able
s
. Cro
s
s
over i
s
a
ge
n
e
tic o
per
ator
that com
b
ine
s
two in
dividual p
a
re
nts
who
will
produce two children.
Probab
ility of cross-over
was determi
ned bet
ween
0.9 to 1.0 [39].
Mutation is
geneti
c
pro
c
esse
s that chang
e the
value of a ge
ne in a ch
ro
moso
me in the
popul
ation. Proba
bility of mutation wa
s defined by 1/
PopSiz
e
a
n
d
the numb
e
r
of gene
rati
ons
(
Nb
Gen
) =
1.4x
PopSize
[39]. While Devired
d
y and Reed [4
0] determin
e
d
the numb
e
r of
gene
ration
s = 2x
PopSize
.
2.3. Model Validation
Model valida
t
ion wa
s co
n
ducte
d by carryin
g
out a
n
expert su
rvey to give
a value
betwe
en 0
-
10
0 agai
nst F
u
zzy-MO
GA too
l
s for
usi
ng in
the ERP
sel
e
ction
and
pe
rforme
d t test
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
9
30
ERP Selec
t
ion Us
ing Fu
zz
y-M
OGA
A
p
p
r
oa
ch:
A
Foo
d
E
n
t
e
rpri
se
Ca
se S
t
udy
(
J
o
k
o R
a
t
o
no
)
1109
The num
be
r of experts
wh
o meet the cri
t
eria
of co
mp
etence, experience and int
e
grity gaine
d
23
experts of SA
P ERP in Indone
sia.
3. A Case Stud
y
A
cas
e
st
udy
simulat
i
o
n
f
o
r
E
R
P
sele
ct
io
n
usin
g Fu
zzy-MOGA was cond
ucte
d in
bake
r
y
food compa
n
y
, PT NIC. The
comp
an
y has b
een
su
ccessful i
n
ERP impl
ementation
a
nd
awarded as
t
he bes
t
prac
tic
e
implement
ation of
SAP
ERP. As
the agro-indus
t
rial c
o
mpany, raw
material a
nd
prod
uct of P
T
NIC h
a
ve the charac
teris
t
ics
:
produc
t s
a
fety
, sea
s
onal, pe
risha
b
le,
sho
r
t delivery cycles a
n
d
sho
r
t expire
d date t
hat sho
u
ld be h
andle
d
by ERP system.
The
weighting of
criteria in each
of hierarchi
c
al level
was
determi
ned by SAP ERP expert survey in
Indone
sia [9], while PT NI
C cro
ss fun
c
ti
onal man
age
ment team d
e
cid
e
d the score of criteri
a
for
the vendors.
4. Results a
nd Analy
s
is
Model has been validated
by SAP ERP
experts
and
gain average score 79.78
on scale
0-10
0. Since
the score mo
re t
han stand
ard certification
pa
ssi
ng
value 70, then
we can
con
c
l
u
d
e
the model val
i
d, relative sig
n
ificant with t test.
(a)
(b)
(c
)
(d)
Figure 6. Plot of Pareto optimum sol
u
tio
n
s
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No. 3, September 20
15 : 1105 – 11
12
1110
NSGA-II p
r
o
c
edure
wa
s
execute
d
in
an
open
source
appli
c
ation
so
ftware
of S
c
il
ab 5.4.1
with
optim
_n
sga
2
fun
c
tion
. Adaptation
wa
s ma
de in
the fitness f
unctio
n
, the
obje
c
tive fun
c
tion
and
penalty, dimensions, t
he number
of decisi
on va
ri
ables and
parameters
of
NSGA-II. In t
h
is
study, popul
a
t
ion size (
PopSiz
e
) wa
s u
s
ed 5
0
with
prob
ability of cro
s
sover 0.
9 and mutati
on
0.02 an
d n
u
m
ber
of ge
ne
ration
s (
Nb
G
e
n
)
5, 10, 2
0
, 70. Th
e resu
lt of pareto o
p
timum
soluti
ons
can
be seen i
n
Figu
re 6. Fi
gure
6(a
)
, 6(b) an
d
6(c) shows ho
w th
e pro
c
e
s
s to
wards
co
nverging
with the g
r
eat
er nu
mbe
r
of
gene
ration
s
(
NbG
e
n
). In th
is case, the a
m
ount of g
e
n
e
ration
20 h
a
s
a relatively converg
ent re
sults. The
s
e
results
indi
ca
te the NSGA-II is an efficient method i
n
terms of
com
puting, u
s
in
g
elitism
an
d
cro
w
d
ed
co
m
paratio
n o
p
e
r
ators that
ma
intain dive
rsit
y,
without u
s
ing
a wide ra
ng
e of additiona
l param
et
ers and the non
-dominate
d
so
rting procedu
re,
resulting in a
faster
conve
r
gent pro
c
e
s
s.
