T
E
L
KO
M
N
I
KA
T
e
lec
om
m
u
n
icat
ion
,
Com
p
u
t
i
n
g,
E
lec
t
r
on
ics
an
d
Cont
r
ol
Vol.
18
,
No.
1
,
F
e
br
ua
r
y
2020
,
pp.
114
~
123
I
S
S
N:
1693
-
6930,
a
c
c
r
e
dit
e
d
F
ir
s
t
G
r
a
de
by
Ke
me
nr
is
tekdikti
,
De
c
r
e
e
No:
21/E
/KP
T
/2018
DO
I
:
10.
12928/
T
E
L
KO
M
NI
KA
.
v18i1.
12733
114
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
T
ime
-
b
ase
d
α
+
m
in
e
r
f
or
m
o
d
e
ll
in
g b
u
si
n
e
ss
p
r
o
c
e
sse
s
u
si
n
g t
e
m
p
o
r
al
p
a
t
t
e
r
n
Yut
ik
a
Am
e
li
a
E
f
f
e
n
d
i
1
,
Riya
n
ar
t
o
S
ar
n
o
2
1
In
fo
rma
t
i
o
n
Sy
s
t
em
S
t
u
d
y
Pro
g
ram,
D
ep
ar
t
men
t
o
f
E
n
g
i
n
eeri
n
g
,
U
n
i
v
ers
i
t
a
s
A
i
r
l
an
g
g
a,
In
d
o
n
es
i
a
2
D
ep
ar
t
men
t
o
f
In
f
o
rmat
i
cs
,
In
s
t
i
t
u
t
T
e
k
n
o
l
o
g
i
Sep
u
l
u
h
N
o
p
emb
er,
In
d
o
n
es
i
a
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
M
a
r
19
,
2019
R
e
vis
e
d
J
un
30
,
20
19
Ac
c
e
pted
J
ul
18
,
20
19
Bu
s
i
n
es
s
p
r
o
ces
s
es
are
i
m
p
l
eme
n
t
e
d
i
n
an
o
rg
a
n
i
za
t
i
o
n
.
W
h
e
n
a
b
u
s
i
n
es
s
p
ro
ce
s
s
i
s
ru
n
,
i
t
g
en
era
t
es
ev
e
n
t
l
o
g
.
O
n
e
t
y
p
e
o
f
ev
en
t
l
o
g
i
s
d
o
u
b
l
e
t
i
me
s
t
am
p
ev
e
n
t
l
o
g
.
D
o
u
b
l
e
t
i
me
s
t
am
p
h
as
t
h
e
s
t
ar
t
a
n
d
c
o
mp
l
et
e
t
i
me
o
f
each
act
i
v
i
t
y
ex
ecu
t
ed
i
n
t
h
e
b
u
s
i
n
es
s
p
ro
ce
s
s
an
d
h
as
a
cl
o
s
e
rel
at
i
o
n
s
h
i
p
w
i
t
h
t
emp
o
ral
p
a
t
t
er
n
.
In
t
h
i
s
p
ap
er,
s
ev
e
n
t
y
p
es
o
f
t
em
p
o
ra
l
p
at
t
er
n
b
et
w
een
act
i
v
i
t
i
e
s
w
ere
p
re
s
en
t
ed
as
e
x
t
e
n
d
e
d
v
ers
i
o
n
o
f
rel
at
i
o
n
s
u
s
ed
i
n
t
h
e
d
o
u
b
l
e
t
i
me
s
t
am
p
ev
en
t
l
o
g
.
Si
n
ce
t
h
e
ev
en
t
l
o
g
w
as
n
o
t
al
w
ay
s
ex
ec
u
t
e
d
in
s
eq
u
en
t
i
a
l
w
ay
,
t
h
er
efo
re
u
s
i
n
g
t
em
p
o
ra
l
p
at
t
ern
,
ev
e
n
t
l
o
g
w
as
d
i
v
i
d
e
d
i
n
t
o
s
ev
era
l
s
mal
l
g
r
o
u
p
s
t
o
mi
n
e
t
h
e
b
u
s
i
n
e
s
s
p
ro
ce
s
s
b
o
t
h
s
eq
u
en
t
i
a
l
an
d
p
aral
l
el
.
Bo
t
h
t
em
p
o
ra
l
p
a
t
t
er
n
an
d
T
i
me
-
b
a
s
ed
α
+
Mi
n
er
al
g
o
r
i
t
h
m
w
ere
u
s
e
d
t
o
mi
n
e
p
ro
ce
s
s
m
o
d
e
l
,
d
e
t
erm
i
n
ed
s
e
q
u
e
n
t
i
al
a
n
d
p
aral
l
el
re
l
at
i
o
n
s
an
d
t
h
e
n
ev
al
u
at
ed
t
h
e
p
r
o
ces
s
mo
d
el
u
s
i
n
g
fi
t
n
e
s
s
v
al
u
e.
T
h
i
s
p
a
p
er
w
a
s
fo
c
u
s
e
d
o
n
t
h
e
ad
v
an
t
a
g
es
o
f
t
emp
o
ral
p
a
t
t
e
rn
i
mp
l
eme
n
t
e
d
i
n
T
i
me
-
b
as
e
d
α
+
Mi
n
er
al
g
o
ri
t
h
m
t
o
mi
n
e
b
u
s
i
n
es
s
p
ro
ce
s
s
.
T
h
e
res
u
l
t
s
al
s
o
cl
earl
y
s
t
at
ed
t
h
a
t
t
h
e
p
r
o
p
o
s
e
d
met
h
o
d
c
o
u
l
d
p
re
s
en
t
b
e
t
t
er
re
s
u
l
t
rat
h
er
t
h
an
t
h
a
t
o
f
o
r
i
g
i
n
a
l
α
+
Mi
n
er
al
g
o
r
i
t
h
m.
K
e
y
w
o
r
d
s
:
Alpha
+
mi
ne
r
B
us
ines
s
pr
oc
e
s
s
model
P
r
oc
e
s
s
dis
c
ove
r
y
P
r
oc
e
s
s
mi
ning
T
e
mpor
a
l
pa
tt
e
r
n
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e
.
C
or
r
e
s
pon
din
g
A
u
th
or
:
I
nf
or
mation
S
ys
tem
S
tudy
P
r
og
r
a
m,
De
pa
r
tm
e
nt
of
E
nginee
r
ing,
Unive
r
s
it
a
s
Air
langga
,
68
Dha
r
maw
a
ngs
a
Da
lam
S
e
lata
n
S
t.
,
Ai
r
langga
,
Gube
ng,
S
ur
a
ba
ya
,
J
a
wa
T
im
u
r
60286
,
(
031
)
5033869
,
I
ndo
ne
s
ia
.
E
mail:
yuti
ka
.
e
f
f
e
ndi@vokas
i.
una
ir
.
a
c
.
id
1.
I
NT
RODU
C
T
I
ON
An
or
ga
niza
ti
on
ha
s
inf
o
r
mation
s
ys
tems
whic
h
a
r
e
us
e
d
to
c
oll
e
c
t,
s
tor
e
a
nd
p
r
oc
e
s
s
da
ta
to
ga
in
inf
or
mation
on
c
e
r
tain
thi
ngs
[
1,
2]
.
T
he
y
c
a
n
a
ls
o
s
uppor
t
mana
ge
ment
a
nd
de
c
is
ion
making
of
bus
ines
s
pr
oc
e
s
s
e
s
[
3]
.
S
ys
tems
in
the
or
ga
niza
ti
on
wi
ll
ge
n
e
r
a
t
e
e
ve
nt
log
whic
h
g
ives
knowle
dge
a
bout
the
a
c
ti
vit
ies
of
bus
ines
s
pr
oc
e
s
s
e
s
r
un
r
e
c
e
ntl
y
or
f
e
w
ye
a
r
s
a
go
[
4,
5]
.
B
e
c
a
us
e
of
the
huge
a
mount
of
e
ve
nt
l
og,
they
c
a
nnot
a
na
lyze
dir
e
c
tl
y.
T
he
r
e
f
or
e
pr
oc
e
s
s
mi
nin
g
tec
hnique,
whic
h
is
a
c
ombi
na
ti
on
of
da
ta
m
i
n
ing
a
nd
mac
hine
lea
r
ning,
is
pr
e
s
e
nt
to
be
a
ble
to
ha
ndl
e
pr
oblems
in
the
e
ve
nt
log
[
4
,
6,
7]
.
I
n
pr
oc
e
s
s
mi
ning
,
the
tec
hnique
r
e
late
d
to
mi
ne
the
e
ve
nt
log
to
ge
t
pr
oc
e
s
s
models
de
s
c
r
ibi
ng
the
r
e
a
l
bus
ines
s
pr
oc
e
s
s
e
s
is
c
a
ll
e
d
pr
oc
e
s
s
dis
c
ov
e
r
y
[
6]
.
P
r
o
c
e
s
s
models
a
c
t
a
s
guidanc
e
f
or
a
na
lyzing
a
nd
ve
r
if
ying
the
pe
r
f
or
manc
e
of
the
e
xis
ti
ng
bus
ines
s
pr
oc
e
s
s
e
s
[
8]
.
I
n
pr
oc
e
s
s
dis
c
ove
r
y,
e
ve
nt
log
is
one
of
the
main
i
nput
ne
e
de
d
[
9]
.
Da
ta
whic
h
mus
t
be
c
ontaine
d
in
the
e
ve
nt
log
a
r
e
a
c
ti
vit
ies
a
nd
ti
mes
tamp
[
10]
.
Ac
ti
vit
ies
de
f
ine
the
na
me
of
e
ve
nts
or
jobs
w
hich
a
r
e
e
xe
c
uted
in
bus
ines
s
pr
oc
e
s
s
e
s
[
4]
.
T
o
ge
t
the
r
e
a
l
ti
me
of
bus
ines
s
pr
oc
e
s
s
e
s
,
we
c
a
n
us
e
ti
mes
tamp
[
11
]
.
T
his
r
e
s
e
a
r
c
h
us
e
s
double
ti
mes
tamp
e
ve
nt
log
b
e
c
a
us
e
it
ha
s
s
tar
t
a
nd
c
ompl
e
te
ti
me.
D
ouble
ti
mes
tamp
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
ime
-
bas
e
d
α
+
mine
r
for
mode
ll
ing
bus
ines
s
pr
oc
e
s
s
e
s
us
ing
tem
por
al
patt
e
r
n
(
Y
uti
k
a
A
me
li
a
E
f
fendi
)
115
e
ve
nt
log
ha
s
a
c
los
e
r
e
lations
hip
with
the
tempo
r
a
l
pa
tt
e
r
n
,
whic
h
is
a
pa
tt
e
r
n
of
e
ve
nt
log
that
oc
c
ur
s
f
r
om
e
a
c
h
a
c
ti
vit
y
pe
r
f
or
med
[
12
,
13]
.
I
n
r
e
a
li
ty,
to
ob
tain
the
bus
ines
s
pr
oc
e
s
s
e
s
,
e
ve
nt
log
is
not
a
lwa
ys
e
xe
c
uted
in
s
e
que
nti
a
l
wa
y
f
r
om
s
tar
t
to
e
nd
[
13]
.
I
t
c
a
n
be
done
by
e
xe
c
uti
ng
the
a
c
ti
vit
ies
in
s
e
que
nti
a
l
a
nd
in
pa
r
a
ll
e
l.
