TELKOM
NIKA
, Vol. 13, No. 4, Dece
mb
er 201
5, pp. 1466
~1
477
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v13i4.2388
1466
Re
cei
v
ed Au
gust 3, 201
5; Re
vised Sept
em
ber
8, 201
5; Acce
pted
Octob
e
r 2, 20
15
Service Engineering Based on
Service Oriented
Architecture Methodology
Suhardi
1
, Ro
bin Doss
2
, Purnomo Yus
t
ianto
3
1
School of Elec
trical Eng
i
ne
eri
ng an
d Informa
tics,
Institute of
T
e
chnol
og
y B
and
un
g, Indon
esia, 40
13
2
2
School Inform
ation T
e
chno
lo
g
y
, D
eaki
n
Uni
v
ersit
y
, Mel
b
o
u
r
ne, Australi
a, 312
5
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: suhard
i
@stei
.
itb.ac.id
1
, p
y
us
tian@
deak
in.e
du.au
3
A
b
st
r
a
ct
Service E
ngi
n
eeri
ng (SE) a
n
d
Servic
e Orie
nted Ar
ch
itectu
re (SOA) origi
nally r
e
si
de o
n
different
plate
aus
of d
i
s
c
ipli
ne. SE
is
d
e
fine
d as
a
set
of acti
v
i
ty intro
duci
ng
a
new
b
u
sin
e
ss serv
ic
e, w
h
ile
SOA i
s
a
technic
a
l a
ppr
oach to re
defi
ne an
enter
pri
s
e busi
ness
pr
ocesses as
a
set of IT
enabl
ed servic
es. T
h
is
pap
er pro
pos
e
s
a SOA emb
edd
ed SE fra
m
ew
ork as
a
c
o
mpre
hens
ive appr
oach in
re
-defin
ing busi
n
ess
service a
nd its
IT
imple
m
e
n
tation. After an
intr
oducti
on, a review
of existing SE fra
m
ew
orks and S
O
A
meth
od
olo
g
i
e
s
is pr
esente
d
i
n
the
pa
per. A
fterw
ard,
a co
mp
lete S
E
fra
m
ew
ork
is pr
o
pose
d
w
i
th se
veral
results on
earl
y
case studi
es. A survey resul
t
s are t
hen pr
e
s
ented to pr
ov
e the usa
b
il
ity and b
e
n
e
fit of th
e
prop
osed fra
m
ew
ork. T
he framew
ork is
des
ign
ed a
nd
pr
o
pose
d
to he
lp
practitio
ners a
nd res
earch
ers
to
cond
uct servic
e eng
in
eeri
ng
by empl
oyin
g p
r
incip
l
es a
nd
method
ol
ogy offered by SOA a
ppro
a
ch.
Ke
y
w
ords
: Se
rvice Eng
i
ne
eri
ng, F
r
amew
ork, SOA,
Busines
s Model C
anva
s
, Service Blue
printi
ng
Copy
right
©
2015 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
The glo
bal
shift toward
s
a se
rvice
-
b
a
s
ed
econom
y has b
r
ou
g
h
t importa
nce to the
con
c
e
p
t of S
e
rvice
Engi
ne
ering
(SE). E
norm
o
u
s
cha
nge
s in
glob
a
lized
bu
sine
ss e
n
viron
m
en
t in
recent yea
r
s
have ma
de
service
inn
o
va
tion a
critic
al prio
rity
for all busi
n
e
s
s
ma
nage
rs.
B
u
si
n
e
ss
entities comp
ete to provide
the best valu
e for cu
stom
e
r
s by taki
ng a
d
vant
age of t
he co
mpetitive
busi
n
e
ss
envi
r
onm
ent. SE method
s hav
e the pote
n
tia
l
to help the
pro
c
e
ss
of se
rvice in
novati
on
to deliver inte
nded a
nd de
signed value
o
f
a service to the cu
stome
r
.
Information
T
e
ch
nolo
g
y (I
T) h
a
s be
en
an imp
o
rta
n
t
comp
one
nt in
the recent g
r
owth
of
the se
rvice i
ndu
stry. It also takes
a central role
in
innovating n
e
w servi
c
e
s
. Therefore, a
s
a
pra
c
tical a
nd
an academi
c
field, SE requires
com
p
lem
entary analy
s
is from IT poi
nt-of-vie
w
. As a
result of thi
s
tren
d, it is
imperative th
at
the
field
sho
u
ld be a
d
vanced and
enla
r
ge
d wi
th
partici
pant
s, not only from busi
n
e
ss an
d
managem
ent
field, but should also invol
v
e practition
e
r
s
and re
se
arch
ers
with IT expertise.
The te
rm ‘
se
rvi
c
e
en
ginee
ring
’
(SE) wa
s first m
entio
ned
du
ring
th
e mid
-
nin
e
tie
s
a
s
a
techni
cal di
scipline in syst
e
m
atic develo
p
ment
and d
e
s
ign of services u
s
ing a
ppropriate m
odel
s,
method
s, a
n
d
tool
s [1]
SE encomp
a
s
ses b
r
oad
range
of a
c
ti
vities sta
r
tin
g
from
bu
sin
e
ss
strategi
c l
a
yer to
pro
c
e
s
s d
e
si
gn a
n
d
its d
e
taile
d implem
ent
ation in te
ch
nical l
a
yer.
The
introdu
ction
Service S
c
ie
nce, Ma
nag
e
m
ent and En
ginee
ring (S
SME)
as
a di
sci
pline reiterates
the emph
asi
z
e on
an in
terdiscipli
nary
approa
ch, combi
n
ing m
anag
ement
and en
gine
e
r
ing
theorie
s, in
cl
uding from IT disciplin
e, in the obj
e
c
ti
ve of improvi
ng the comp
etitiveness in
the
orga
nization
based on g
r
o
w
ing
servi
c
e
s
need
s in the
busin
ess en
vironme
n
t [2]. To handl
e such
compl
e
x und
ertaki
ng, a fra
m
ewo
r
ks a
n
d
method
s to
guide the p
r
o
c
ess is requi
re
d [3], which
can
c
o
ver through the tec
hnic
a
l as
pec
t
in the IT layer.
An emergi
ng
appro
a
ch from IT discipl
ine,
Service
Oriente
d
Architecture (SO
A
), wa
s
introdu
ce
d wi
th the capabi
lity of bridging the bus
in
e
ss p
r
o
c
e
ss (BP) requi
rem
ents with the
IT
developm
ent
pro
c
e
s
ses im
plemente
d
a
s
servi
c
e
s
co
mpone
nt [4]. Compl
e
me
nting SE with SOA
approa
ch, th
erefo
r
e,
sho
u
l
d be
an i
dea
l app
roa
c
h
fo
r bu
sin
e
ss
se
rvice
develo
p
m
ent to a
n
swer
the comp
etitiveness ch
all
enge an
d the deman
d fo
r agility in the se
rvice i
ndu
stry. Whil
e a
methodi
cal f
r
amework to
guide
a
se
rvice
engi
nee
ri
ng p
r
o
c
e
s
s i
s
req
u
ire
d
, a
formali
z
e
d
a
nd
pra
c
tical
fram
ewo
r
k which
can
gui
de
an
appli
c
atio
n o
f
the SE
with
SOA app
ro
ach is still l
a
rge
l
y
untou
che
d
. Some ea
rly propo
sition
s ha
ve already
b
een mad
e
[5] [6], but
these works did
no
t
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 : 146
6 – 1477
1467
delved into t
he detail
s
of
SOA method
ology, and
di
d not p
r
ovide
d
ca
se
studi
es a
s
a
proo
f of
c
o
nc
ep
t.