Results in
a
c
cordan
ce
with the rule
s of
thumb [39] with
the numbe
r o
f
generatio
ns
1.4x50 = 70
can be seen in
Figure 6
(
d
)
.
Since M
OGA
provide
s
pa
reto optimum
solutio
n
s
whi
c
h on
e an
d o
t
hers
non d
o
m
inated
solutio
n
then
we
can
cho
o
s
e o
ne soluti
on in pa
reto
optimum
solu
tions. All sol
u
tion in the pa
reto
optimum solu
tion rate ven
dor
X1
a
s
th
e best fitne
s
s value with th
e parti
cipatio
n coeffi
cient. To
make
cle
a
r, b
y
taking on
e of the clu
s
ter
solutio
n
(Fig
u
r
e 6(d)), obtai
ned fi
tness v
a
lue an
d ven
dor
partici
pation
coeffici
ent a
s
p
r
e
s
ente
d
in Tabl
e 2.
Vendo
r
X1
got the hi
g
hest p
a
rti
c
ip
ation
coeffici
ent score, follo
wed
by vendo
r
X4
. Top ma
na
gement of P
T
NIC fin
a
lly cho
o
se
X1
as
a
vendor a
nd consultant for ERP implem
entation in
the comp
any after con
s
id
erin
g the final score
and vend
or commitment to
post-im
plem
entation servi
c
e
s
.
Table 2. Fina
l sco
re of fitness value an
d partici
patio
n coeffici
ents
Fitne
ss Valu
e
Participa
t
io
n Co
efficie
n
ts
f1
f2
X1 X2 X3
X4 X5
-3.235349
2.339288
0.866373
0.173267
0.100742
0.568810
0.290837
5. Conclusio
n
Fuzzy-M
OGA
appro
a
ch h
a
s be
en dev
elope
d and
applie
d to assi
st mana
g
e
ment in
makin
g
comp
lex and co
mp
licated d
e
ci
si
ons o
n
ERP
sele
ction p
r
o
c
e
ss. Th
e ap
proa
ch
ha
s a
l
so
been valid
ate
d
by experts
and throug
h a ca
se
study
simulatio
n
on
mass ba
ke
ry food enterpri
s
e
in Indone
sia.
Thus, the Fuzzy-MO
GA is to be
one
of the best alternatives a
p
p
roa
c
h fo
r ERP
sele
ction
with
optimization
of important selectio
n crite
r
ia.
5.1. Future
Resear
ch
Advance
d
re
sea
r
ch i
s
to
develop
opti
m
izat
ion
tech
nique
s
with
more
than
two fitness
function
s an
d
many con
s
traints.
Ackn
o
w
l
e
dg
ements
The Fu
zzy -
MOGA devel
oped in thi
s
work i
s
a part
of SMART-T
I
N © resea
r
ch proje
c
t,
Bogor Ag
ricul
t
ural Unive
r
sity.
Referen
ces
[1]
Ra
yn
er N, W
o
ods J.
ERP Str
a
tegy: W
h
y Do
You N
e
e
d
On
e an
d Key
Con
s
ider
ations for
Defini
ng On
e.
Gartner RAS C
o
re Res
earch.
201
1; 2(4):1-9.
[2]
Ghosh
R. A
C
o
mpre
hens
ive
Stud
y o
n
E
R
P
F
a
ilures
Stress
ing
on
R
e
luct
a
n
ce to
C
h
a
nge
as
a
Caus
e
of Failure.
Jour
nal of Marketi
n
g and Ma
na
ge
me
nt
. 2012; 3(
1): 123-1
34.
[3]
Panor
ama
CG.
2013 ERP R
e
port.
A Panora
m
a Cons
ultin
g
Soluti
ons Res
e
arch Re
port. 2013.
[4]
Gupta PC, Kumar A. Evalua
tion of ERP Li
fe
C
y
cl
e on th
e Scale of ER
P Implementat
ion F
a
il
ure.
Internatio
na
l Journ
a
l of Res
e
arch
in IT
, Manage
ment an
d Engi
neer
in
g
. 20
12; 2(8): 61-6
8
.
[5]
Munke
l
t T
,
Volker S. ERP s
y
stems: aspects of
selectio
n, impleme
n
tatio
n
and susta
i
na
b
l
e op
erati
ons.