T
o
know
the
e
xa
c
t
a
c
ti
vit
ies
e
xe
c
uted
in
s
e
que
nti
a
l
or
in
pa
r
a
ll
e
l,
one
e
a
s
y
wa
y
to
f
ind
out
is
to
pa
y
a
tt
e
nti
on
to
the
ti
mes
tamp
of
e
a
c
h
a
c
ti
vit
y
[
12]
.
T
he
r
e
lation
is
de
f
ined
a
s
pa
r
a
ll
e
l
if
the
ti
mes
tamp
ove
r
laps
or
in
ter
ms
of
li
nking
one
or
mor
e
a
c
ti
vit
ies
wit
h
one
or
mor
e
a
c
ti
vit
ies
e
xe
c
uted
a
f
ter
wa
r
ds
,
m
e
a
nwhile
s
e
que
nti
a
l
r
e
lation
ha
s
the
ti
mes
tamp
whic
h
doe
s
not
ove
r
lap
a
nd
a
bout
li
nking
one
a
c
ti
vit
y
a
nd
a
nother
a
c
ti
vit
y
e
xe
c
uted
a
f
ter
wa
r
ds
[
13]
.
B
e
c
a
us
e
of
that
matter
s
,
us
ing
tempo
r
a
l
pa
tt
e
r
n
,
e
ve
nt
log
is
divi
de
d
int
o
s
e
ve
r
a
l
s
mall
gr
oups
to
make
m
ini
ng
pr
oc
e
s
s
mode
ls
a
nd
their
r
e
lations
e
a
s
y.
I
n
thi
s
r
e
s
e
a
r
c
h,
s
e
ve
n
types
of
tempor
a
l
pa
tt
e
r
n
be
twe
e
n
a
c
ti
vit
ies
a
r
e
pr
e
s
e
nted
a
s
e
xtende
d
ve
r
s
ion
of
r
e
lations
us
e
d
in
the
double
ti
mes
tamp
e
ve
nt
l
og.
T
his
pa
pe
r
p
r
opos
e
s
T
im
e
-
ba
s
e
d
α
+
M
iner
(
T
B
A+
M
)
a
lgor
it
hm
to
mi
ne
a
pr
oc
e
s
s
model
a
long
with
their
r
e
lations
ba
s
e
d
on
the
e
ve
nt
log.
M
odif
ied
T
im
e
-
ba
s
e
d
Alph
a
M
iner
(
M
T
B
AM
)
[
13]
wa
s
int
r
oduc
e
d
to
upgr
a
de
the
α
a
lgor
it
hm
ba
s
e
d
on
t
im
e
int
e
r
va
l
.
How
e
ve
r
,
thi
s
a
lgor
it
hm
c
a
nnot
mi
ne
length
one
loop
(
L
1L
)
a
nd
length
two
loop
(
L
2L
)
in
the
e
ve
nt
log
to
be
modelled
in
a
bus
ines
s
pr
oc
e
s
s
.
W
he
n
c
ompar
e
d
to
or
igi
na
l
α
+
a
lgo
r
it
hm
[
14
,
15
]
,
the
pr
opos
e
d
a
lgo
r
it
hm
c
a
n
de
tec
t
ove
r
lapping
pr
oc
e
s
s
e
s
us
ing
tempor
a
l
pa
tt
e
r
n
a
n
d
identif
ies
them
a
s
pa
r
a
ll
e
l
pr
oc
e
s
s
e
s
by
de
f
ini
ng
ne
w
thr
e
s
holds
a
nd
a
ls
o
c
a
n
dis
c
ove
r
length
one
loop
(
L
1L
)
a
nd
length
two
loop
(
L
2L
)
.
α
+
a
lgor
it
hm
i
s
c
h
os
e
n
be
c
a
us
e
it
is
the
upgr
a
de
d
ve
r
s
ion
o
f
α
a
lgor
it
hm
[
16
]
.
How
e
ve
r
,
the
dis
a
dva
ntage
s
of
α
+
a
lgor
it
h
m
a
r
e
it
c
a
nnot
dis
ti
nguis
h
the
dif
f
e
r
e
nc
e
s
be
twe
e
n
AN
D
a
nd
OR
in
pa
r
a
ll
e
l
r
e
lations
a
nd
s
ti
ll
us
e
s
e
que
nti
a
l
r
e
lations
be
twe
e
n
a
c
ti
vit
ies
to
obtain
a
bu
s
ines
s
pr
oc
e
s
s
.
I
n
other
wor
ds
,
α
+
a
lgor
it
hm
c
a
n
only
de
ter
mi
ne
whe
ther
the
r
e
lations
of
bus
ines
s
pr
oc
e
s
s
e
s
a
r
e
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
(
AN
D,
XO
R
)
in
s
e
que
nti
a
l
wa
y.
T
o
c
ove
r
the
s
hor
tcomings
of
the
α
+
a
lgor
it
hm
,
the
mod
if
ica
ti
on
of
α
+
a
lgor
i
thm
int
o
T
im
e
-
ba
s
e
d
α
+
a
lgor
it
hm
(
T
B
A+
M
)
is
p
r
opos
e
d
in
or
de
r
to
be
a
ble
to
dis
c
ove
r
pr
oc
e
s
s
models
,
including
dis
c
ove
r
ing
L
1L
a
n
d
L
2L
,
a
s
we
ll
a
s
their
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
(
AN
D,
OR
,
XO
R
)
by
de
ter
mi
ning
ne
w
thr
e
s
holds
s
o
that
the
bus
ines
s
pr
oc
e
s
s
c
a
nnot
only
be
dis
c
ove
r
e
d
by
us
ing
s
e
que
nti
a
l
wa
y.
T
o
s
um
up,
thi
s
r
e
s
e
a
r
c
h
f
o
c
us
e
s
on
us
ing
tempor
a
l
pa
tt
e
r
n
,
modi
f
ying
α
+
a
lgor
it
hm
in
to
T
B
A+
M
a
lgor
it
hm
a
nd
us
ing
f
it
ne
s
s
va
lue
a
s
e
va
luation
of
dis
c
ove
r
e
d
pr
oc
e
s
s
models
.
T
his
r
e
s
e
a
r
c
h
pa
pe
r
c
ons
is
ts
of
f
our
s
e
c
ti
ons
.
Our
pr
oc
e
s
s
dis
c
ove
r
y
a
ppr
oa
c
h
r
e
late
d
to
th
is
r
e
s
e
a
r
c
h
will
be
e
xplaine
d
in
s
e
c
ti
on
2.
T
he
e
xpe
r
im
e
ntal
r
e
s
ult
a
nd
d
is
c
us
s
ion
will
be
pr
e
s
e
nted
in
s
e
c
ti
on
3.
L
a
s
t,
the
c
onc
lus
ion
will
be
c
onc
luded
t
his
r
e
s
e
a
r
c
h
pa
pe
r
in
s
e
c
ti
on
4
.
2.
RE
S
E
AR
CH
M
E
T
HO
D
I
n
s
e
c
ti
on
2,
we
pr
e
s
e
nt
the
e
ve
nt
log,
tempor
a
l
pa
tt
e
r
n
a
nd
int
e
gr
a
ted
dis
c
ove
r
y
a
pp
r
oa
c
h
f
o
r
dis
c
ove
r
ing
bus
ines
s
pr
oc
e
s
s
e
s
in
thi
s
r
e
s
e
a
r
c
h.
2.
1.
E
ve
n
t
l
og
W
e
e
va
luate
a
n
e
ve
nt
log
o
f
P
r
oduc
ti
on
f
r
om
P
T
.
YM
;
a
ya
r
n
p
r
oduc
ti
on
c
ompany
in
J
a
ka
r
ta,
I
ndone
s
ia.
W
e
c
a
ll
it
a
s
E
L
1.
T
a
ble
1
s
hows
pa
r
t
o
f
E
L
1
whic
h
c
ontains
C
a
s
e
I
D,
li
s
t
of
a
c
ti
vit
ies
,
s
tar
t
ti
me,
a
nd
c
ompl
e
te
ti
me
.
E
L
1
ha
s
100
c
a
s
e
s
a
nd
11
a
c
ti
vit
ies
.
T
he
a
c
ti
vit
ies
a
r
e
C
ott
on
B
ales
,
B
low
ing
and
P
ick
ing,
C
ar
ding,
P
r
e
C
ar
ding,
Supe
r
L
ap,
C
omb
ing,
Dr
aw
ing,
Dr
aw
ing
I
I
,
R
ov
ing,
Spinn
ing,
W
in
ding
.
I
n
the
e
nd,
P
e
t
r
i
Ne
t
f
or
m
[
15
,
17
]
is
us
e
d
to
pr
e
s
e
nt
t
he
f
inal
r
e
s
ult
of
bus
ines
s
pr
oc
e
s
s
in
our
e
xpe
r
im
e
nt.
2.
2.
T
e
m
p
or
al
p
at
t
e
r
n
T
he
model
o
f
the
tempor
a
l
pa
tt
e
r
n
is
ba
s
e
d
on
t
he
c
a
tegor
y
of
the
ti
mes
tamp,
e
s
pe
c
ially
double
ti
mes
tamp
to
mi
ne
the
pr
oc
e
s
s
model.
R
izka
e
t
a
l.
int
r
oduc
e
d
the
us
e
of
ti
me
-
ba
s
e
d
in
pr
oc
e
s
s
mi
ning
a
lgor
it
hm
[
18]
.
I
n
he
r
r
e
s
e
a
r
c
h,
the
de
f
ini
t
ion
of
t
im
e
-
ba
s
e
d
in
pr
oc
e
s
s
mi
ning
a
lgor
it
hm
c
ons
is
ts
o
f
be
for
e
and
me
e
ts
,
ov
e
r
laps
,
c
ontai
ns
,
is
-
fi
nis
he
d
-
by
,
e
qu
als
,
a
nd
s
tar
ts
.
I
n
th
is
r
e
s
e
a
r
c
h,
we
pr
e
s
e
nt
s
e
ve
n
types
of
tempor
a
l
pa
tt
e
r
n
be
twe
e
n
a
c
ti
vit
ies
whic
h
a
r
e
e
xtende
d
ve
r
s
ion
of
s
tanda
r
d
r
e
lations
in
bus
ines
s
pr
oc
e
s
s
e
s
.
I
f
pr
e
vious
ly
ther
e
we
r
e
only
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
,
a
f
ter
modi
f
ica
ti
on,
we
divi
de
the
s
e
que
nti
a
l
r
e
lations
int
o
be
f
o
r
e
a
nd
mee
t,
whe
r
e
a
s
pa
r
a
ll
e
l
r
e
lation
be
c
omes
the_s
ame
_s
tar
t_t
ime
,
the_s
ame
_c
omple
te_time
,
ov
e
r
lap,
c
ontai
n
a
nd
eq
ual
.
T
a
ble
2
s
hows
pr
oc
e
s
s
models
f
or
a
l
l
tempor
a
l
pa
tt
e
r
ns
.
De
f
ini
ti
on
2.
1
.
T
he
r
e
a
r
e
e
v
e
nt
log
(
EL
)
a
nd
tr
ac
e
(
)
whe
r
e
in
EL
.
T
he
types
of
tempo
r
a
l
pa
tt
e
r
n
be
twe
e
n
a
c
ti
vit
y
X
a
nd
a
c
ti
vit
y
Y;
X
(
X
s
,
X
f
)
a
nd
Y
(
Y
s
,
Y
f
)
,
a
c
c
or
ding
to
whic
h
X
,
Y
EL
c
a
n
be
c
las
s
if
ied
int
o
be
for
e
,
me
e
t,
the_s
ame
_s
tar
t_t
ime
,
the_s
ame
_c
om
plete
_ti
me
,
ov
e
r
lap
,
c
ontai
n
a
nd
e
qual.