This pa
per p
r
opo
se
s a fo
rmali
z
ed SE fram
ework, spe
c
ifically embedd
ed wit
h
SOA
approa
ch. T
h
ree
charact
e
risti
c
s are
con
s
id
ere
d
i
n
formul
atin
g the p
r
op
o
s
ed f
r
ame
w
ork:
simpli
city, co
nformity and
uniformity. Th
e frame
w
o
r
k i
s
de
sig
ned
to
simplify the
pro
c
e
s
s in te
rms
of easine
s
s of adoption
and implem
e
n
tation, and
at the same time it also ensu
r
e
s
the
compl
e
tene
ss and
app
ropriate
ne
ss
of the solu
t
i
on provided
for given
situation in
an
orga
nization.
The exi
s
ten
c
e of the fra
m
ewo
r
k is
also
desi
gne
d to create th
e unif
o
rmity of se
rv
ice
engin
eeri
ng pro
c
e
ss acro
ss multiple
i
m
pleme
n
tati
o
n
s. Th
e pro
positio
n
of the fram
ework is
pre
s
ente
d
in this pa
per
with ca
se stu
d
ie
s and
survey
results on its i
m
pleme
n
tatio
n
.
2. Ser
v
ice Engineering F
r
ame
w
o
r
ks
The d
e
finition
of se
rvice
e
ngine
erin
g wi
th re
g
a
rd
to
e-servi
c
e
sco
pe is sp
ecifie
d as “
an
approa
ch that
pro
v
ide
s
a di
sci
pline for
using
m
odels a
nd tech
niqu
e
s
to guide th
e
unde
rsta
ndin
g
,
stru
cture, de
sign, im
plementat
ion, dep
loym
ent, docum
entation,
operation, maintena
nce and
m
odification of
e-servi
c
e
s
” [7]. Within this
sc
ope, the
objec
tive
of SE framework
is to provide t
h
e
servi
c
es in IT layer,
specifi
c
ally in the f
o
rm
of
web services whi
c
h
w
ill all
o
w a
high degree
of
automated i
n
tera
ction a
m
ong
se
rvices. The
r
efo
r
e,
it is high
ly desirable
to redefin
e
the
architectu
re o
f
the enterpri
s
e usin
g app
ro
ach
e
s fro
m
Service
Orie
nted Architectu
re (SOA
).
From SE poi
nt of view, two frame
w
o
r
ks for desi
gnin
g
(ne
w
)
se
rvice
s
we
re
sel
e
cted to
be a com
parative base to
propo
se the
new fram
e
w
ork. Th
e first
framework wa
s take
n from
cla
ssi
c
NS
D
con
c
e
p
t, prop
ose
d
by
Lin
a
nd
Hsi
e
h
[8], and
app
rop
r
i
a
tely nam
ed
as
Ne
w Servi
c
e
Develo
pment
(NS
D
) f
r
am
ewo
r
k. Ba
se
d on ITIL
, t
he second f
r
amework
wa
s introdu
ce
d
as
Servic
e Sys
t
em Developm
ent Process (SSDP) [9].
Both frame
w
orks p
r
op
ose
the combi
n
a
t
ion use of v
a
riou
s ma
na
gement a
nalysis tool
s
su
ch a
s
: Fea
s
ibility Study, Socio T
e
chn
o
-Economi
c
al
Analysis, En
vironme
n
tal
Scanni
ng, Trend
Analys
is
, B
C
G matrix, and Quality Func
tion
Deplo
y
ment (Q
FD). A com
pari
s
on
can
be
m
ade
betwe
en the
two focusi
ng
on the ap
pro
a
ch i
n
de
sign
ing a servi
c
e,
spe
c
ifically in modeli
ng t
h
e
serv
i
c
e.
The first fra
m
ewo
r
k, the NSD fram
ework, def
in
es t
he Service
Desig
n
as two
distinct
stage
s: (1) service
mod
e
li
ng a
nd
(2) service
im
ple
m
entation. T
he
servi
c
e m
odelin
g itself
i
s
comp
osed f
r
om fou
r
com
pone
nts:
(1)
prod
uct
mod
e
l, (2
) p
r
o
c
e
s
s m
odel,
(3
)
resou
r
ce mo
del,
and (4) ma
rketing co
ncept
. Among othe
rs, the
UML
notation is
e
m
ployed a
s
a
tool for prod
uct
model. Se
rvice
blue
printi
ng te
chni
que
is
su
gge
ste
d
a
s
a
p
r
o
c
ess m
odel
tool. Th
e ma
in
prop
osed to
o
l
s of
servi
c
e
desi
gn in
NS
D fra
m
ewor
k is
the QF
D matrix.QFD
i
s
sugg
este
d as
a
tool to visuali
z
e th
e
cu
sto
m
er
nee
ds du
ring
the
se
rvice
develo
p
m
ent p
r
o
c
e
ss.
A modified
Q
F
D i
s
also
su
gge
st
ed a
s
a
service-pl
anni
ng
matrix to
d
i
splay
cu
sto
m
er requi
re
ments, te
chn
i
cal
measures, target values, a
nd com
petitive analyse
s in
the form of House of Quali
t
y (HOQ).
The se
con
d
frame
w
o
r
k (SSDP) defi
nes
the se
rvice
d
e
si
gn and develop
ment
a
s
analyzi
ng service req
u
irem
ents into the i
dentificatio
n of (1) servi
c
e
entities funct
i
ons, (2
) servi
c
e
interfaces,
(3
) se
rvice inte
rope
ra
bility,
and (4)
se
rvice level a
g
re
ements.
Whil
e not explicit
ly
mentionin
g
SOA approa
ch
, this definition ba
sically embodi
es th
e SOA desig
n approa
ch. In
modelin
g the servi
c
e, SSDP utilize
s
a se
rvic
e
meta-mod
el
[10] to he
lp stakehold
e
rs
con
c
e
p
tuali
z
e
the servi
c
e value chain.
The two fram
ewo
r
ks
are in
here
n
tly an it
erat
ing
p
r
o
c
e
s
s of
contin
u
ous imp
r
ove
m
ent, and
thus the
re
sul
t
of each
stag
e is a fe
edb
a
ck fo
r
the
nex
t iteration. Also, tests
coul
d
be cond
ucte
d
in the end of
each
step, ensurin
g co
n
f
ormity of
the step re
sults with the inp
u
t spe
c
ificati
ons
stated in the
begin
n
ing of the step, befo
r
e deci
d
ing to
contin
ue to the next step.
A combin
atio
n and ge
neralizatio
n bet
wee
n
t
he two has al
so b
een propo
se
d, defining
the Gen
e
ral
Service E
ngi
neeri
ng F
r
am
ewo
r
k [11]. It
is
comp
osed
from fou
r
g
r
oup
s of a
c
tivities:
(1) Id
entificati
on, (2
)
De
sig
n
, (3) Develo
pment, an
d (4) O
peration.