Internatio
na
l Journ
a
l of Infor
m
at
i
on Syste
m
s and Proj
ect Mana
ge
me
nt
. 201
3; 1(2): 25-
39.
[6]
Dezd
ar S, Aini
n S. Critical
S
u
ccess F
a
ctor
s for Erp Impl
ementati
on: In
sights from a
Middl
e-Easter
n
Cou
n
tr
y
.
Mid
d
l
e
-East Journ
a
l
of Scientific Re
search
. 20
11;
10(6): 79
8-8
0
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
ERP Selec
t
ion Us
ing Fu
zz
y-M
OGA
A
p
p
r
oa
ch:
A
Foo
d
E
n
t
e
rpri
se
Ca
se S
t
udy
(
J
o
k
o R
a
t
ono
)
1111
[7]
Sanchez NG,
Bernal L
EP.
D
e
termin
a
tion
of
Critica
l
Succ
e
ss F
a
ctors in
Impleme
n
tin
g
a
n
ERP
S
y
stem:
A Field Study
in Mexican Enterprises.
Infor
m
ati
on T
e
ch
no
logy for Dev
e
lo
pment.
201
3; 1
3
(3): 293-
30
9.
[8]
Uta A, Intorsurean
u I, Mihalc
a
R. Crit
eria for
T
he Selectio
n of ERP soft
w
a
re.
Informatic
a Econo
mica.
200
7; 2(42): 63
-66.
[9]
Ratono J, Seminar KB, Arkem
an Y, Suro
so AI.
T
he Develo
pment of Enterpris
e
Res
ource Pl
ann
in
g
(ERP) Selecti
o
n Metho
dol
og
y Using D
e
lta M
ode
l.
Internatio
nal Jo
urn
a
l of Informatio
n
T
e
c
hno
logy
an
d
Business Management.
2
0
14;
31(1): 1-12.
[10]
F
i
sher DM, F
i
s
her SA, Kia
ng
MY. Evaluati
n
g
mid-lev
e
l E
R
P
soft
w
a
re
. Jour
nal
of Co
mp
ute
r
Informati
o
n
System
s
. 20
04
; 45(1): 38-46.
[11]
W
e
i CC, Ch
ie
n CF
, W
ang
MJJ. An AHP-
base
d
a
ppro
a
c
h to ERP s
ystem selecti
o
n
.
Internatio
na
l
Journ
a
l of Prod
uction Eco
n
o
m
ics.
2005; 9
6
(1
): 47-62.
[12]
Lia
o
XW
, Li Y,
Lu B. A model
for selectin
g a
n
ERP s
y
st
em base
d
on
lin
gui
stic informatio
n
processi
ng
.
Information System
s
. 20
07; 3
2
(7): 100
5-1
0
1
7
.
[13]
Sh
yur HJ. A Semi-Structure
d Process for
ER
P S
y
stem
s Evaluati
on: A
ppl
yi
ng An
al
ytic Net
w
o
r
k
Process.
Journ
a
l of e-Busi
nes
s
. 2003; 5(1): 1
05-1
22.
[14]
A
y
a
g
Z
.
Eval
uatin
g si
mulat
i
on s
o
ftw
are
altern
atives th
roug
h ANP
. Procee
din
g
s of
the 201
1
Internatio
na
l C
onfere
n
ce
on
Industria
l E
ngi
neer
ing
a
n
d
Operatio
ns M
ana
geme
n
t. K
ual
a
Lump
u
r.
201
1;
[15]
W
e
i CC, W
ang
MJJ. A comprehe
nsiv
e fram
e
w
ork for sel
e
cting an ERP s
y
stem.
Internat
ion
a
l Jour
na
l
of Project Man
age
ment.
200
4
;
22(2): 161-1
6
9
.
[16]
Asgari
M, All
a
hverd
ilo
o M,
S
a
mkha
ni S. A
Compre
he
nsiv
e F
r
ame
w
o
r
k f
o
r Se
lectin
g th
e ERP
S
y
stem
in Ira
n
Kh
odr
o
Comp
an
y.
E
u
rope
an J
our
na
l of Ec
ono
mic
s
, F
i
nanc
e a
n
d
Ad
ministrativ
e
Sci
ences
.
201
1; 38(2
011)
: 7-19.
[17]
Lie
n
CT
, Chan HS. A Selection Mo
del
for ERP S
y
stem b
y
Ap
pl
yi
ng F
u
zz
y
AH
P Approac
h
.
Internatio
na
l Journ
a
l of T
he
Co
mp
ut
er, the Internet an
d Mana
ge
me
n.
20
07; 15(3): 5
8
-7
2.
[18]
Onut S, Efendigi
l T
.