De
f
ini
ti
on
2
.
2
.
B
a
s
e
d
on
e
v
e
nt
log
(
EL
)
,
tr
ac
e
(
)
a
nd
a
c
ti
vit
ies
X,Y
EL
,
the
r
e
lations
of
p
r
oc
e
s
s
models
both
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
c
a
n
be
dif
f
e
r
e
n
ti
a
ted
a
s
f
oll
ows
:
N
otR
e
lat
e
d,
#
iff
∧
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
1
,
F
e
br
ua
r
y
2020
:
114
-
123
116
Se
que
nc
e
,
→
if
f
>
∧
P
ar
all
e
l,
∥
iff
>
∧
>
∨
{
∨
@
∨
∨
≈
∨
}
A
N
D,
Y
X
•
iff
Y
X
||
and
ther
e
is
no
Y
X
in
in
E
L
OR
,
Y
X
if
f
Y
X
||
and
ther
e
is
Y
X
in
in
E
L
X
OR
,
Y
X
iff
Y
X
||
if
ther
e
is
only
X
or
Y
in
in
E
L
T
a
ble
1.
T
he
f
r
a
gment
o
f
E
L
1
C
a
s
e
I
D
L
is
t
of
A
c
ti
vi
ti
e
s
S
ta
r
t
T
im
e
C
ompl
e
te
T
im
e
C
a
s
e
I
D
L
is
t
of
A
c
ti
vi
ti
e
s
S
ta
r
t
T
im
e
C
ompl
e
te
T
im
e
Y
M
01
C
ot
to
n B
a
le
s
20/
06/
2017
09.09
20/
06/
2017
12.28
Y
M
02
C
ot
to
n B
a
le
s
23/
06/
2017
04.48
23/
06/
2017
15.15
Y
M
01
B
lo
w
in
g a
nd
P
ic
ki
ng
20/
06/
2017
13.42
20/
06/
2017
19.12
Y
M
02
B
lo
w
in
g a
nd
P
ic
ki
ng
23/
06/
2017
16.44
23/
06/
2017
22.11
Y
M
01
C
a
r
di
ng
20/
06/
2017
23.41
21/
06/
2017
08.32
Y
M
02
P
r
e
C
a
r
di
ng
23/
06/
2017
23.26
24/
06/
2017
07.48
Y
M
01
P
r
e
C
a
r
di
ng
21/
06/
2017
08.16
21/
06/
2017
10.10
Y
M
02
C
a
r
di
ng
24/
06/
2017
04.19
24/
06/
2017
07.48
Y
M
01
S
upe
r
L
a
p
21/
06/
2017
10.46
21/
06/
2017
15.34
Y
M
02
P
r
e
C
a
r
di
ng
24/
06/
2017
07.55
24/
06/
2017
10.10
Y
M
01
C
ombi
ng
21/
06/
2017
16.57
21/
06/
2017
18.09
Y
M
02
S
upe
r
L
a
p
24/
06/
2017
11.33
25/
06/
2017
23.02
Y
M
01
D
r
a
w
in
g
21/
06/
2017
18.09
21/
06/
2017
20.15
Y
M
02
C
ombi
ng
25/
06/
2017
00.30
25/
06/
2017
03.02
Y
M
01
D
r
a
w
in
g I
I
21/
06/
2017
19.15
22/
06/
2017
10.51
Y
M
02
R
ovi
ng
25/
06/
2017
03.16
25/
06/
2017
05.11
Y
M
01
R
ovi
ng
22/
06/
2017
10.51
22/
06/
2017
16.28
Y
M
02
S
pi
nni
ng
25/
06/
2017
06.00
25/
06/
2017
08.25
Y
M
01
S
pi
nni
ng
22/
06/
2017
18.28
22/
06/
2017
23.42
Y
M
02
W
in
di
ng
25/
06/
2017
09.03
25/
06/
2017
12.45
Y
M
01
W
in
di
ng
22/
06/
2017
00.22
23/
06/
2017
04.48
T
a
ble
2.
F
r
a
gment
of
doub
le
ti
mes
tamp
E
L
2
be
fo
r
e
:
w
he
r
e
in
>
iff
≤
m
e
e
t
:
w
he
r
e
in
>
iff
≤
P
r
oc
e
s
s
m
od
e
l:
th
e
_s
am
e
_c
om
pl
e
t
e
_t
ime
:
w
he
r
e
in
iff
=
∧
<
∧
<
P
r
oc
e
s
s
m
od
e
l:
ov
e
r
la
p:
w
he
r
e
in
iff
>
∧
<
c
ont
ai
n:
w
he
r
e
in
@
iff
<
∧
>
∧
>
e
qual
:
w
he
r
e
in
≈
iff
=
∧
=
P
r
oc
e
s
s
m
od
e
l:
th
e
_s
am
e
_s
ta
r
t_
ti
m
e
:
w
he
r
e
in
iff
=
∧
=
P
r
oc
e
s
s
m
od
e
l:
2.
3.
I
n
t
e
gr
at
e
d
d
is
c
ove
r
y
ap
p
r
oac
h
T
he
dis
c
ove
r
y
a
pp
r
oa
c
h
f
o
r
the
ove
r
a
ll
bus
ines
s
pr
oc
e
s
s
is
s
hown
in
F
ig
ur
e
1.
T
his
method
us
e
s
e
ve
nt
log
(
E
L
)
a
s
a
n
input
pr
oc
e
s
s
.
F
o
r
a
ll
c
a
s
e
s
,
we
ne
e
d
to
de
ter
mi
ne
the
tempo
r
a
l
pa
tt
e
r
n
a
n
d
c
r
e
a
te
pr
oc
e
s
s
models
ba
s
e
d
on
tempor
a
l
pa
tt
e
r
n
.
F
r
om
the
pr
oc
e
s
s
models
,
we
c
a
n
obtain
one
bus
ines
s
pr
oc
e
s
s
.
T
he
ne
xt
s
tep
is
to
c
las
s
if
y
a
ll
r
e
lations
o
f
bus
ines
s
pr
oc
e
s
s
(
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l)
.
T
o
e
va
luate
the
bus
ines
s
pr
oc
e
s
s
,
we
c
a
lcula
te
the
f
it
ne
s
s
va
lue
[
19
-
21]
a
nd
c
ompar
e
it
with
the
o
r
igi
na
l
a
lgor
i
thm
.
I
n
the
e
nd,
bus
ines
s
pr
oc
e
s
s
a
nd
it
s
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
(
A
ND
,
OR
,
XO
R
)
r
e
lations
a
r
e
ob
taine
d.
T
his
r
e
s
e
a
r
c
h
pr
opos
e
d
how
to
dis
c
ove
r
pr
oc
e
s
s
models
us
ing
T
im
e
-
ba
s
e
d
α
+
M
iner
a
lgor
it
hm
a
n
d
tempor
a
l
pa
tt
e
r
n
.
W
he
n
the
pr
oc
e
s
s
model
is
obta
ined,
we
c
a
lcula
te
the
f
it
ne
s
s
va
lue
to
e
va
luate
the
pr
oc
e
s
s
model
obtaine
d
by
T
B
A+
M
a
lgor
it
hm
a
s
our
pr
op
os
e
d
met
hod
a
nd
the
pr
oc
e
s
s
model
ge
ne
r
a
ted
by
or
igi
na
l
α
+
a
lgor
it
hm.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
ime
-
bas
e
d
α
+
mine
r
for
mode
ll
ing
bus
ines
s
pr
oc
e
s
s
e
s
us
ing
tem
por
al
patt
e
r
n
(
Y
uti
k
a
A
me
li
a
E
f
fendi
)
117
F
igur
e
1.
P
r
oc
e
s
s
dis
c
ove
r
y
a
ppr
oa
c
h
S
tep
1.
E
ve
nt
log
(
E
L
)
is
the
input
f
or
thi
s
a
ppr
oa
c
h.
Ge
n
e
r
a
ll
y,
C
a
s
e
I
D,
L
is
t
o
f
Ac
ti
vit
ies
a
nd
T
im
e
s
tamp
a
r
e
ne
e
de
d
whe
n
mi
ning
the
pr
oc
e
s
s
models
.
S
tep
2.
E
L
a
s
input
then
is
us
e
d
to
mi
ne
the
pr
oc
e
s
s
mo
de
ls
us
ing
tempor
a
l
pa
tt
e
r
n
a
nd
T
im
e
-
ba
s
e
d
α
+
M
iner
a
lgor
it
hm.
T
he
s
teps
of
pr
oc
e
s
s
dis
c
ove
r
y
u
s
ing
T
im
e
-
ba
s
e
d
α
+
M
iner
a
lgor
it
hm:
S
tep
2.
1
.
De
f
ine
the
tempor
a
l
pa
tt
e
r
n
f
or
a
ll
c
a
s
e
s
in
E
L
S
tep
2.
2
.
Ge
ne
r
a
te
pr
oc
e
s
s
models
ba
s
e
d
on
the
tempor
a
l
p
a
tt
e
r
n
f
or
a
ll
c
a
s
e
s
,
including
the
r
e
lations
of
a
ll
a
c
ti
vit
ies
.
T
he
s
teps
to
ge
ne
r
a
te
pr
oc
e
s
s
models
:
S
tep
2.
2
.
1
Ge
ne
r
a
te
s
e
t
o
f
tr
a
ns
it
ion
(T
L
)
=
{
∈
|
∃
∈
}
(
1)
S
tep
2.
2
.
2
Ge
ne
r
a
te
s
e
t
of
input
(T
I
)
=
{
∈
|
∃
=
(
)
}
(
2)
S
tep
2.
2
.
3
Ge
ne
r
a
te
s
e
t
of
output
(T
o
)
=
{
∈
|
∃
=
(
)
}
(
3)
S
tep
2.
2
.
4
Ge
ne
r
a
te
the
plac
e
s
(
P
L
)
=
{
(
,
)
|
(
,
)
∈
}
∪
{
,
}
(
4)
S
tep
2.
3
.
De
ter
mi
ne
a
ll
tr
a
c
e
s
of
E
L
whic
h
c
a
n
be
obtaine
d
f
r
o
m
the
pr
oc
e
s
s
models
.
T
o
ge
t
a
ll
the
t
r
a
c
e
s
,
the
pr
oc
e
s
s
models
f
r
om
S
tep
2.
2
.
a
r
e
s
tac
ke
d
int
o
one
pr
oc
e
s
s
model.
T
he
tr
a
c
e
s
s
hould
include
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
.
S
tep
2.
4.
Obta
in
one
bus
ines
s
pr
oc
e
s
s
a
s
we
ll
a
s
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
S
tep
2.
5
.
I
de
nti
f
y
s
e
t
o
f
L
oop
(
L
L
)
,
na
mely
len
gth
one
loop
(
L
1L
)
a
nd
length
two
loop
(
L
2L
)
f
r
om
E
L
=
{
∈
T
|
∃
=
1
2
…
∈
;
∈
{
1
,
2
,
…
,
}
[
=
−
1
∧
=
]
}
(
5)
S
tep
2.
6
.