This
pap
er i
s
a foll
ow-up
on
this
con
c
e
p
t, by providi
ng
elabo
ration
d
e
tails i
n
to the
pro
p
o
s
ed
fra
m
ewo
r
k a
nd
by focu
sin
g
t
h
e
highlight on t
he role of SO
A methodolo
g
y in SE Framewo
r
k.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Servi
c
e Engi
neeri
ng Ba
se
d on Service Oriente
d
Architecture Met
hodol
og
y
(Su
hardi
)
1468
3. SOA Meth
odologies
SOA is d
e
fin
ed a
s
softwa
r
e a
r
chitectu
re ba
sed
on l
oosely co
upl
ed softwa
r
e
servi
c
e
s
whi
c
h inte
grated into a
distri
buted
com
putin
g
system, b
y
mean
s of
se
rvice
-
o
r
ie
nted
prog
ram
m
ing
[12]. In SOA perspe
c
tive, se
rvice
s
a
r
e the
buildin
g blo
c
ks of a
n
enterpri
s
e.
An
enterp
r
i
s
e i
s
defined
by its pool
of se
rvice
s
, bot
h
available i
n
te
rnally an
d p
ublicly, an
d the
intera
ction p
a
ttern betwe
en the se
rvices. The serv
i
c
e interactio
n
is not boun
d internally i
n
an
enterp
r
i
s
e, but the busin
ess land
scap
e or the bu
sine
ss enviro
n
ment is also comp
rised
of
intera
cting se
rvice
s
amo
n
g
various o
r
g
a
n
izatio
ns
. Th
us, the well
-b
eing of an en
terpri
se a
nd the
busi
n
e
ss e
n
vironm
ent is de
termine
d
by t
he implem
ent
ation quality of these servi
c
e
s
.
Since its in
ception in th
e
mid 20
00’s,
the SOA ap
proa
ch
ha
s e
x
perien
c
e
d
g
r
owth
in
maturity an
d
rate
of a
d
o
p
tion. While
several m
a
in
co
ncepts were
inde
ed
converg
ed, S
O
A
contri
butors
are
still unable to provide on a si
ngular
cohesiv
e
view on the SOA adoption.
Variou
s met
hodol
ogie
s
a
r
e offere
d in
the form
of step-by-ste
p
guidan
ce in
desig
ning t
h
e
impleme
n
tation of SOA. F
r
om compilin
g works on
SOA methodol
ogie
s
[13] [
14] [15] [16]
[17],
we
can
coll
e
c
t at lea
s
t 15
different met
hodol
ogie
s
. T
w
o mo
st po
p
u
lar SOA m
e
thodolo
g
ies a
r
e
(1) IBM'
s SOMA [18] and (2
) Thom
as Erl'
s SO
A [19], often
labeled a
s
Mainst
ream
SOA
methodol
ogy (MSOAM) [15
]
.
IBM’s SOMA is a highly re
comm
end
ed SOA method due to its co
mpre
hen
sive
ness and
its vast indu
st
rial ad
option [
20] [14] [17]. It cove
rs
a co
mplete cy
cle
of servi
c
e en
ginee
ring, fro
m
busi
n
e
ss
sid
e
to the technical impl
em
entation.
The
method co
n
s
ist
s
of six main stag
es:
(1)
busi
n
e
ss tra
n
sformation
analysi
s
, (2
) identificatio
n, (3)
spe
c
i
f
ication, (4
) reali
z
ation,
(5)
impleme
n
tation, and (6) d
eployment
-m
onitorin
g
. Each
sta
ge in
-tu
r
n con
s
ist
s
of three to four
sub
st
age
s.
Table 1. Stag
es of SOMA Methodol
ogy [1]
A variation of SOMA prop
ose
s
the u
s
e
of Compon
e
n
t Busine
ss
Model (CBM), as an
analytical too
l
in the early stage of the
methodol
o
g
y, during sta
g
e
1 to 3. CBM is basi
c
all
y
a
matrix of enterprise’s
busine
ss
com
petencies against account
ability level, i.e. directing
,
controlling
an
d executing [21]. The resulting elem
en
ts of the mat
r
ix are d
e
fine
d as b
u
si
ne
ss
comp
one
nts.
Each b
u
si
ne
ss compo
nent
is defin
ed to
have attribute
s
of: (1
) pu
rp
ose, (2)
activi
ty,
(3)
re
so
ur
ce, (4)
gov
e
r
na
n
c
e,
a
nd (5
) s
e
rv
ice
s
offe
re
d. The
matrix
ca
n b
e
u
s
e
d
to determine
the
importa
nce of each b
u
si
ne
ss
com
pone
n
t, and ulti
mately to define new
servi
c
e o
ffering.
A compl
e
te
SOMA meth
odolo
g
y, therefore,
can
b
e
co
nsi
dered
as SE fra
m
ewo
r
k in
itself. Yet, ad
opting SO
MA inde
pend
entl
y
has be
en
made
difficult
due
to the
li
mited availa
b
ility
of detailed
SOMA reference ma
terial.
Combining
SOMA with
SE framework therefore
will
necessitate t
he
removal
of
seve
ral
pa
rt
of the
m
e
thod
ology d
ue to
t
he p
r
o
c
e
s
s re
dund
an
cy. Wi
th
the help f
r
o
m
a detail
e
d
referen
c
e,
SOMA meth
odolo
g
y ca
n
still be
useful in buil
d
ing
SE
frame
w
ork, e
s
pe
cially in th
e spe
c
ificatio
n stage a
nd o
n
wa
rd.
Another well
-kno
wn
S
O
A methodol
ogy is
the
one
de
vised by T
h
o
m
as E
r
l. In h
i
s bo
ok
[19-6] he el
a
borate
d
a def
initive refere
n
c
e fo
r SOA i
m
pleme
n
tatio
n
, MSOAM. The metho
d
o
l
og
y
con
s
i
s
ts of se
ven stage
s of
activities: (1
) Ontol
ogy De
finition, (2) B
u
sin
e
ss Mo
d
e
l Alignment, (3)
Service O
r
ie
nted Analysi
s
, (4) Service Oriente
d
De
s
i
gn, (5) Se
rv
ice D
e
v
e
lopm
ent, (6) Serv
i
c
e
Testing, a
nd
(7) Se
rvice
Deployment. T
he emp
h
a
s
is
of MSOAM elaboration i
s
durin
g an
alysis
and de
sig
n
a
c
tivities. This
is evident by the num
b
e
r of
sub-stag
es d
e
fined for the
two stag
es.
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 : 146
6 – 1477
1469
While
providi
ng detail
ed p
r
escription
o
n
analy
s
is
a
nd de
sig
n
a
c
tivities, MSOAM doe
s
not provide
el
aborate g
u
ida
n
ce
for
activities
dur
i
ng bot
h
prelimina
r
y busi
n
e
s
s
anal
ysis and
po
st
-
desi
gn. The
s
e lacks m
a
ke it some
wh
at appr
opri
a
te to be co
mbined
with
SE framework.
MSOAM met
hodol
ogy st
re
ngth in
analy
s
is and
de
sig
n
sta
ges is q
u
ite usef
ul to
be ad
opted,
but
care
sho
u
ld
be ta
ken
to
ensure
that n
o
a
c
tivity red
unda
ncy i
s
p
e
rform
ed. Ad
ditional
activities
sho
u
ld al
so
be define
d
in
the SE framewo
r
k to fill the gaps fo
r pre
-
de
sig
n
and po
st-d
esign
activities in original MSOA
M.