A the
o
rical mo
de
l des
i
gn for ER
P soft
w
a
re se
lectio
n proces
s under th
e
constrai
nts of c
o
st an
d q
ual
it
y
:
A fuzz
y a
ppr
o
a
ch.
Jo
urna
l of
Intelli
ge
nt & F
u
zz
y
Syste
m
s.
201
0;
21(
6):
365-
378.
[19]
A
y
a
ğ
Z
,
Özd
e
mir RG. An
intel
lig
ent
ap
proac
h to E
R
P soft
w
a
re s
e
lectio
n thro
ug
h fuzz
y ANP.
Internatio
na
l Journ
a
l of Prod
uction R
e
se
arch
. 2007; 4
5
(10)
: 2169-2
1
9
4
.
[20]
Ceb
e
ci
U. F
u
z
z
y
A
H
P-b
a
sed
decisi
o
n
su
ppo
rt s
y
st
em f
o
r s
e
lecti
ng E
R
P s
y
stems
in
te
xtil
e i
ndustr
y
b
y
usin
g bal
anc
ed
scorecard.
Jo
urna
l Expert System Ap
plic
ati
ons.
200
9; 36(
5):890
0-8
909.
[21]
Karsak EE, Öz
ogu
l CO. An
i
n
tegrated
d
e
ci
si
on m
a
kin
g
app
roach
for ERP
s
y
stem
se
lecti
on.
Journal
Expert System
Appl
icatio
ns.
2
009; 36(
1): 660
-667.
[22]
Yazga
n
HR, B
o
ran S, Gozte
pe
K. An ER
P soft
w
a
r
e
sel
e
ction pr
ocess
w
i
th
usin
g ar
tificial n
eura
l
net
w
o
rk b
a
se
d
on an
al
ytic n
e
t
w
o
r
k proc
es
s appro
a
ch.
J
ourn
a
l Exp
e
rt System Ap
plic
ations
. 2
009
;
36(5): 92
14-
92
22.
[23]
Ozalp V, Yi
git
O, Maria MC.
ERP cons
ulta
n
t
selectio
n pr
o
b
lem
usin
g AH
P, fuzz
y
A
H
P
and
ANP: A
case study
in T
u
rkey
.
E
3
Jour
nal of Busi
ness
Manag
e
m
e
n
t and Eco
n
o
m
ics
. 2012; 3(3): 1
06-1
17.
[24]
Erol I, Barut M
.
Devel
o
p
i
ng
k
n
o
w
l
e
d
ge
bas
e
d
dec
isi
on s
u
p
port too
l
for e
n
t
erprise r
e
so
ur
ces pl
an
nin
g
(ERP) soft
w
a
r
e
selection.
Ankara Ün
iversites
i
Siyasa
l Bilg
ile
r Fakültesi Der
g
isi
. 20
03; 58(
1): 72-96.
[25]
Olson D
L
, Wu DD. Multi
p
le
Criteria
Ana
l
ys
is for Eval
uati
o
n of Informati
o
n
S
y
stem Risk
.
Asia-Pacifi
c
Journ
a
l of Ope
r
ation
a
l Res
ear
ch.
2011; 2
8
(1)
:
1-25.
[26]
Jaha
nsh
ahi
H, F
a
rhadz
areh B
,
F
o
tuhi H, Gol
pour A, Mok
h
tari MB. A Ne
w
Algorit
hm for
ERP S
y
st
e
m
Selecti
on B
a
s
ed o
n
F
u
zz
y
DEMAT
E
L Approac
h.
Ad
vance
s
i
n
En
vi
ro
nm
en
ta
l
Bi
o
l
o
g
y
.
2013;
7(9):
250
9-25
21.
[27]
ISO/IEC
25010.
Systems and
softw
are engi
n
eeri
ng - Systems
an
d softw
are Quality Re
qu
ire
m
ents a
n
d
Evalu
a
tion (SQ
uaRE) - Qual
ity mod
e
l
. 20
11.
[28]
Vorst CV. Appr
oach f
o
r Sel
e
c
t
ing ERP
Softw
a
r
e
at Mid-s
i
zed C
o
mp
ani
e
s
Reflecti
ng
Cr
itical S
u
ccess
Factors.
Journ
a
l of US-Ch
i
na
Publ
ic Ad
min
i
s
t
ration
. 201
2; 9
(
9): 1057-
10
68
.
[29]
Chu T
C
, Lin R
H
. Evaluati
ng
Supp
liers vi
a a
T
o
tal Integral
Valu
e bas
ed F
u
zz
y
MC
DM Appro
a
ch.