S
pe
c
if
y
the
type
of
pa
r
a
ll
e
l
r
e
lations
:
XO
R
,
OR
or
AND
XO
R
r
e
lation
I
f
A
v
g
P
P
M
≤
M
in
A
SR
in
E
L
,
then
X
OR
(
6)
OR
r
e
lation
I
f
M
in
A
S
R
≤
A
v
g
P
P
M
≤
A
v
g
A
S
R
in
E
L
,
then
OR
(
7)
AN
D
r
e
lation
I
f
A
v
g
A
S
R
≤
A
v
g
P
P
M
in
E
L
,
then
A
N
D
(
8)
whe
r
e
:
M
in
A
S
R
:
mi
nim
um
va
lue
o
f
a
ll
s
e
que
nti
a
l
r
e
lations
in
E
L
A
v
g
A
S
R
:
a
ve
r
a
ge
of
a
ll
s
e
que
nti
a
l
r
e
lations
in
E
L
A
v
g
P
P
M
:
a
ve
r
a
ge
of
pa
r
a
ll
e
l
r
e
lations
with
the
s
a
me
pa
r
e
nt
a
c
ti
vit
y
in
pr
oc
e
s
s
model
(
the
f
r
e
que
nc
y
both
dir
e
c
tl
y
a
nd
indi
r
e
c
tl
y
of
e
a
c
h
a
c
ti
vit
y
f
oll
owe
d
by
other
a
c
ti
vit
y
)
S
tep
2.
7
.
Dis
play
the
f
inal
r
e
s
ult
of
bus
ines
s
pr
oc
e
s
s
in
P
e
tr
i
Ne
t
O
ut
put
:
B
us
in
e
s
s
pr
oc
e
s
s
wi
t
h s
e
qu
e
nt
i
a
l
a
nd
par
a
l
l
e
l
r
e
la
ti
on
s
I
nput
:
E
ve
nt
l
og
D
e
te
r
mi
ne
t
e
mpor
a
l
pa
tt
e
r
n f
or
a
ll
c
a
s
e
s
C
r
e
a
te
pr
oc
e
s
s
mode
ls
f
or
a
ll
c
a
s
e
s
C
la
s
s
if
y t
he
r
e
la
ti
ons
o
f
bu
s
i
ne
s
s
pr
oc
e
s
s
(
s
e
qu
e
nt
i
a
l
a
nd pa
r
a
ll
e
l)
E
va
lu
a
te
f
it
ne
s
s
va
lu
e
of
bus
in
e
s
s
pr
oc
e
s
s
e
s
ob
t
a
i
ne
d b
y
T
B
A+
M
a
nd A
+
M
a
lg
or
it
hm
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
1
,
F
e
br
ua
r
y
2020
:
114
-
123
118
(
)
=
(
,
,
)
(
9)
whe
r
e
:
α
:
A
bus
ines
s
pr
oc
e
s
s
in
P
e
tr
i
Ne
t
f
o
r
m
P
:
plac
e
I
n
the
bus
ines
s
pr
oc
e
s
s
,
it
is
s
ymbol
ize
d
by
c
ir
c
le
s
ha
pe
a
nd
na
med
a
s
p1,
p2
,
p3
,
e
tc
.
T
:
tr
a
ns
it
ion
I
n
the
bus
ines
s
pr
oc
e
s
s
,
it
s
hows
the
a
c
ti
vit
ies
with
s
qua
r
e
s
ha
pe
F
:
f
unc
ti
on
I
n
the
bus
ines
s
pr
oc
e
s
s
,
f
unc
ti
on
mea
ns
a
n
a
r
c
.
F
=
{P
x
T
}
mea
ns
the
a
r
c
c
onne
c
ts
the
plac
e
(
P
)
a
nd
tr
a
ns
it
ion
(
T
)
S
tep
3.
E
va
luation
o
f
the
pr
oc
e
s
s
model.
W
e
c
a
lcula
te
t
he
f
it
ne
s
s
va
lue
a
f
ter
p
r
oc
e
s
s
model
is
obtaine
d
[
19]
,
a
nd
then
we
c
ompar
e
the
r
e
s
ult
s
be
twe
e
n
p
r
oc
e
s
s
model
obtaine
d
by
T
im
e
-
ba
s
e
d
α
+
M
iner
a
lg
or
it
hm
a
nd
pr
oc
e
s
s
model
ge
ne
r
a
ted
f
r
om
α
+
M
iner
a
lgor
it
hm
.
T
he
f
it
ne
s
s
va
lue
is
c
a
lcula
ted
us
ing
(
10)
.
=
(
10)
whe
r
e
:
Tf
:
the
f
i
tnes
s
va
lue
C
aptC
as
e
s
:
number
o
f
c
a
ptur
e
d
c
a
s
e
s
in
E
L
L
ogC
as
e
s
:
number
o
f
c
a
s
e
s
in
E
L
3.
RE
S
UL
T
S
AN
D
AN
AL
YSI
S
I
n
thi
s
s
e
c
ti
on,
we
do
the
e
xpe
r
im
e
nt
whic
h
wil
l
pr
e
s
e
nt
the
r
e
s
ult
s
that
ou
r
pr
opos
e
d
method
c
a
n
mi
ne
bus
ines
s
pr
oc
e
s
s
e
s
us
ing
T
im
e
-
ba
s
e
d
α
+
M
iner
a
lgor
it
hm
a
nd
tempor
a
l
pa
tt
e
r
n
.
As
ment
ioned
in
s
e
c
ti
on
2.
1
.
,
a
r
e
a
l
-
li
f
e
e
ve
nt
log
is
tes
ted
a
nd
a
na
lyze
d
i
n
thi
s
e
xpe
r
im
e
nt.
E
ve
nt
log
E
L
1
is
in
double
ti
mes
tamp
a
s
e
xplaine
d
in
T
a
ble
1
.
3.
1.
E
xp
e
r
im
e
n
t
al
r
e
s
u
lt
s
F
r
om
e
ve
nt
log
E
L
1
pr
e
s
e
nted
in
T
a
ble
1
,
we
ha
v
e
100
c
a
s
e
s
a
nd
11
a
c
ti
vit
ies
to
be
im
pleme
nted
in
thi
s
e
xpe
r
im
e
nt.
B
a
s
e
d
on
a
ll
s
teps
e
xplaine
d
in
s
e
c
ti
o
n
2.
3
.
,
we
ne
e
d
to
de
ter
mi
ne
the
input
a
nd
th
e
type
of
tempor
a
l
pa
tt
e
r
n
whic
h
a
r
e
us
e
d
f
or
a
ll
c
a
s
e
s
.
All
c
a
s
e
s
in
the
e
ve
nt
log
E
L
1
ha
ve
the
s
a
me
input
whic
h
is
a
c
ti
vit
y
C
ott
on
B
ales
.
T
he
r
e
s
ult
s
of
tempor
a
l
pa
tt
e
r
n
f
or
two
c
a
s
e
s
(
YM
01
a
nd
YM
02)
of
E
L
1
a
r
e
p
r
e
s
e
nted
in
T
a
ble
3
.
E
L
1
c
ons
is
ts
of
be
for
e
,
ov
e
r
lap,
me
e
t,
t
he
_s
ame
_c
omple
te_tim
e
,
the_s
ame
_s
tar
t_t
ime
a
nd
c
ontai
n
f
or
a
ll
100
c
a
s
e
s
.
How
e
ve
r
,
the
tempo
r
a
l
pa
t
ter
n
e
qual
is
not
e
xplaine
d
in
thi
s
e
xpe
r
im
e
nt.
T
his
is
be
c
a
us
e
e
qual
doe
s
not
oc
c
ur
in
the
e
ve
nt
log
E
L
1
a
f
ter
we
a
na
lys
is
it
ba
s
e
d
on
the
s
tar
t
ti
me
a
nd
c
ompl
e
te
t
im
e
.
Af
ter
a
ll
tempor
a
l
pa
tt
e
r
ns
f
r
om
the
E
L
1
we
r
e
obtaine
d,
we
ne
e
d
to
ge
ne
r
a
te
the
pr
oc
e
s
s
models
including
the
r
e
lations
o
f
a
ll
a
c
ti
vit
ies
ba
s
e
d
on
th
e
tempor
a
l
pa
tt
e
r
n
f
o
r
a
ll
c
a
s
e
s
by
f
oll
owing
the
r
e
s
ult
s
in
T
a
ble
3
a
nd
the
de
f
ini
ti
on
in
s
e
c
ti
on
2.
2
.
I
n
a
ddi
ti
on,
we
ne
e
d
to
c
r
e
a
te
the
t
r
a
ns
it
ion
a
nd
plac
e
,
a
nd
a
ls
o
de
ter
mi
ne
the
input
a
nd
output
f
or
e
a
c
h
c
a
s
e
by
f
ol
lowing
(
1)
,
(
2)
,
(
3
)
,
(
4)
.
T
a
ble
3.
T
e
mpor
a
l
pa
tt
e
r
n
f
or
two
c
a
s
es
C
a
s
e
I
D
L
is
t
of
A
c
ti
vi
ti
e
s
C
a
s
e
I
D
L
is
t
of
A
c
ti
vi
ti
e
s
Y
M
01
C
ot
to
n ba
le
s
be
fo
r
e
B
lo
w
in
g a
nd P
ic
ki
ng
Y
M
02
C
ot
to
n ba
le
s
be
fo
r
e
B
lo
w
in
g a
nd P
ic
ki
ng
Y
M
01
B
lo
w
in
g a
nd P
ic
ki
ng
be
fo
r
e
C
a
r
di
ng
Y
M
02
B
lo
w
in
g a
nd P
ic
ki
ng
be
fo
r
e
P
r
e
C
a
r
di
ng
Y
M
01
C
a
r
di
ng
ov
e
r
la
p
P
r
e
C
a
r
di
ng
Y
M
02
P
r
e
C
a
r
di
ng
th
e
_s
am
e
_c
om
pl
e
te
_t
ime
C
a
r
di
ng
Y
M
01
P
r
e
C
a
r
di
ng
be
fo
r
e
S
upe
r
L
a
p
Y
M
02
C
a
r
di
ng
be
fo
r
e
P
r
e
C
a
r
di
ng
Y
M
01
S
upe
r
L
a
p
be
fo
r
e
C
ombi
ng
Y
M
02
P
r
e
C
a
r
di
ng
be
fo
r
e
S
upe
r
L
a
p
Y
M
01
C
ombi
ng
be
fo
r
e
D
r
a
w
in
g
Y
M
02
S
upe
r
L
a
p
be
fo
r
e
C
ombi
ng
Y
M
01
D
r
a
w
in
g
ov
e
r
la
p
D
r
a
w
in
g I
I
Y
M
02
C
ombi
ng
be
fo
r
e
R
ovi
ng
Y
M
01
D
r
a
w
in
g I
I
m
e
e
t
R
ovi
ng
Y
M
02
R
ovi
ng
be
fo
r
e
S
pi
nni
ng
Y
M
01
R
ovi
ng
be
fo
r
e
S
pi
nni
ng
Y
M
02
S
pi
nni
ng
be
fo
r
e
W
in
di
ng
Y
M
01
S
pi
nni
ng
be
fo
r
e
W
in
di
ng
B
a
s
e
d
on
E
L
1,
C
a
s
e
I
D
YM
01
ha
s
a
c
ti
vit
y
C
ott
on
B
ales
a
s
input
a
nd
a
c
ti
vit
y
Spinn
ing
a
s
output
a
s
we
ll
a
s
C
a
s
e
I
D
YM
02.