Table 2. Stag
es of MSOA
M Methodol
o
g
y [6]
1.
Define Re
lev
a
nt Ontology
2.
Align Rele
va
nt
B
u
s
i
nes
s
Models
3.
Per
f
orm
Ser
v
ic
e Ori
e
nt
ed A
n
al
ysi
s
4.
Per
f
orm
Ser
v
ic
e Ori
e
nt
ed D
e
s
i
gn
5. D
e
ve
lop Se
rvic
es
6. D
e
ve
lop Tes
t
S
e
rv
ice
Oper
at
ions
7. De
ploy S
e
rv
ice
s
3.
1
.
De
f
i
ne
B
u
s
i
n
e
s
s
Re
q
u
ir
em
en
t
3
.
2.
Id
ent
if
y Aut
o
m
a
ti
o
n
S
ystem
3
.
3.
Mo
de
l Ca
nd
id
ate
Se
r
v
ice
s
3
.
3
.
1.
De
c
o
m
p
o
s
e B
u
s
i
ne
s
s
Pr
oc
e
s
s
3
.
3
.
2.
I
d
e
n
t
i
f
y
O
p
era
t
io
n Ca
nd
id
at
e
3.
3.3
.
Ab
stra
ct Or
ch
estr
at
io
n L
o
g
i
c
3.
3.4
.
C
r
eat
e S
e
r
v
ice
C
a
n
d
i
d
a
t
es
3.
3.5
.
R
e
fi
n
e
&
Ap
pl
y Se
rvi
c
e
Or
ie
nt
ati
o
n
3.
3.6
.Ide
n
ti
fy Ser
v
i
c
e
C
o
mp
osi
t
i
o
n
s
3.
3.7
.
R
e
vi
se O
per
a
t
i
on
Gr
oup
i
n
g
3.
3.8
.
An
al
yze
Pr
ocessi
n
g
R
e
q
u
i
r
em
en
ts
3
.
3
.
9.
I
d
en
t
i
f
y
A
p
pli
c
at
io
n
S
e
r
v
ic
e
O
p
er
at
io
ns
4.
1
.
Co
mp
os
e
SO
A
4
.
2. D
e
sign Ent
i
t
y
-
C
e
n
tric
Bu
si
n
e
ss
S
e
rv.
4.
3
Des
i
gn
Ap
pl
ic
at
io
n
S
e
rv
ic
e
s
4.
1.1
.
C
h
o
s
e S
e
r
v
i
c
e L
a
yer
4.
1.2
.
Po
si
ti
on
Co
r
e
St
and
ar
ds
4.
1.3
.
C
h
o
o
se
SOA
Exte
nsi
o
n
s
4
.
4. D
e
sign Task-Cent
r
i
c
B
u
sine
ss
Se
r
v
.
4
.
5 D
e
sig
n
S
e
r
v
. O
r
i
e
n
t
ed
Bu
si
ness
Pr
ocess
3.
3.1
0
. C
r
ea
te
App
l
. Se
rvi
c
e
C
and
i
dat
es
3.
3.1
1
.R
evi
s
e
Se
rvi
c
e
C
o
m
posi
t
i
o
ns
3.
3.1
2
. R
e
vise
Op
er
ati
o
n Gr
o
upi
n
g
4.4
.
1.
D
e
fi
ne
w
o
r
k
fl
ow
l
o
gi
c.
4
.
4.2
.
De
rive
in
itial
i
n
te
rface
.
4.4
.
3.
Ap
pl
y Se
rvi
c
e
Or
ie
nta
t
i
o
n
.
4.4
.
4.
Sta
n
d
a
r
d
i
z
e S
e
r
v
i
c
e In
te
rfa
c
e.
4.4
.
5.
Ide
n
ti
fy
Re
qu
ir
e
d
P
r
o
c
essi
ng
.
4.5
D
e
si
gn
Se
rv.
Or
i
ent
ed
Busi
n
e
ss Pr
o
c
ess
4.3
.
1.
R
e
vi
ew
E
x
ist
i
n
g
S
e
r
v
i
c
es.
4.3
.
2.
C
onf
ir
m
C
o
n
t
ext
4.3
.
3.
Derive In
itial In
terface
4.3
.
4.
Ap
pl
y Se
rvi
c
e
-
Or
i
e
n
t
at
io
n.
4.3
.
5.
Sta
n
d
a
r
d
i
z
e S
e
r
v
i
c
e In
te
rfa
c
e
4.3
.
6.
Ad
d Sp
ecu
l
a
t
i
v
e Fe
atu
r
es
4.5
.
1.
Ma
p Out Inte
ra
cti
o
n
Sce
nar
i
o
s
4.5
.
2.
D
e
si
gn
Pr
oce
ss Se
rvi
c
e
Int
e
r
f
ace
4.5
.
3.
For
m
al
i
z
e Pa
r
t
ne
r S
e
r
v
. C
o
nver
sa
tio
n
4.5
.
4.
De
fi
ne
Pr
oce
ss Lo
gi
c.
4
.
4.
5
.
A
lig
n I
n
te
ra
c
t
io
n S
c
en
ar
io
s
&
R
e
f
ine
The hea
rt of SOA desig
n activity is
the
spe
c
ifica
t
ion activity,
employing m
odelin
g
notation. A review arti
cle
[22] reveals
six av
ailable
notation
s
for SOA modelin
g: (1) SOA-RM,
Referen
c
e M
odel for SOA
,
(2) SOA-RFA SOA Referen
c
e Archit
ecture Fou
n
d
a
tion, (3) SO
A
Ontology,
(4
) SOM
F
, S
e
rvice
-
O
r
ient
ed M
odelin
g
Fra
m
e
w
ork, (5) PIM4
SOA, Platform-
Indepe
nde
nt Model fo
r S
O
A, and (6)
SOAML, SOA Modelin
g
Lang
uage. F
o
r a
simpl
e
SOA
impleme
n
tation, a ba
si
c
UML o
r
a
Th
omas E
r
l’s
n
o
tation is
suff
icient, but fo
r a co
mplex
SO
A
impleme
n
tation, SOAML,
as a
n
exten
s
i
on of UM
L,
is sug
g
e
s
ted d
ue to its referential availabi
lity
and vast ad
o
p
tion in seve
ral SOA methodolo
g
ies.
4. The Propo
sed Frame
w
ork
The propo
se
d frame
w
ork i
s
drawn from
generalized
SE approa
ch,
describ
ed in
se
con
d
part of this
article,
with spe
c
ific u
s
e
of
busin
ess
analysi
s
tool
s, and
comb
ined with S
O
A
methodol
ogy, in o
r
de
r to
achieve
a
more
prac
ti
cal
ap
pro
a
ch
in servi
c
e engin
eeri
ng. The
prop
osed fra
m
ewo
r
k con
s
ist
s
of fou
r
stage
s: (1
) identificatio
n stage, (2) d
e
sig
n
sta
ge,
(3)
developm
ent stage a
nd (4
) deployment
stage.