Asia
Pacific Man
a
g
e
ment Rev
i
ew
. 2011; 1
6
(4): 5
21-5
34.
[30]
Deb K, A
g
ra
w
a
l S, Prata
p
A, Me
yariv
an T
.
A fast elitist n
on-d
o
m
in
ated
so
rting
ge
netic
alg
o
rith
m f
o
r
mu
lti-ob
jectiv
e opti
m
i
z
at
ion: N
S
GA-II.
In: proceed
ings
of th
e p
a
ral
l
el
pr
obl
em so
lvin
g fro
m
natur
e V
I
(PPSN-VI) conference. 2002.
[31]
Reza
ei J, Davo
odi M. Multi-o
b
j
e
ctive mo
d
e
ls f
o
r lot-sizi
ng
w
i
t
h
supp
li
er sele
ction.
Internati
ona
l Jour
n
a
l
Producti
on Eco
n
o
m
ics
. 20
11; 130(
201
1): 77-
86.
[32]
Luo Y
L
, Liu M,
Hao Z
,
Liu D.
An improv
ed
NSGA
-II algorit
hm for multi-o
b
jectiv
e travel
li
ng sal
e
sma
n
prob
lem.
T
E
LKOMNIKA Indon
esia
n Journ
a
l o
f
Electrical Eng
i
ne
erin
g
. 201
4; 12(6): 441
3-4
418
[33]
Qizuan
g, Yan
g
G,
T
ang R,
Xi
ang G. Mu
lti-o
b
ject
iv
e o
p
timi
zation
al
gorith
m
s desi
gn
bas
ed o
n
s
upp
ort
vector regress
i
on metam
ode
ll
ing.
T
E
LKOMNIKA Indon
esi
an Jo
urna
l
of Electrical E
n
g
i
neer
ing
. 20
13
;
11(1
1
): 640
6-6
412.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No. 3, September 20
15 : 1105 – 11
12
1112
[34]
Ponsic
h A, Azzaro-Pa
n
tel C,
Domen
e
ch S,
Pi
bou
le
au L.
Constra
i
nt ha
n
d
lin
g strateg
i
e
s
in ge
netic
alg
o
rithms a
p
p
licatio
n to
opti
m
al b
a
tch p
l
an
t desig
n.
Ch
e
m
ic
al E
ngi
ne
er
ing
an
d Proc
e
ssing: Proc
ess
Intensificati
o
n
.
200
8; 47(3): 26
5-51
4.
[35]
Venter G. Rev
i
e
w
of Optimiz
a
tion T
e
chni
que
s
.
Enc
y
c
l
op
ed
i
a
of Aer
o
spac
e
Engi
ne
erin
g. South Africa.
201
0.
[36]
Arkeman Y, S
e
minar, KB, Guna
w
an H. Algoritm
a Ge
netik
a: T
eori da
n A
p
likas
in
ya
u
n
tu
k Bisnis
d
a
n
Industri
.
Bogor
(ID): IPB Press. 2012.
[37]
Hain
es A
L
, M
ills K
L
, Fil
lib
e
n
JJ. D
e
termi
nin
g
re
lativ
e
i
m
portanc
e
an
d b
e
st settin
g
s
for g
e
n
e
tic
alg
o
rithm contr
o
l par
ameters.
Evoluti
onary C
o
mputati
on.
20
10; 201
0(1): 1-
22.
[38]
Re
xh
epi
A, Ma
xh
un
i A, Dik
a
A.
Anal
ysis
of t
he im
pact of
p
a
rameters v
a
l
u
es on
the Ge
n
e
tic Alg
o
rithm
for T
SP.
Internation
a
l Jo
urna
l of Computer S
c
ienc
e Issues.
201
3;10(
1): 15
8-16
4.
[3
9
]
R
e
ed
P, Min
s
h
e
r
B, Gol
d
b
e
r
g
D
E
.
D
e
sig
n
i
n
g
a
co
mp
e
t
en
t si
mpl
e
g
e
n
e
ti
c a
l
g
o
r
i
t
h
m
for se
a
r
ch
and
optimiz
ation.
Water Resourc
e
s Research.
2
000; 36(
12): 37
57-3
761.
[40]
Devire
dd V,
Ree
d
P.
An
Efficient Des
i
g
n
Metho
dol
og
y for the No
ndo
minate
d
S
o
rted Gen
e
tic
Algorit
h
m
-II.
Genetic a
nd Evo
l
utio
nar
y Com
p
utati
on C
onfer
ence (GECCO
03). Chic
ago. 2
003.
Evaluation Warning : The document was created with Spire.PDF for Python.