T
r
a
ns
it
ion
p
r
e
s
e
nts
the
na
me
of
a
c
ti
vit
y
,
mea
nwhile
plac
e
is
de
f
ined
to
c
on
ne
c
t
two
a
c
ti
vit
ies
in
a
P
e
tr
i
Ne
t
f
or
m
.
I
n
thi
s
e
xpe
r
im
e
nt
,
w
e
na
me
the
plac
e
a
s
p1,
p2
,
p3
,
e
tc.
F
igur
e
2
s
hows
the
pr
oc
e
s
s
models
ba
s
e
d
on
T
a
ble
3
f
or
C
a
s
e
I
D
YM
01
a
nd
YM
02.
F
or
the
r
e
s
t
of
98
c
a
s
e
s
,
we
do
the
s
a
me
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
ime
-
bas
e
d
α
+
mine
r
for
mode
ll
ing
bus
ines
s
pr
oc
e
s
s
e
s
us
ing
tem
por
al
patt
e
r
n
(
Y
uti
k
a
A
me
li
a
E
f
fendi
)
119
s
teps
a
s
C
a
s
e
I
D
YM
01
a
nd
YM
02.
F
igur
e
3
p
r
e
s
e
nts
the
dis
c
ove
r
e
d
pr
oc
e
s
s
models
f
o
r
C
a
s
e
I
D
Y
M
01
a
n
d
YM
02
a
nd
a
ls
o
e
mphas
ize
s
that
a
ll
c
a
s
e
s
of
E
L
1
c
ons
is
t
of
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
.
F
igur
e
2.
P
r
oc
e
s
s
models
ge
ne
r
a
ted
f
r
om
tempo
r
a
l
pa
tt
e
r
n
f
or
c
a
s
e
I
D
YM
01
a
nd
YM
02
F
igur
e
3.
P
r
oc
e
s
s
models
f
or
c
a
s
e
I
D
YM
01
a
nd
Y
M
02
Ne
xt
s
tep,
we
a
ls
o
ne
e
d
to
ge
t
a
ll
tr
a
c
e
s
f
r
om
a
ll
c
a
s
e
s
of
E
L
1.
T
r
a
c
e
s
r
e
c
or
ds
s
e
qu
e
nc
e
of
a
c
ti
vit
ies
whic
h
be
long
to
the
s
a
me
c
a
s
e
[
22]
.
T
he
e
a
s
y
wa
y
to
ge
t
a
ll
tr
a
c
e
s
,
the
two
p
r
oc
e
s
s
models
in
F
igu
r
e
3
a
nd
the
other
98th
c
a
s
e
s
a
r
e
s
ta
c
ke
d
int
o
one
bus
i
n
e
s
s
pr
oc
e
s
s
.
All
tr
a
c
e
s
of
E
L
1
a
r
e
pr
e
s
e
nted
in
T
a
ble
4.
T
he
tot
a
l
number
o
f
t
r
a
c
e
s
in
E
L
1
a
r
e
12
tr
a
c
e
s
.
B
e
s
ides
ge
tt
ing
the
number
o
f
t
r
a
c
e
s
,
we
a
ls
o
ge
t
the
bus
ines
s
pr
oc
e
s
s
of
E
L
1
whic
h
i
s
the
r
e
s
ult
of
pr
oc
e
s
s
dis
c
ov
e
r
y
us
ing
T
B
A+
M
a
lgor
it
hm.
T
he
b
us
ines
s
pr
oc
e
s
s
of
E
L
1
is
s
hown
in
F
igur
e
4.
Af
ter
the
bus
ines
s
pr
oc
e
s
s
of
E
L
1
including
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
wa
s
obtaine
d,
now
we
mus
t
identif
y
whe
ther
thi
s
e
ve
nt
log
E
L
1
c
ontains
length
one
loop
(
L
1L
)
a
nd
length
two
loop
(
L
2L
)
by
f
oll
owing
(
5)
.
L
1L
a
nd
L
2L
mea
n
on
e
a
c
ti
vit
y
in
one
c
a
s
e
may
be
e
x
e
c
uted
mul
ti
ple
ti
mes
.
T
his
c
ondit
ion
is
c
omm
only
known
a
s
loop
[
16
,
23
,
24
]
.
W
e
do
a
na
lyzing
f
o
r
a
ll
100
c
a
s
e
s
to
f
ind
L
1L
a
nd
L
2L
,
a
nd
the
r
e
s
ult
s
s
how
that
ther
e
a
r
e
s
ome
a
c
ti
vit
ies
whic
h
e
xe
c
uted
mul
ti
ple
ti
mes
in
E
L
1.
T
a
ble
5
pr
e
s
e
nts
the
a
c
ti
vit
ies
whic
h
a
r
e
c
a
tegor
ize
d
a
s
L
1L
a
nd
L
2L
in
the
pr
oc
e
s
s
models
.
C
a
se
I
D
Y
M
0
1
C
a
se
ID
Y
M
0
2
C
a
se
I
D
Y
M
0
1
C
a
se
ID
Y
M
0
2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
1
,
F
e
br
ua
r
y
2020
:
114
-
123
120
T
a
ble
4.
T
r
a
c
e
s
of
E
L
1
T
R
A
C
E
S
1
T
R
A
C
E
S
2
T
R
A
C
E
S
3
T
R
A
C
E
S
4
T
R
A
C
E
S
5
T
R
A
C
E
S
6
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
S
upe
r
L
a
p
S
upe
r
L
a
p
S
upe
r
L
a
p
S
upe
r
L
a
p
S
upe
r
L
a
p
S
upe
r
L
a
p
C
ombi
ng
C
ombi
ng
C
ombi
ng
C
ombi
ng
C
ombi
ng
C
ombi
ng
D
r
a
w
in
g
D
r
a
w
in
g
D
r
a
w
in
g I
I
D
r
a
w
in
g I
I
R
ovi
ng
R
ovi
ng
D
r
a
w
in
g I
I
D
r
a
w
in
g I
I
D
r
a
w
in
g
D
r
a
w
in
g
S
pi
nni
ng
S
pi
nni
ng
R
ovi
ng
R
ovi
ng
R
ovi
ng
R
ovi
ng
W
in
di
ng
W
in
di
ng
S
pi
nni
ng
S
pi
nni
ng
S
pi
nni
ng
S
pi
nni
ng
W
in
di
ng
W
in
di
ng
W
in
di
ng
W
in
di
ng
T
R
A
C
E
S
7
T
R
A
C
E
S
8
T
R
A
C
E
S
9
T
R
A
C
E
S
10
T
R
A
C
E
S
11
T
R
A
C
E
S
12
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
C
ot
to
n B
a
le
s
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
B
lo
w
in
g a
nd
P
ic
ki
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
a
r
di
ng
C
a
r
di
ng
P
r
e
C
a
r
di
ng
S
upe
r
L
a
p
S
upe
r
L
a
p
S
upe
r
L
a
p
S
upe
r
L
a
p
P
r
e
C
a
r
di
ng
P
r
e
C
a
r
di
ng
C
ombi
ng
C
ombi
ng
C
ombi
ng
C
ombi
ng
S
upe
r
L
a
p
S
upe
r
L
a
p
D
r
a
w
in
g
D
r
a
w
in
g
D
r
a
w
in
g I
I
D
r
a
w
in
g I
I
C
ombi
ng
C
ombi
ng
R
ovi
ng
R
ovi
ng
R
ovi
ng
R
ovi
ng
R
ovi
ng
D
r
a
w
in
g
S
pi
nni
ng
S
pi
nni
ng
S
pi
nni
ng
S
pi
nni
ng
S
pi
nni
ng
R
ovi
ng
W
in
di
ng
W
in
di
ng
W
in
di
ng
W
in
di
ng
W
in
di
ng
S
pi
nni
ng
W
in
di
ng
F
igur
e
4.
B
us
ines
s
pr
oc
e
s
s
o
f
E
L
1
dis
c
ove
r
e
d
by
u
s
ing
the
pr
opos
e
d
method
T
a
ble
5.
Ac
ti
vit
ies
e
xe
c
uted
mul
ti
ple
ti
mes
in
E
L
1
L
is
t
of
A
c
ti
vi
ti
e
s
L
1L
L
2L
P
r
e
C
a
r
di
ng
Y
e
s
-
C
a
r
di
ng
-
Y
e
s
3.
2.
De
t
e
r
m
in
in
g
p
ar
all
e
l
r
e
lat
ion
s
o
f
d
is
c
ove
r
e
d
b
u
s
in
e
s
s
p
r
oc
e
s
s
W
he
n
dis
c
ove
r
e
d
bus
ines
s
pr
oc
e
s
s
is
obtaine
d,
the
ne
xt
s
tep
in
ou
r
pr
oc
e
s
s
dis
c
ove
r
y
a
ppr
oa
c
h
is
to
de
ter
mi
ne
the
type
o
f
pa
r
a
ll
e
l
r
e
lations
;
XO
R
,
O
R
,
AN
D.
W
e
us
e
(
6
)
,
(
7)
,
a
nd
(
8)
to
s
pe
c
if
y
the
type
of
pa
r
a
ll
e
l
r
e
lations
.
T
he
f
r
e
que
nc
ies
of
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
whic
h
a
r
e
c
a
lcula
ted
f
r
om
a
ll
c
a
s
e
s
of
E
L
1
a
r
e
a
c
ti
vit
y
C
ott
on
B
ales
a
nd
a
c
ti
vit
y
B
low
ing
and
P
ick
ing
ha
ve
100
r
e
lations
,
a
c
ti
vit
y
B
low
ing
and
P
ick
ing
a
nd
a
c
ti
vit
y
C
ar
ding
ha
ve
77
r
e
lations
,
a
c
ti
vit
y
B
low
ing
and
P
ick
ing
a
nd
a
c
ti
vit
y
P
r
e
C
ar
ding
c
ons
is
t
of
23
r
e
lations
.
M
e
a
nwhile,
a
c
ti
vit
ies
C
a
r
ding
a
nd
Supe
r
lap
,
a
c
ti
vit
ies
P
r
e
C
ar
ding
a
nd
Supe
r
lap
ha
ve
23
a
nd
77
r
e
s
pe
c
ti
ve
ly.
Ac
ti
vit
y
Supe
r
lap
a
nd
a
c
ti
vit
y
C
ombing
ha
ve
100
r
e
lations
in
the
E
L
1.
I
n
a
ddit
ion,
Ac
ti
vit
y
C
ombing
whic
h
pa
ir
with
a
c
ti
vi
ti
e
s
Dr
aw
ing
,
Dr
aw
ing
I
I
a
nd
R
ov
ing
c
ons
is
t
of
7
3,
12,
15
r
e
s
pe
c
ti
ve
ly.
Ac
ti
vit
ies
Dr
aw
ing
a
nd
Dr
aw
ing
I
I
ha
ve
60
r
e
lations
,
whe
r
e
a
s
a
c
ti
vit
ies
Dr
aw
ing
I
I
a
nd
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
ime
-
bas
e
d
α
+
mine
r
for
mode
ll
ing
bus
ines
s
pr
oc
e
s
s
e
s
us
ing
tem
por
al
patt
e
r
n
(
Y
uti
k
a
A
me
li
a
E
f
fendi
)
121
Dr
aw
ing
ha
ve
9
r
e
lations
.
Ac
ti
vit
ies
Dr
aw
ing
a
n
d
Dr
aw
ing
I
I
lea
d
to
a
c
ti
vit
y
R
ov
ing
c
ons
is
t
of
13
a
nd
3
r
e
lations
.