The first sta
ge, the id
ent
ification
stag
e, is
the
bu
siness
side
a
nalysi
s
in
wh
ich n
e
w
(ele
ctro
nic) service
s
pote
n
t
ial are identif
ied and p
r
op
ose
d
with an
alysis of exist
i
ng co
ndition
of
the whol
e org
anization enti
t
y. The desig
n stage, a
s
th
e se
con
d
sta
ge, con
s
i
s
ts
of both bu
sin
e
ss
desi
gn an
d techni
cal d
e
si
g
n
pro
c
e
s
s. Th
e se
rvice
s
are desi
gne
d or rede
sign
ed i
n
this sta
ge with
rega
rd to the
existing condi
tion of the org
anizati
on. Implementin
g the frame
w
o
r
k f
o
r the first tim
e
in an
org
ani
zation i
s
also an
opp
ortu
nity to adopt
SOA app
ro
ach i
n
the
o
r
gani
zatio
n
o
f
IT
system. The
deci
s
io
n of employing SO
A in top-do
wn or bottom
-
u
p
mann
ers is
made d
u
rin
g
this
stage. The
thi
r
d stag
e,
the devel
opm
ent
stage, i
s
a
so
ftware
en
gine
ering
p
r
o
c
e
s
s co
nformi
ng t
o
SOA approa
ch. The la
st
stage, the
deployme
nt st
age,
con
s
ists of migrati
ng to pro
d
u
c
tion
environ
ment,
along with
applying mo
nitoring m
e
a
s
ures a
nd setting up SOA Governa
n
ce to
ensure the
co
nformity of syst
ems
for future improvements
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Servi
c
e Engi
neeri
ng Ba
se
d on Service Oriente
d
Architecture Met
hodol
og
y
(Su
hardi
)
1470
Figure 1. Pro
posed Servi
c
e Enginee
rin
g
Fram
ewo
r
k
4.1. Identific
a
tion Stag
e
In the identif
ication
stag
e, new
se
rvice
(
s)
pote
n
tial
for a
cu
stom
er i
s
identifie
d and
defined. The
unde
rlying m
o
tive
for
i
nno
vating
ne
w serv
ice
i
s
to
improve
custo
m
er val
ue. In
that
sen
s
e, the
service i
dentificati
on
stage
must co
nsi
s
t of steps to
identify services
with hi
gh
cu
stome
r
val
ue. Busi
ne
ss Model
Canv
as
(BMC) [2
3
]
is u
s
e
d
in t
h
is
stag
e a
s
an a
nalysi
s
a
nd
modelin
g too
l
of the fi
rm
to identify co
mpone
nts
wit
h
st
rong
pot
ential to im
p
r
ove the
val
ue
prop
ositio
n to the custom
ers. Nine blo
cks of BM
C are
the analytical
base
s
for im
proving the firm
value. The potential of improvem
ent might aris
e fro
m
custo
m
er
segment, reve
nue stream, co
st
stru
cture, value pro
p
o
s
itio
n, or in key a
c
tivities.
Figure 2. Business Mod
e
l Canva
s
Tem
p
late
The ide
n
tifica
tion stag
e sh
ould
start wit
h
"as-is BM
C", but the rea
l
result of the BMC
analysi
s
i
s
the ne
wly propo
sed BM
C: the "to-b
e
BMC". Thi
s
ne
w BM
C rep
r
e
s
ent
s
the
prop
ositio
n o
f
se
rvice
inn
o
vation. As
an exa
m
ple,
the de
ci
sion
to improve t
he Key A
c
tivities
block might require a crea
tion of
service innovation
in supp
orti
n
g
activities. Therefo
r
e, a ne
w
sup
port
se
rvice
sho
u
ld b
e
desi
gne
d an
d implem
ente
d
to incre
a
se
the effective
ness of the
key
activity pro
c
e
ss. T
h
is type
of se
rvice
can be
cate
g
o
rized a
s
"
s
e
r
vice
-a
s-sup
p
o
rt". In anoth
e
r
ca
se, the
de
cisi
on
of improvement mi
g
h
t re
side
in
t
he Valu
e Pro
positio
n blo
c
k. In thi
s
ca
se a
new
se
rvice
sho
u
ld b
e
propo
sed fo
r the custom
er
s,
in t
he f
o
r
m
of
dire
ct
cu
st
ome
r
s f
a
cin
g
servi
c
e
s
. This type of service can be
ca
t
egori
z
e
d
as
“servi
ce
-a
s-p
r
odu
ct”.
Another com
m
on to
ol to i
d
entify the
service
in
novatio
n is the
qu
est
i
onnai
re
meth
od. Th
e
method i
s
re
gularly u
s
ed
durin
g a ma
rket re
se
arch
activity. The use of q
u
e
s
tionnai
re i
s
pa
rt of
true requi
rem
ent investig
ation, as th
e requireme
nt
is
defined ba
sed
on actu
al intera
ction wi
th
pro
s
pe
ctive
cu
stome
r
s [5
]. An analysi
s
from
th
e q
uestio
nnai
re
data might i
n
stigate
a n
e
w
servi
c
e inn
o
vation. For example, the q
uestio
nnai
re result might sugge
st an im
provem
ent in the
firm key activ
i
ties. Therefo
r
e a service innovat
ion
ca
n be intro
d
u
c
ed from “se
r
vice-as-sup
port”
categ
o
ry to enhan
ce the a
c
tual key acti
vities.
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 : 146
6 – 1477
1471
Figure 3. Stages of Identifi
c
ation Stage
Another
com
m
on form of true requi
rem
ent invest
igati
on is ob
se
rva
t
ion. Obse
rva
t
ion is a
method
to
ca
pture
the i
n
fo
rmation
ba
se
d on
the
re
al
-wo
r
ld
situ
ation. Ob
se
rvati
on p
r
o
c
e
s
s
can
also
in
clude
que
stionn
aire
, or inte
rview activiti
es.
During
the
ob
servation th
e rese
arche
r
m
u
st
dire
ctly visits or expe
rien
ce
s
the
site
of actu
al b
u
s
ine
s
s p
r
o
c
e
s
s an
d
cu
sto
m
er t
r
an
sa
cti
on
perfo
rmed.
A servi
c
e i
n
novation in t
hese day
s freque
nt
ly lea
d
s to th
e im
plementatio
n
of ne
w
Information T
e
ch
nolo
g
y (IT) compo
nen
t [24]. An
input from IT asse
ssm
ent is t
herefo
r
e
will be
valuable i
n
th
e ide
n
tificatio
n
sta
ge. Th
e
latest
tre
nd i
n
IT, su
ch
as cloud
com
puti
ng, could
se
rve
as an imp
o
rta
n
t element in the servi
c
e in
novation.
The fin
a
l result of a
n
id
e
n
tification
sta
ge i
s
p
r
e
s
e
n
t
ed in
the fo
rm of
"to-b
e
BMC"
compl
e
me
nte
d
with sum
m
ary of the
servi
c
e inn
o
vation idea.
Both comp
onent
s sh
oul
d be
formali
z
ed a
s
a servi
c
e ide
n
tification do
cument.
4.2. Design
Stage
The de
sig
n
st
age of the p
r
opo
sed fram
ewo
r
k i
s
divid
ed into two
sub-stage
s: (1
) se
rvice
pro
c
e
s
s de
si
gn a
nd
(2
) S
O
A de
sign.
T
he g
oal
of th
e
first
su
b-stag
e, the
s
e
r
v
ic
e p
r
oc
ess
de
s
i
g
n
,
is to cre
a
te the de
sign of
the service defined
fro
m
analysi
s
in the identificat
ion stag
e. The
se
con
d
sub
-
stage, SOA
d
e
sig
n
, is the
sub
-
sta
g
e
to
elabo
rate
the
de
sign
u
s
ing
SOA a
pproa
ch
and metho
dol
ogy.