L
a
s
t,
a
c
ti
vit
y
R
ov
ing
a
nd
a
c
ti
vit
y
Sp
inn
ing
,
a
c
ti
vit
y
Sp
inni
ng
a
nd
a
c
ti
vit
y
W
indi
ng
ha
ve
t
he
s
a
me
100
r
e
lations
.
All
of
them
a
r
e
in
s
e
que
nti
a
l
r
e
lati
ons
.
F
or
a
c
ti
vit
ies
C
ar
ding
a
nd
P
r
e
C
ar
ding
a
nd
a
c
ti
vit
ies
Dr
aw
ing
,
Dr
aw
ing
I
I
a
nd
R
ov
ing
ha
ve
100
a
nd
69
r
e
s
pe
c
ti
ve
ly.
T
he
y
a
r
e
in
pa
r
a
ll
e
l
r
e
lations
.
F
or
the
ne
xt
s
tep,
we
ne
e
d
to
c
a
lcula
te
m
ini
mum
a
ll
s
e
que
nti
a
l
r
e
lations
,
a
ve
r
a
ge
pa
r
a
ll
e
l
r
e
lations
a
nd
a
ve
r
a
ge
a
ll
s
e
que
nti
a
l
r
e
lations
.
T
a
ble
6
pr
e
s
e
nts
the
r
e
s
ult
s
of
E
L
1
f
or
the
va
lue
of
M
in
ASR
,
Avg
P
P
M
a
nd
Avg
ASR
.
F
r
om
T
a
ble
6
we
ge
t
the
in
f
or
ma
ti
on
that
a
c
ti
vit
ies
Dr
aw
ing
,
Dr
aw
ing
I
I
a
nd
R
ov
i
ng
ha
ve
the
va
lue
of
Avg
P
P
M
higher
than
the
va
lue
o
f
M
i
n
ASR
,
but
the
va
lue
of
Avg
P
P
M
is
the
s
a
me
a
s
the
va
lue
of
Avg
ASR
.
S
o,
the
f
inal
r
e
lation
of
a
c
ti
vit
ies
Dr
aw
ing
,
Dr
aw
ing
I
I
a
nd
R
ov
ing
is
OR
a
s
e
xpla
ined
(
7)
.
W
hil
e
,
f
or
a
c
ti
vit
ies
C
ar
ding
a
nd
P
r
e
C
ar
ding
,
th
e
va
lue
of
Avg
P
P
M
is
higher
than
the
va
lue
of
A
vg
ASR
.
B
a
s
e
d
on
(
8)
,
the
r
e
lation
is
AN
D.
T
he
ve
r
y
las
t
s
tep
of
our
pr
opos
e
d
a
ppr
oa
c
h,
we
pr
e
s
e
nt
the
AN
D
a
nd
OR
r
e
lations
in
P
e
tr
i
Ne
t
f
or
m
int
o
the
dis
c
ove
r
e
d
bus
ines
s
pr
oc
e
s
s
of
E
L
1
by
f
oll
owing
(
9
)
.
F
i
gu
r
e
5
s
hows
the
f
inal
r
e
s
ult
of
the
bus
ines
s
pr
oc
e
s
s
of
E
L
1
a
long
with
AN
D,
OR
r
e
lations
a
nd
L
1L
,
L
2L
.
T
a
ble
6.
T
he
va
lue
of
mi
n
ASR
,
a
vg
P
P
M
a
nd
a
vg
ASR
to
de
ter
mi
ne
the
pa
r
a
ll
e
l
o
f
E
L
1
P
a
r
a
ll
e
l
A
c
ti
vi
ti
e
s
T
he
va
lu
e
of
M
in
A
S
R
T
he
va
lu
e
of
A
vg P
P
M
T
he
va
lu
e
of
A
vg A
S
R
P
a
r
a
ll
e
l
G
a
te
w
a
y
C
a
r
di
ng
3
100
52.3
AND
P
r
e
C
a
r
di
ng
D
r
a
w
in
g
52.3
OR
D
r
a
w
in
g I
I
R
ovi
ng
F
igur
e
5.
A
bus
ines
s
pr
oc
e
s
s
of
E
L
1
with
AN
D,
O
R
,
L
1L
,
L
2L
by
us
ing
T
B
A+
M
a
lgor
it
hm
3.
3.
E
valu
at
ion
of
t
h
e
b
u
s
in
e
s
s
p
r
oc
e
s
s
I
n
thi
s
r
e
s
e
a
r
c
h,
two
c
ompar
is
on
s
teps
we
r
e
done
to
e
va
luate
the
bus
ines
s
pr
oc
e
s
s
.
F
ir
s
tl
y,
c
ompar
ing
the
pa
r
a
ll
e
l
r
e
lations
be
twe
e
n
the
pr
opos
e
d
a
lgor
it
hm
(
T
B
A+
M
)
a
nd
the
o
r
igi
na
l
a
l
gor
it
hm
(
α+
M
iner
)
.
S
e
c
ondly,
the
f
it
ne
s
s
va
lue
of
the
b
us
ines
s
pr
oc
e
s
s
is
c
a
lcula
ted
us
ing
(
10)
.
T
a
ble
7
pr
e
s
e
nts
the
pa
r
a
ll
e
l
r
e
lations
dis
c
ove
r
e
d
by
us
ing
α+
M
i
ne
r
a
lgor
it
hm
a
nd
F
igu
r
e
6
s
hows
the
bus
ines
s
pr
oc
e
s
s
in
the
P
e
tr
i
Ne
t
f
or
m
f
or
the
s
a
me
c
a
s
e
s
of
E
L
1
in
T
a
ble
1
obtaine
d
by
α+
M
iner
a
lgo
r
it
hm.
T
he
main
di
f
f
e
r
e
nc
e
be
twe
e
n
F
igur
e
s
5
a
nd
6
is
we
c
a
n
obtain
OR
r
e
lation
in
the
bus
ines
s
pr
oc
e
s
s
in
the
F
igu
r
e
5
,
mea
nwhile
we
c
a
n
only
dis
ti
nguis
h
AN
D
r
e
lation
in
the
F
igur
e
6
a
mong
a
c
ti
vit
ies
Dr
a
wing,
D
r
a
wing
I
I
a
nd
R
oving.
Ge
ne
r
a
ll
y,
pr
oc
e
s
s
dis
c
ove
r
y
us
ing
the
or
igi
na
l
α+
M
iner
a
lgor
it
hm
us
e
s
s
ingl
e
ti
mes
tamp
e
ve
nt
log
[
16
,
25,
26
]
.
I
t
c
a
nnot
dis
ti
nguis
h
the
di
f
f
e
r
e
nc
e
s
be
twe
e
n
AN
D
a
nd
OR
in
pa
r
a
ll
e
l
r
e
lations
.
T
he
or
igi
na
l
α
+
M
iner
a
lgor
it
hm
ha
s
diffi
c
ult
ies
in
in
ter
pr
e
ti
ng
OR
in
a
bus
ines
s
pr
oc
e
s
s
.
T
his
is
be
c
a
us
e
the
o
r
i
ginal
α
+
M
iner
a
lgor
it
hm
only
ha
s
2
pa
r
a
ll
e
l
r
e
lations
,
na
mely
AN
D
a
nd
XO
R
[
16,
24
]
.
I
f
one
a
c
ti
vit
y
is
e
xe
c
uted
e
xa
c
tl
y
onc
e
in
bus
ines
s
pr
oc
e
s
s
,
then
it
is
c
a
tegor
ize
d
a
s
XO
R
,
mea
nwhile
AN
D
ha
ppe
ns
whe
n
one
or
mor
e
a
c
ti
vit
ies
a
r
e
a
ll
owe
d
to
be
e
xe
c
uted
mor
e
than
onc
e
in
bus
ines
s
pr
oc
e
s
s
.
T
he
r
e
f
or
e
,
OR
in
the
bus
ines
s
pr
oc
e
s
s
will
be
e
qua
ted
with
XO
R
or
AN
D,
e
ve
n
though
in
r
e
a
li
ty
thos
e
thr
e
e
r
e
lations
(
AN
D,
O
R
,
XO
R
)
ha
ve
dif
f
e
r
e
nc
e
s
in
how
to
e
xe
c
ute
a
c
ti
vit
ies
in
the
e
ve
nt
log
[
4,
6,
13]
.
W
e
c
a
n
c
onc
lu
de
that
the
p
r
opos
e
d
a
lgor
it
hm
is
be
tt
e
r
in
de
ter
mi
ning
pa
r
a
ll
e
l
r
e
lations
(
AN
D,
OR
)
than
the
or
igi
na
l
a
lgor
it
hm
.
T
he
c
a
lcula
ti
on
of
f
it
ne
s
s
va
lue
be
c
omes
the
las
t
s
tep
f
o
r
e
va
luating
the
bus
ines
s
pr
oc
e
s
s
.
F
it
ne
s
s
va
lue
is
de
c
lar
e
d
good
if
the
va
lue
is
c
lo
s
e
r
to
1
[
1
9,
20]
.
T
he
c
a
lcula
ti
on
r
e
s
ult
s
of
f
it
ne
s
s
va
lue
pr
e
s
e
nted
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
1
,
F
e
br
ua
r
y
2020
:
114
-
123
122
T
a
ble
8
give
the
inf
o
r
mation
that
the
bus
ines
s
pr
oc
e
s
s
obtaine
d
by
us
ing
the
pr
opos
e
d
a
lgor
it
hm
f
it
s
the
r
e
a
li
ty
be
tt
e
r
than
that
of
the
o
r
igi
na
l
a
lgo
r
it
h
m.
All
c
a
s
e
s
in
E
L
1
c
a
n
be
c
a
ptur
e
d
a
nd
c
a
n
be
modelled
int
o
a
bus
ines
s
pr
oc
e
s
s
.
F
r
om
the
e
va
luation
s
tage
,
we
c
a
n
c
onc
lude
tha
t
our
pr
opos
e
d
a
lgor
it
h
m
c
a
n
mi
ne
the
bus
ines
s
pr
oc
e
s
s
a
s
we
ll
a
s
the
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
(
AN
D,
OR
,
XO
R
)
r
e
lations
,
whic
h
c
a
nnot
be
ob
taine
d
by
the
or
ig
inal
a
lgor
it
hm.
T
he
r
e
s
ult
s
of
f
it
ne
s
s
va
lu
e
a
ls
o
c
lea
r
ly
s
tate
that
the
ne
w
p
r
opos
e
d
a
lgor
it
hm
gives
the
be
s
t
va
lue.
He
nc
e
,
T
i
me
-
ba
s
e
d
α
+
M
iner
a
lgor
it
hm
c
a
n
be
a
n
e
xtende
d
ve
r
s
ion
of
α
+
M
iner
a
lgo
r
it
hm
in
pr
oc
e
s
s
dis
c
ove
r
y.
F
igur
e
6.
A
bus
ines
s
pr
oc
e
s
s
of
E
L
1
with
AN
D,
A
ND
,
L
1L
,
L
2L
ge
ne
r
a
ted
by
α+
M
iner
T
a
ble
7.
T
he
pa
r
a
ll
e
l
r
e
lations
o
f
E
L
1
dis
c
ove
r
e
d
us
ing
α
+
M
iner
P
a
r
a
ll
e
l
A
c
ti
vi
ti
e
s
P
a
r
a
ll
e
l
R
e
la
ti
ons
C
a
r
di
ng
P
r
e
C
a
r
di
ng
AND
D
r
a
w
in
g
D
r
a
w
in
g I
I
R
ovi
ng
AND
T
a
ble
8.
T
he
c
ompa
r
is
on
of
f
it
ne
s
s
va
lu
e
of
E
L
1
A
lg
or
it
hms
F
it
ne
s
s
V
a
lu
e
T
im
e
-
ba
s
e
d
α
+
M
in
e
r
1.000
α+
M
in
e
r
0.954
4.
CONC
L
USI
ON
T
his
r
e
s
e
a
r
c
h
pa
pe
r
f
oc
us
e
d
on
int
r
oduc
ing
a
ne
w
a
ppr
oa
c
h
to
mi
ne
the
bus
ines
s
pr
oc
e
s
s
e
s
us
ing
tempor
a
l
pa
tt
e
r
n
whic
h
im
pleme
nted
in
T
i
me
-
ba
s
e
d
α
+
M
iner
(
T
B
A+
M
)
a
lgor
it
hm.
T
he
pr
opos
e
d
method
de
f
ined
s
e
ve
n
tempor
a
l
pa
tt
e
r
ns
whic
h
include
d
both
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
r
e
lations
.
T
he
y
a
r
e
be
fo
r
e
,
me
e
t,
the_s
ame
_s
ta
r
t_t
ime
,
the_s
ame
_c
omple
te_time
,
ov
e
r
lap
,
c
ontai
n
a
nd
e
qual
.
T
B
A+
M
a
lgor
it
hm
wa
s
the
e
xtende
d
ve
r
s
ion
o
f
α
+
M
iner
a
lgor
it
hm.
T
B
A+
M
c
ombi
ne
d
with
tempor
a
l
pa
tt
e
r
n
we
r
e
us
e
d
to
mi
ne
the
p
r
oc
e
s
s
model
f
r
o
m
the
e
ve
nt
log
.
C
ompar
e
d
to
or
igi
na
l
α
+
M
iner
a
lgor
it
hm,
thi
s
modi
f
ied
a
lgor
it
hm
c
ould
de
tec
t
o
ve
r
lapping
ti
mes
tamp
us
ing
tempor
a
l
pa
tt
e
r
n
a
nd
identif
ies
it
a
s
a
pa
r
a
ll
e
l
pr
oc
e
s
s
,
dis
c
ove
r
e
d
length
one
loop
(
L
1L
)
,
length
two
loop
(
L
2L
)
a
nd
s
pe
c
if
ied
the
s
e
que
nti
a
l
a
nd
pa
r
a
ll
e
l
(
AN
D,
OR
,
XO
R
)
r
e
lations
in
a
c
or
r
e
c
t
wa
y.
M
e
a
nwhile
α
+
M
iner
a
lgor
it
hm
wa
s
not
a
ble
to
dis
ti
nguis
h
the
dif
f
e
r
e
nc
e
s
be
twe
e
n
AN
D
a
nd
OR
i
n
pa
r
a
ll
e
l
r
e
lations
.
T
he
e
xpe
r
im
e
ntal
r
e
s
ult
s
pr
e
s
e
nted
that
our
ne
w
p
r
oc
e
s
s
dis
c
ove
r
y
a
ppr
oa
c
h
c
ould
mi
ne
the
bus
ines
s
pr
oc
e
s
s
e
s
with
pa
r
a
ll
e
l
AN
D
a
nd
OR
r
e
lations
,
whic
h
c
ould
not
be
obtaine
d
by
the
o
r
igi
na
l
α+
M
iner
a
lgor
it
hm.
I
n
a
ddit
ion
,
the
r
e
s
ult
s
of
f
it
ne
s
s
va
lu
e
a
ls
o
c
lea
r
ly
s
tate
d
that
the
ne
w
method
a
ls
o
ga
ve
be
tt
e
r
r
e
s
ult
r
a
ther
than
that
o
f
the
or
igi
na
l
α
+
M
iner
a
l
go
r
it
hm
AC
KNOWL
E
DGE
M
E
NT
S
T
he
c
ompl
e
ti
on
of
thi
s
r
e
s
e
a
r
c
h
c
ould
not
ha
ve
b
e
e
n
pos
s
ibl
e
without
the
s
uppor
t
f
r
om
C
omput
e
r
L
a
bor
a
tor
y,
De
pa
r
tm
e
nt
of
I
nf
o
r
mation
S
ys
tem
s
,
Unive
r
s
it
a
s
Air
langga
a
nd
I
nf
o
r
mation
M
a
na
ge
ment
L
a
bor
a
tor
y,
De
pa
r
tm
e
nt
o
f
I
nf
or
matics
,
I
ns
ti
tut
T
e
knologi
S
e
puluh
Nope
mber
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
T
ime
-
bas
e
d
α
+
mine
r
for
mode
ll
ing
bus
ines
s
pr
oc
e
s
s
e
s
us
ing
tem
por
al
patt
e
r
n
(
Y
uti
k
a
A
me
li
a
E
f
fendi
)
123
RE
F
E
RE
NC
E
S
[1
]
X
.
W
an
g
,
P.
H
u
an
g
,
S.
W
u
,
an
d
Z
.
Sh
i
,
“
Meas
u
reme
n
t
o
f
Sy
n
er
g
et
i
c
D
e
g
ree
o
f
E
n
t
er
p
ri
s
e
E
-
C
o
mmerce
Bu
s
i
n
es
s
Pro
ces
s
,
”
F
o
u
r
t
h
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
e
r
e
n
ce
o
n
B
u
s
i
n
es
s
In
t
el
l
i
g
en
ce
a
n
d
F
i
n
a
n
c
i
a
l
E
n
g
i
n
eer
i
n
g
,
2011
,
doi
:
1
0
.
1
1
0
9
/
BIFE
.
2
0
1
1
.
8
1
.
[2
]
J
.
G
u
o
an
d
Y
.
Z
o
u
,
“
A
Bu
s
i
n
es
s
Pro
ces
s
E
x
p
l
o
rer:
Rec
o
v
er
i
n
g
Bu
s
i
n
e
s
s
Pro
ce
s
s
e
s
fro
m
Bu
s
i
n
e
s
s
A
p
p
l
i
ca
t
i
o
n
s
,
”
in
15
th
W
o
r
ki
n
g
C
o
n
f
er
e
n
ce
o
n
R
ever
s
e
E
n
g
i
n
ee
r
i
n
g
,
2
0
1
8
,
doi
:
1
0
.
1
1
0
9
/
W
CR
E
.
2
0
0
8
.
2
5
.
[3
]
A
.
Si
en
o
u
,
A
.
P.
K
ard
u
ck
,
E
.
L
ami
n
e
,
a
n
d
H
.
Pi
n
g
a
u
d
,
“
Bu
s
i
n
es
s
Pro
c
es
s
an
d
Ri
s
k
M
o
d
e
l
s
E
n
r
i
ch
men
t
:
Co
n
s
i
d
e
rat
i
o
n
s
fo
r
Bu
s
i
n
e
s
s
In
t
el
l
i
g
en
ce
,
”
IE
E
E
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
e
-
B
u
s
i
n
es
s
E
n
g
i
n
eer
i
n
g
,
2
0
1
8
,
doi
:
1
0
.
1
1
0
9
/
IC
E
BE
.
2
0
0
8
.
1
2
3
.
[4
]
Y
.
A
.
E
ffen
d
i
an
d
R.
Sarn
o
,
“
Imp
l
emen
t
a
t
i
o
n
o
f
t
h
e
Se
man
t
i
c
W
eb
i
n
Bu
s
i
n
es
s
Pro
ces
s
Mo
d
el
i
n
g
U
s
i
n
g
Pet
r
i
N
et
s
,
”
in
In
t
er
n
a
t
i
o
n
a
l
C
o
n
f
er
e
n
ce
o
n
I
n
f
o
r
m
a
t
i
o
n
a
n
d
Co
m
m
u
n
i
c
a
t
i
o
n
s
Tec
h
n
o
l
o
g
y
,
2
0
1
8
,
v
o
l
.
7
4
1
-
746
,
doi
:
1
0
.
1
1
0
9
/
ICO
IA
C
T
.
2
0
1
8
.
8
3
5
0
7
2
4
.
[5
]
R.
Say
l
am
a
n
d
O
.
K
.
Sa
h
i
n
g
o
z
,
“
Pro
ce
s
s
m
i
n
i
n
g
i
n
b
u
s
i
n
es
s
p
r
o
ces
s
man
ag
eme
n
t
:
Co
n
cep
t
s
an
d
ch
a
l
l
e
n
g
es
,
”
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
E
l
ec
t
r
o
n
i
cs
,
Co
m
p
u
t
er
a
n
d
Co
m
p
u
t
a
t
i
o
n
(ICE
CCO
)
,
2
0
1
3
,
doi
:
1
0
.
1
1
0
9
/
IC
E
CCO
.
2
0
1
3
.
6
7
1
8
2
4
6
.
[6
]
W
.
M.
P.
V
an
d
er
A
al
s
t
,
“
Pro
ces
s
mi
n
i
n
g
:
d
i
s
co
v
er
y
,
co
n
fo
rm
an
ce
an
d
en
h
a
n
ceme
n
t
o
f
b
u
s
i
n
e
s
s
p
ro
ce
s
s
es
,
”
Sp
ri
n
g
er
Sci
e
n
ce
an
d
Bu
s
i
n
e
s
s
Me
d
i
a,
2
0
1
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
6
4
2
-
1
9
3
4
5
-
3.
[7
]
W
.
M.
P.
V
an
d
er
A
al
s
t
an
d
H.
M.
W.
V
erb
eek
,
“
Pro
ces
s
d
i
s
c
o
v
er
y
an
d
co
n
fo
rma
n
ce
ch
eck
i
n
g
u
s
i
n
g
p
as
s
a
g
es
,
”
F
u
n
d
a
m
e
n
t
a
In
f
o
r
m
a
t
i
c
a
e
,
v
o
l
.
1
3
1
,
n
o
.
1
,
p
p
.
1
0
3
-
1
3
8
,
2
0
1
1
.
[8
]
J
.
J
o
e,
T
.
E
mmat
t
y
,
Y
.
Bal
l
al
,
an
d
S.
K
u
l
k
arn
i
,
“
Pro
ces
s
mi
n
i
n
g
fo
r
p
ro
j
ect
man
a
g
emen
t
,
”
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
en
ce
o
n
D
a
t
a
M
i
n
i
n
g
a
n
d
A
d
va
n
ced
Co
m
p
u
t
i
n
g
(S
A
P
I
E
NCE
)
,
2
0
1
6
,
d
o
i
:
1
0
.
1
1
0
9
/
SA
PI
E
N
C
E
.
2
0
1
6
.
7
6
8
4
1
4
2
.
[9
]
A
.
A
.
K
al
en
k
o
v
a,
W
.
M.
P.
V
an
d
er
A
al
s
t
,
I.
A
.
L
o
mazo
v
a,
an
d
V.
A.
Ru
b
i
n
, “
Pro
ces
s
mi
n
i
n
g
u
s
i
n
g
BPMN
:
rel
a
t
i
n
g
ev
en
t
l
o
g
s
a
n
d
p
r
o
ces
s
mo
d
el
s
,
”
S
o
f
t
w
a
r
e
&
S
y
s
t
e
m
s
M
o
d
el
i
n
g
,
2
0
1
7
,
v
o
l
.