Duri
ng service pro
c
e
ss d
e
s
ign, the se
rvice bl
uep
rint
ing techni
que
[25] is prop
ose
d
to
rep
r
e
s
ent ho
w
the
servi
c
e sho
u
ld be del
ivered
to
the
cu
stome
r
. Se
rvice
blue
pri
n
ting techniq
u
e
is
norm
a
lly u
s
e
d
to d
e
scribe
existing
se
rvice
or to
rep
r
e
s
ent th
e
se
rvice i
nnovatio
n
.
The te
ch
niq
ue
define
s
five layers fo
r se
rvice intera
cti
on: (1
) phy
si
cal eviden
ce,
(2) cu
stom
e
r
action, (3
) on-
stage
co
ntact
employee,
(3) ba
ck-stag
e
conta
c
t em
pl
oyee an
d (5
)
sup
port p
r
o
c
ess. In additi
on
to se
rvice
blu
eprintin
g
te
ch
nique, Bu
sin
e
ss p
r
o
c
e
s
s diagram (BPD), with
its
B
u
sin
e
ss
P
r
o
c
ess
Modelin
g No
tation (BPMN) a
s
the to
ol, can b
e
u
s
ed to el
abo
rate the p
r
o
c
e
ss of
serv
ice
innovation.
Figure 3. Service Bluep
rint
Template
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Servi
c
e Engi
neeri
ng Ba
se
d on Service Oriente
d
Architecture Met
hodol
og
y
(Su
hardi
)
1472
The intention
of the service
design
sub
-
stage is to have an ov
ervie
w
of the servi
c
e. The
result from th
is process wi
ll be use
d
as a refe
re
nce
to determin
e
the se
rvice o
peratio
n in th
e
SOA implem
entation sub-stage.
The se
co
nd sub
-
sta
g
e
in desi
gn stag
e
is
t
he SOA
impleme
n
tation
sub
-
stage
. In this
sub
-
sta
ge, a
n
SOA meth
odolo
g
y will guide the
cre
a
ti
on of se
rvice de
sig
n
in
IT terms. An
SOA
methodol
ogy
commo
nly involves an i
dentificatio
n
activity followed by de
si
gn activity. The
purp
o
se of i
d
entification
is to ide
n
tify the candi
d
a
te
servi
c
e
s
, whil
e the
purpo
se of the
de
si
gn
stage i
s
to d
e
fine the serv
ice
s
sp
ecifi
c
a
t
ion,
such a
s
servi
c
e
contract and
ch
oreography. In the
prop
osed f
r
a
m
ewo
r
k, so
me pa
rt of
SOA se
rv
ice
identification
ha
s al
rea
d
y bee
n p
e
rfo
r
med
durin
g the previous
step, in the
form of to-be BMC,
service bl
uep
ri
nt and BPMN. These a
r
tifacts
are th
e sta
r
ti
ng poi
nt for
Service i
denti
f
ication
in
the
SOA metho
d
ology. For ex
ample, the
case
of adopting T
homa
s
Erl’s
MSOAM methodol
ogy, the fi
rst two step
s of it, Ontology Definition
an
d
Busine
ss Mo
del Alignm
en
t, can b
e
si
m
p
lified by
em
ploying the
result
s of BM
C an
alysi
s
. T
he
same id
ea is also a
ppli
c
abl
e to steps 3.1
(Bu
s
i
ness Re
qui
rement Defini
tion) and 3.
3.1
(De
c
o
m
po
se
Busine
ss Pro
c
e
s
s) in
whi
c
h the p
r
eviou
s
result from
Busin
e
ss P
r
oce
s
s Di
ag
ra
m
and Servi
c
e
Bluepri
n
ting can be u
s
ed.
Figure 4. SOA Methodolo
g
ies in
De
sig
n
Stage
4.3. Dev
e
lop
m
ent Stag
e
For a full-fle
d
ged SOA im
plementatio
n, several co
m
m
ercial platfo
rms a
r
e avail
able to
sup
port serv
ice-ori
ented
comp
uting a
ppro
a
ch
of SOA, such
as IBM We
bSphe
re, SAP
NetWeaver,
BEA Weblogi
c, and Oracl
e
SOA Suite.
An open
source alternati
v
e also exi
s
ts as
J
B
oss
Enterpris
e
SOA Plat
form.
For ou
r goal
with the pro
p
o
se
d frame
w
ork in a
simpl
e
deployme
nt, Micro
s
oft's
Wind
ows
Comm
uni
cati
on Fo
und
atio
n (WCF) te
ch
nology i
s
re
commen
ded
d
ue to it
s
simp
le develo
p
me
nt
requi
rem
ent
comp
ared to
other platform. Selecti
n
g
WCF
will h
e
lp defini
ng
activities in
stage
4.1.2 (Positio
n Core Stand
ard
s
) an
d 4.1
.
3 (C
ho
ose SOA Extensio
ns) of MSOA
M. Using
WCF
as th
e
co
re
stan
da
rd
a
nd SOA
technolo
g
y al
so
sim
p
lify sta
ge 4.2
(De
s
i
gn e
n
tity-ce
n
t
ric
busi
n
e
ss
services) and 4.
3 (De
s
ig
n ap
plicatio
n se
rvice) d
ue to th
e automation
of WDSL (Web
Service
Defin
i
tion Lang
uag
e) gen
eration
in WCF.
Duri
ng
and i
n
the e
nd
of develop
men
t
stage,
seve
ral te
sts fo
r t
he
system m
u
st b
e
con
d
u
c
ted. Due to the nature of SOA implem
e
n
tation as d
i
stribute
d
-m
e
s
sagin
g
syst
em,
p
e
r
for
m
in
g un
it te
s
t
an
d sys
te
m
te
st fo
r la
rge
scal
e
or
enterpri
s
e
-
wide
SOA
system
ca
n b
e
a
chall
engin
g
unde
rtake. Propo
sition fo
r testing
SOA
(we
b
)
servi
c
e
s
involves gene
rating t
e
st
requi
rem
ents by using Se
rvice Me
ssag
e Flow
Di
ag
ram (SMF
D)
cal
c
ulatin
g e
v
ery con
n
e
c
tions
variation, and
the use of test
tools for
web se
rvice, such a
s
SoapUI to WebInjec
t [26].
Before movin
g
to deployment stage, a User
Accepta
n
ce Te
st (UA
T
) must be p
e
rform
e
d
to ensu
r
e
con
f
ormity betwe
en the re
sult
of devel
opme
n
t and the re
sult from a
nal
ysis an
d de
si
gn
stage, al
so to
verify that th
e servi
c
e sy
st
em meet the use
r
re
quire
ment.
4.4. Deplo
y
ment Stag
e
A
su
ccessful testing re
sult can be deplo
y
ed
and
furth
e
r mo
nitored
and e
nha
nce
d
for the
system. Oth
e
r
wi
se, the
proce
s
s
mu
st
be re
peate
d
by analyzi
n
g
and d
e
termi
n
ing
step
s to
be
reiterated, in
analysi
s
, de
si
gn o
r
p
r
ototyping
stage,
a
nd repe
ating
the sta
ge to f
i
x the deviati
on.