16
,
n
o
.
4
,
p
p
.
1
0
1
9
-
1
0
4
8
.
[1
0
]
H.
M.
W
.
V
erb
eek
,
J.
C.
A.
M
.
Bu
i
j
s
,
B.
F
.
V
an
D
o
n
g
e
n
,
an
d
W.
M.
P.
V
an
d
er
A
al
s
t
,
“
Pro
M
6
:
T
h
e
Pro
ce
s
s
Mi
n
i
n
g
T
o
o
l
k
i
,
”
v
o
l
.
6
1
5
,
2
0
1
0
.
[1
1
]
M.
R
.
Peñ
a
an
d
S.
Bay
o
n
a
-
O
r
é
,
“
Pro
ces
s
Mi
n
i
n
g
an
d
A
u
t
o
ma
t
i
c
Pro
ces
s
D
i
s
co
v
ery
,”
7
th
In
t
er
n
a
t
i
o
n
a
l
Co
n
f
e
r
en
ce
o
n
S
o
f
t
wa
r
e
P
r
o
ce
s
s
Im
p
r
o
vem
e
n
t
(CI
M
P
S
)
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
CIMPS.
2
0
1
8
.
8
6
2
5
6
2
1
.
[1
2
]
S.
S
.
Pi
n
t
er
an
d
M.
G
o
l
a
n
i
,
“
D
i
s
co
v
eri
n
g
w
o
r
k
fl
o
w
mo
d
e
l
s
fr
o
m
act
i
v
i
t
i
es
’
l
i
fe
s
p
an
s
,
”
Co
m
p
u
t
e
r
s
i
n
In
d
u
s
t
r
y
,
v
o
l
.
53
,
n
o
.
3
,
p
p
.
2
8
3
-
2
9
6
,
2
0
1
4
.
[1
3
]
Y
.
A
.
E
ffe
n
d
i
an
d
R.
Sarn
o
,
“
M
o
d
e
l
i
n
g
Para
l
l
e
l
Bu
s
i
n
e
s
s
Pr
o
ces
s
U
s
i
n
g
M
o
d
i
fi
e
d
T
i
me
-
b
a
s
ed
A
l
p
h
a
M
i
n
er
,
”
In
t
e
r
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
In
n
o
v
a
t
i
ve
Co
m
p
u
t
i
n
g
,
In
f
o
r
m
a
t
i
o
n
a
n
d
Co
n
t
r
o
l
,
v
o
l
.
14
,
n
o
.
5
,
2
0
1
8
,
doi
:
1
0
.
2
4
5
0
7
/
i
j
i
c
i
c.
1
4
.
0
5
.
1
5
6
5
.
[1
4
]
W
.
M.
P.
V
an
d
er
A
al
s
t
,
“
Bu
s
i
n
es
s
p
r
o
ces
s
man
a
g
eme
n
t
as
t
h
e
“
K
i
l
l
er
A
p
p
”
fo
r
Pe
t
ri
n
et
s
,
”
S
o
f
t
wa
r
e
&
S
y
s
t
em
s
M
o
d
el
i
n
g
,
v
o
l
.
14
,
n
o
.
2
,
p
p
.
6
8
5
-
6
9
1
,
2
0
1
5
.
[1
5
]
W.
M.
P
.
V
an
d
er
A
al
s
t
an
d
C.
C.
St
ah
l
,
“
Mo
d
e
l
i
n
g
b
u
s
i
n
es
s
p
ro
ce
s
s
e
s
:
A
p
e
t
ri
n
et
-
o
ri
e
n
t
e
d
ap
p
r
o
a
ch
,”
MIT
Pres
s
,
2
0
1
1
.
[1
6
]
A.
K.
A
.
d
e
Med
ei
r
o
s
,
B.
F.
v
an
D
o
n
g
e
n
,
W
.
M.
P.
v
an
d
er
A
a
l
s
t
,
an
d
A
.
J.
M.
M.
W
ei
j
t
ers
.
“
Pro
ce
s
s
M
i
n
i
n
g
:
E
x
t
en
d
i
n
g
t
h
e
α
-
a
l
g
o
ri
t
h
m
t
o
Mi
n
e
Sh
o
r
t
L
o
o
p
s
,”
2
0
0
4
.
[1
7
]
N
.
T
ax
,
N
.
Si
d
o
r
o
v
a,
R.
H
aak
ma
,
an
d
W
.
M.
P.
V
an
d
er
A
al
s
t
,
“
Mi
n
i
n
g
p
r
o
ces
s
mo
d
e
l
d
es
cr
i
p
t
i
o
n
s
o
f
d
ai
l
y
l
i
fe
t
h
r
o
u
g
h
ev
e
n
t
a
b
s
t
ract
i
o
n
,
”
P
r
o
ceed
i
n
g
s
o
f
S
A
I
In
t
el
l
i
g
e
n
t
S
y
s
t
e
m
s
C
o
n
f
er
e
n
ce
,
p
p
.
83
-
1
0
4
,
2
0
1
6
.
[1
8
]
R.
A
.
Su
t
ri
s
n
o
w
a
t
i
,
H
.
Bae,
L
.
D
o
n
g
h
a,
an
d
K
.
Mi
n
s
o
o
,
“
Pro
ces
s
Mo
d
el
D
i
s
c
o
v
er
y
b
as
e
d
o
n
A
c
t
i
v
i
t
y
L
i
fes
p
an
,
”
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
Tech
n
o
l
o
g
y
I
n
n
o
va
t
i
o
n
a
n
d
I
n
d
u
s
t
r
i
a
l
M
a
n
a
g
em
e
n
t
,
p
p
.
1
3
7
-
1
5
6
,
2
0
1
4
,
doi
:
1
0
.
1
0
1
6
/
j
.
e
s
w
a.
2
0
1
4
.
0
5
.
0
5
5
.
[1
9
]
Y
.
A
.
E
ffen
d
i
an
d
R.
Sarn
o
,
“
Co
n
fo
rma
n
ce
Ch
eck
i
n
g
E
v
a
l
u
a
t
i
o
n
o
f
Pro
ces
s
D
i
s
c
o
v
er
y
U
s
i
n
g
Mo
d
i
f
i
e
d
A
l
p
h
a+
+
Mi
n
er
A
l
g
o
ri
t
h
m
,
”
In
t
er
n
a
t
i
o
n
a
l
S
e
m
i
n
a
r
o
n
A
p
p
l
i
c
a
t
i
o
n
f
o
r
Tech
n
o
l
o
g
y
o
f
I
n
f
o
r
m
a
t
i
o
n
a
n
d
Co
m
m
u
n
i
ca
t
i
o
n
,
p
p
.
4
3
5
-
4
4
0
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
IS
E
MA
N
T
IC.
2
0
1
8
.
8
5
4
9
7
7
0
.
[2
0
]
A
.
Bu
rat
t
i
n
,
F.
M.
Mag
g
i
,
a
n
d
A
.
Sp
er
d
u
t
i
,
“
C
o
n
f
o
rma
n
ce
ch
ec
k
i
n
g
b
as
e
d
o
n
m
u
l
t
i
-
p
ers
p
ect
i
v
e
d
ec
l
arat
i
v
e
p
r
o
ces
s
mo
d
e
l
s
,”
E
x
p
er
t
S
ys
t
em
s
wi
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
65
,
p
p
.
194
-
2
1
1
,
2
0
1
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
es
w
a.
2
0
1
6
.
0
8
.
0
4
0
.
[2
1
]
W
.
Ch
o
my
a
t
,
an
d
W
.
Premc
h
ai
s
w
a
d
i
,
“
Pr
o
ces
s
m
i
n
i
n
g
o
n
me
d
i
ca
l
t
reat
me
n
t
h
i
s
t
o
ry
u
s
i
n
g
c
o
n
f
o
rman
ce
ch
ec
k
i
n
g
,”
in
14
th
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ce
o
n
ICT
a
n
d
Kn
o
w
l
ed
g
e
E
n
g
i
n
eer
i
n
g
(ICT&KE
)
,
2
0
1
6
,
doi
:
1
0
.
1
1
0
9
/
IC
T
K
E
.
2
0
1
6
.
7
8
0
4
1
0
2
.
[2
2
]
W.
M.
P
.
V
an
d
er
A
al
s
t
,
A
.
A
d
ri
an
s
y
a
h
,
an
d
B.
F.
V
an
D
o
n
g
en
,
“
Cau
s
a
l
N
et
s
:
A
Mo
d
e
l
i
n
g
L
an
g
u
ag
e
T
a
i
l
o
red
t
o
w
ard
s
Pro
ces
s
D
i
s
c
o
v
er
y
,
”
J.
P
.
K.
B
.
Ko
n
i
g
,
CO
N
CU
R
-
Co
n
c
u
r
r
en
cy
Th
e
o
r
y
,
Sp
r
i
n
g
er
Berl
i
n
H
ei
d
el
b
erg
,
pp.
28
-
42
,
2
0
1
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
6
4
2
-
2
3
2
1
7
-
6
_
3
.
[2
3
]
D.
C
.
So
fi
e,
C
.
J
an
,
an
d
P.
G
eert
,
“
Imp
ro
v
i
n
g
t
h
e
q
u
a
l
i
t
y
o
f
t
h
e
H
eu
r
i
s
t
i
c
s
Mi
n
er
i
n
Pr
o
m
6
.
2
.
,”
E
xp
e
r
t
S
ys
t
em
s
wi
t
h
A
p
p
l
i
ca
t
i
o
n
s
,
v
o
l
.
41
,
n
o
.
1
7
,
p
p
.
7
6
7
8
-
7
6
9
0
,
2
0
1
4
.
[2
4
]
L
i
J
,
L
i
u
D
an
d
Y
an
g
B.
“
Pro
ces
s
Mi
n
i
n
g
:
E
x
t
e
n
d
i
n
g
t
h
e
A
l
p
h
a
al
g
o
r
i
t
h
m
t
o
mi
n
e
d
u
p
l
i
cat
e
t
a
s
k
s
i
n
p
r
o
ces
s
l
o
g
s
,
”
A
d
v
an
ce
s
i
n
W
eb
an
d
N
e
t
w
o
rk
T
e
ch
n
o
l
o
g
i
e
s
,
an
d
In
f
o
r
mat
i
o
n
Man
a
g
emen
t
,
p
p
.
3
9
6
-
4
0
7
,
2
0
0
7
.
[2
5
]
A
.
Ro
g
g
e
-
S
o
l
t
i
,
R.
S
.
Man
s
,
W
.
M.
P
.
V
an
d
er
A
al
s
t
,
an
d
M.
W
es
k
e
,
“
Rep
ai
r
i
n
g
ev
en
t
l
o
g
s
u
s
i
n
g
t
i
me
d
p
ro
ces
s
mo
d
e
l
s
,”
O
T
M
Co
n
f
ed
e
r
a
t
ed
I
n
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
e
n
ces
,
2
0
1
3
.
[2
6
]
W
.
L
i
j
i
e
a
n
d
W
.
J
i
a
n
mi
n
,
“
D
et
ect
i
n
g
i
mp
l
i
c
i
t
D
ep
e
n
d
e
n
ci
es
Bet
w
ee
n
T
as
k
s
fr
o
m
E
v
en
t
L
o
g
s
,”
Sch
o
o
l
o
f
So
f
t
w
are
,
T
s
i
n
g
h
u
a
U
n
i
v
er
s
i
t
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.