Entering the
deployme
nt stage, any measu
r
e
s
for m
anag
ement chang
e of the organi
zatio
n
is
plann
ed and
executed to
ensu
r
e a smooth tran
sit
i
on from the
existing system. Monitori
ng
activities sho
u
ld
also
be condu
cted
to observe
t
he
real-wo
r
ld be
h
a
vior an
d pe
rforman
c
e of t
h
e
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 : 146
6 – 1477
1473
newly i
n
stall
e
d service
system and
its in
frastruc
tu
re
a
s
a
me
an
s fo
r
continu
o
u
s
l
y
improvin
g t
h
e
servi
c
e. Fo
r
enlarging th
e
impl
ementati
on of SOA a
ppro
a
ch, SO
A governa
nce sho
u
ld al
so
be
desi
gne
d and
put into place.
The followi
ng
table summ
arizes the
co
nce
p
t and to
ols empl
oyed
along with
artifact
s
prod
uced du
ri
ng ea
ch sta
g
e
of the framewo
r
k.
Table 3. Prop
ose
d
SE Fra
m
ewo
r
ks To
o
l
s and Artifa
cts
Stage
Tool
s
A
r
t
i
fac
t
1. Identif
y
True
requireme
nt
BMC
Questionnaire re
sult
Observation ana
y
s
is
As-Is BMC
Service Innovation
To-be BM
C
Business Ser
v
i
c
e Catalog
2. Design
Service Blueprinting
BPMN
SO
A Method
(e.q.
MSOAM/SOMA
)
Service Blueprint
BPD
UML / SO
AML
IT Service Catalo
g
3. Develop
SO
A Platform
(e.q. WCF
)
IT Service
Implementation
Unit and s
y
stem
test
r
e
sults
UAT result
4. Deplo
y
SOA Gove
rnanc
e
SOA Policy
&Sta
ndards
Monitoring meas
ures
5. Case Stud
ies
As an
atte
mpt to form
alize
the
propo
sed
Serv
ice En
gine
ering F
r
ame
w
ork,
the
frame
w
ork
was in
cre
m
ent
ally built by usi
ng
com
p
onent
s of the prop
osed
frame
w
ork a
n
d
unde
rtaki
ng the followi
ng case
studie
s
. The detail
s
are pre
s
ente
d
b
e
low.
5.1. First Ca
se Stud
y
The first case study wa
s
perfo
rmed b
e
f
ore a
definiti
v
e formal fra
m
ewo
r
k wa
s
devise
d
.
The locus of the study was a State Palm Plantat
ion Firm. The goal
wa
s to provid
e an integrate
d
Enterpri
se
Reso
urce Plan
ning (E
RP) b
a
se
d on
SOA
approa
ch. As a service e
ngine
erin
g work,
the study em
ployed SOM
A
as its SOA
methodol
ogy.
From the th
ree ide
n
tificati
on techniq
u
e
s
in
SOMA: doma
i
n decompo
si
tion, goal se
rvice modeli
n
g and existin
g
system a
n
a
l
yst, only the
first
two are pe
rforme
d, due t
o
the minim
a
l IT resou
r
ces availa
ble
in the firm. In modeli
ng the
servi
c
e
s
the
work enh
an
ces th
e u
s
e
of
UML
with
SO
AML in th
e di
agra
mmatical
form
s of
Service
Specification,
Service Inte
rface
an
d S
e
rvice
R
eali
z
ation. Finally
, a prototype
of the
se
rvice
desi
gn wa
s p
r
odu
ce
d by using
WCF te
chn
o
logy. Thi
s
ca
se
study serve
d
as a
basi
s
to asse
ss
the context of a SOA Methodolo
g
y with
in a Service E
ngine
erin
g proce
s
s.
5.2. Second Cas
e
Stud
y
The se
con
d
ca
se study was a se
rvice
engin
eeri
ng
study on th
e p
r
ocess of la
n
d
u
s
ag
e
permit in the
National La
nd Authority [27]. T
he stu
d
y mainly employed SO
MA as the b
a
si
c
methodol
ogy, but only p
e
rf
orme
d Domai
n
De
co
mpo
s
it
ion an
d G
oal
Service
Mod
e
ling du
ring
th
e
identificatio
n stage. Servi
c
e bl
uep
rintin
g wa
s u
s
ed
to descri
be the bu
sine
ss pro
c
e
ss
of the
servi
c
e. Seve
ral mod
e
ling
notation
s
we
re use
d
in the study: UML’
s
Sequen
ce
Di
agra
m
, Service
Comp
one
nt Specification,
Subsy
s
tem Dep
end
enc
y
Diag
ram,
and
an in
stantiati
on of th
e SO
MA
referen
c
e a
r
chitecture. WS
DL
was
also
use
d
for d
e
scribi
ng a
nd f
o
rmali
z
in
g web services,
and
finally a p
r
oto
t
ype wa
s
dev
elope
d by
usi
ng
WCF
Framework
and
PHP SOAP.
In this case
s
t
udy,
several co
mp
onent
s of the frame
w
ork, i.e. t
he modeli
ng tools, are introdu
ce
d an
d tested.
5.3. Third Ca
se Stud
y
The
third ca
se
study wa
s done after
th
e
first
draft of
the fram
ewo
r
k
wa
s fo
rmal
ized. T
he
ca
se
study e
m
ployed the f
u
ll pro
p
o
s
ed f
r
ame
w
o
r
k i
n
applying SOA
approa
ch for managi
ng da
ta
silo
s in
the
G
o
vernm
ent St
atistical
Agen
cy. The
ca
se study wa
s sta
r
ted with strat
egic
an
alysi
s
of
the agen
cy with BMC and
followe
d by a serie
s
of BPMN to descri
be and to an
alyze bu
sin
e
ss
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Servi
c
e Engi
neeri
ng Ba
se
d on Service Oriente
d
Architecture Met
hodol
og
y
(Su
hardi
)
1474
pro
c
e
s
ses. Due to the focus on the internal a
gen
cy
process, se
rvice bl
uep
rint
ing tool wa
s only
use
d
lightly
to spe
c
ify th
e data di
sse
m
inati
on
se
rvice ba
se
d
on reque
st f
r
om
cu
stome
r
s.
MSOAM was sel
e
cte
d
a
s
the SOA
m
e
thod
and
was
employe
d
du
ring
the
desi
gn
stag
e
to
prod
uce thre
e layers of service: task-centric,
entity-centri
c, and
appli
c
at
ion ut
ility.
Services and
servi
c
e co
mp
osition
is de
scribe
d
by u
s
ing
sta
nda
rd notation.
F
i
nally, the se
rvice i
n
terfa
c
e is
descri
bed by
usin
g WSDL.
5.4. Fourth Case Study
The fo
urth
case
stu
d
y al
so
sel
e
ct
s th
e Gove
rnm
e
nt Statistical
Agency
a
s
t
he
study
locu
s, but focusin
g
on the desi
gn of pro
posed ex
tern
al data se
rvice for the age
ncy. The wo
rk
wa
s al
so
don
e by follo
win
g
the
com
p
le
ted an
d fo
rm
alize
d
fra
m
e
w
ork. T
he
work p
r
od
uced
two
BMC dia
g
ra
ms: a
s
-i
s BM
C to d
e
scrib
e existing
sit
uation a
nd to
-be BM
C to
prop
ose the
new
servi
c
e. T
h
e
SWO
T
a
nal
ysis
wa
s u
s
ed b
e
twe
en
the two
BM
Cs to
und
erstand
the
ex
isting
busi
n
e
ss mo
del and to va
lidate the ne
ed for the
propo
sed
servi
c
e in the ag
e
n
cy. Service
blue
printing
an
d
BPMN di
agra
m
were
the
n
use
d
to
de
si
g
n
bu
sin
e
ss
specifi
c
ation
o
f
the
servi
c
e.
For
SOA de
sign,
MSOAM was empl
oyed
wi
th Thom
as
E
r
l’s
sta
nda
rd notation co
m
p
lemente
d
wi
th
basi
c
UM
L di
agra
m
s: Use
Ca
se, Cla
s
s Diag
ram a
nd
Sequen
ce
Di
agra
m
.
5.5. Usabilit
y
Test
De
spite seve
ral re
quired field adju
s
tme
n
ts
and im
ple
m
entation variations, the a
doption
attempt on these case
studies
proves t
he usability Service Engi
neering F
r
am
ewor
k. T
o
further
test and eval
uate the co
n
c
ept
s, the propo
sed fr
a
m
ewo
r
k
wa
s taught in an IT maste
r
de
gree
cla
s
s
f
o
r
a
se
mest
e
r
1
.
Each
stud
ent
the
n
assig
ned
to perfo
rm a ca
se study on
an
o
r
ga
nization
by desi
gnin
g
and d
e
velo
ping a
proto
t
ype of
tech
nology-ena
bl
ed serv
i
c
es by
adoptin
g th
e
methodol
ogy from the fram
ewo
r
k.
Table 4. Que
s
tionn
aire Structure
Hy
pot
h
e
si
s Q
u
e
s
t
i
on
s
1. Frame
w
ork is eas
y
to compre
h
end
Q1,
Q5,
Q6
2. Frame
w
ork is eas
y
to ado
pt an
d implement
Q3,
Q8,
Q9,
Q10
,
Q11
3. Frame
w
ork he
lps communicate
project
Q2,
Q4
4. Frame
w
ork is straightfor
w
a
r
d
Q7
At the end
o
f
the se
me
ster, a
su
rvey
wa
s
c
ond
uct
ed to 3
0
stu
dents of the
cla
s
s to
asse
ss the v
a
lue
s
a
nd
be
nefits of
the f
r
ame
w
o
r
k.
T
he fo
cu
s of
the q
u
e
s
tionn
aire
is to te
st the
simpli
city of t
he fra
m
e
w
ork, op
eratio
na
lized
un
der
four hypothe
ses de
sc
ribe
d
in tabl
e 4.
The
respon
ds a
r
e
given unde
r 5-scale
li
kert
resp
ond, from “st
r
on
gly-agr
e
e
” to “strongly disagre
e
”.
The que
stion
s
and thei
r re
sults a
r
e p
r
e
s
ented in the followin
g
ch
art
s
.
1)
M
a
ster
of I
n
for
m
atics,
I
T
Concen
tr
ation,
School of
E
l
ectr
i
cal
E
ngineer
ing and I
n
form
atics,
Bandung I
n
stit
ute of T
echnolo
g
y
,
2014.
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 : 146
6 – 1477
1475
Figure 5. Survey Result
s o
n
Comp
re
hen
sion to the Framework
The re
sult
s from figure
5 show that mo
re
than 80% o
f
the resp
ond
ents ag
re
e that they
unde
rsta
nd a
nd able to
ad
opt the frame
w
ork. On
th
e other h
and,
while n
o
re
sp
onde
nts p
r
ovi
d
e
negative fee
dba
ck
reg
a
rding hi
s un
d
e
rsta
ndin
g
,
o
n
ly about 40
% of resp
on
dents fe
el truly
unde
rsta
nd t
he
role
of e
a
ch
sta
ge
a
nd tool
s
em
ployed i
n
th
e fram
ework.
The
later re
sults
provide
a m
o
tivation to ela
borate
the
st
age
s de
scri
ption furth
e
r
an
d en
rich the
rational fo
r ea
ch
tool it employed.
Figure 6. Survey Result
s o
n
Stages Cl
arity and Practi
cality
Delving
furth
e
r i
n
to e
a
ch
su
ccessive
st
age
i
n
figu
re
6,
we
ca
n
see the
imp
r
o
v
ement
potential
s
of
stag
es
and
tools
elab
oration. All of
re
spo
nde
nt
is a
g
ree o
n
the
cla
r
ity and
pra
c
ticality of the identifica
t
ion stage. Also, 80%
of the re
spo
nde
nt has simil
a
r opinion for the
desi
gn stag
e. The rate of agre
e
me
nt among re
sp
on
dents d
r
op
s to 60% for the third (Devel
op)
and
fou
r
th (Deploy) stag
e. The re
sults
f
o
r
the
thir
d
a
nd fou
r
th
sta
ge a
r
e
rathe
r
expect
ed d
u
e
to
the com
p
lexity of development req
u
ire
d
in the
third
stage, an
d the limited form
alizatio
n defi
ned
in the frame
w
ork fo
r the fo
urth sta
ge. Neverthele
s
s, these re
sult
s also
sup
port t
he motivation
on
further works for
stag
e
descriptio
n
,
es
pec
ially for the las
t
two s
t
ages
.
In terms
of b
enefits of the
prop
osed fra
m
ew
o
r
k, figure 7 sh
ows th
at more th
an
90% of
the respon
de
nts agree tha
t
the
framework
can a
c
tua
lly help an organi
zation to introdu
ce a n
e
w
10
%
71%
19%
Q1:
I
u
nde
rs
t
a
n
d
th
e
con
t
e
x
t
u
a
l
fu
n
c
t
i
o
n
of
SE
F
Ag
ree+
Ag
r
e
e
Ne
u
t
r
a
l
Di
s
a
g
r
e
e
D
i
s
a
g
r
ee+
10%
74%
16%
Q3
:
I
ca
n
u
t
il
iz
e
SE
F
fo
r
my
ow
n
con
t
e
x
t
A
g
ree+
Ag
re
e
Ne
u
t
r
a
l
Di
s
a
g
r
e
e
Di
s
a
g
r
e
e
+
3%
42
%
55
%
Q5
.
I
un
de
r
s
ta
nd
th
e
fu
n
c
t
i
o
n
of
eac
h
st
a
g
e
in
SE
F
A
g
ree+
Ag
re
e
Ne
u
t
r
a
l
Di
sa
g
r
e
e
Di
sa
g
r
e
e
+
42%
58%
Q6.
I
un
de
r
s
ta
nd
th
e
fu
n
c
t
i
o
n
of
all
m
o
de
l
i
ng
too
l
s
in
SE
F
A
g
ree+
Ag
re
e
Ne
u
t
r
a
l
Di
s
a
g
r
e
e
D
i
s
a
g
r
ee+
0%
10
%
20
%
30
%
40
%
50
%
60
%
70
%
80
%
90
%
10
0%
1.
Ide
n
t
i
f
y
2
.
De
s
i
g
n
3.
De
v
e
l
o
p
4
.
De
p
l
o
y
Q8
‐
Q
11:
St
a
g
e
is
Cl
e
a
r
&
Pr
a
c
t
i
c
a
l
D
i
s
a
g
r
ee+
Di
s
a
g
r
e
e
Ne
u
t
r
a
l
Ag
r
e
e
Ag
r
e
e+
Evaluation Warning : The document was created with Spire.PDF for Python.