Co
m
pu
t
er
Science
a
nd
I
nfo
r
m
a
t
io
n T
ec
hn
o
lo
g
ies
Vo
l.
7
,
No
.
3
,
No
v
em
b
er
20
26
,
p
p
.
3
3
7
~
3
4
5
I
SS
N:
2722
-
3
2
2
1
,
DOI
:
1
0
.
1
1
5
9
1
/csi
t
.
v
7
i
3
.
p
3
3
7
-
3
4
5
337
J
o
ur
na
l
ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
p
r
ime.
co
m/in
d
ex
.
p
h
p
/csi
t
Risk
-
int
eg
ra
ted
c
o
ntractor a
llo
ca
ti
o
n in Zimba
bwe
’
s
t
imber
v
a
lue
c
ha
in
T
a
v
eng
wa
No
rm
a
n,
B
ria
n
M
up
ini
S
c
h
o
o
l
o
f
I
n
f
o
r
ma
t
i
o
n
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
H
a
r
a
r
e
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
,
H
a
r
a
r
e
,
Z
i
m
b
a
b
w
e
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
May
19
,
2
0
2
6
R
ev
is
ed
J
u
n
10
,
2
0
2
6
Acc
ep
ted
J
u
l
13
,
2
0
2
6
In
Zi
m
b
a
b
we
,
t
h
e
c
o
m
m
e
rc
ial
fo
re
stry
in
d
u
str
y
is
re
li
a
n
t
o
n
a
larg
e
p
ro
p
o
rti
o
n
o
f
o
u
ts
o
u
rc
e
d
h
a
r
v
e
st
a
n
d
m
il
li
n
g
c
o
n
trac
to
rs
wh
e
re
a
s
e
x
isti
n
g
e
n
terp
rise
re
so
u
rc
e
p
lan
n
in
g
(E
RP
)
sy
ste
m
s
re
c
o
rd
c
o
m
p
lete
d
t
ra
n
sa
c
ti
o
n
s
in
ste
a
d
o
f
p
re
d
ictin
g
c
o
n
trac
t
o
r
fa
il
u
re
b
e
fo
re
a
ss
ig
n
i
n
g
a
c
ti
v
it
ies
.
We
in
tro
d
u
c
e
th
e
d
y
n
a
m
ic
re
so
u
rc
e
a
ll
o
c
a
ti
o
n
fra
m
e
wo
rk
(DRA
F
)
,
a
risk
-
in
teg
ra
ted
d
e
c
isio
n
-
s
u
p
p
o
rt
fra
m
e
wo
rk
th
a
t
su
p
p
lem
e
n
ts
th
e
p
r
o
b
a
b
il
it
y
o
f
c
o
n
trac
to
r
fa
il
u
re
d
e
riv
e
d
fr
o
m
X
G
Bo
o
st
to
a
n
o
n
-
d
o
m
in
a
ted
so
rti
n
g
g
e
n
e
ti
c
a
lg
o
rit
h
m
II
(NSG
A
-
II)
m
u
lt
i
-
o
b
jec
ti
v
e
o
p
t
imiz
e
r.
It
d
i
v
id
e
s
c
o
n
trac
to
r
-
b
lo
c
k
-
m
il
l
c
o
m
b
i
n
a
ti
o
n
s,
m
in
imi
z
e
s
c
o
st
a
n
d
e
x
p
e
c
ted
d
e
lay
,
a
n
d
m
a
x
imiz
e
s
risk
-
a
d
ju
ste
d
ti
m
b
e
r
re
c
o
v
e
ry
a
n
d
o
p
e
ra
ti
o
n
a
l
re
li
a
b
i
li
ty
.
We
c
re
a
ted
th
e
so
lu
ti
o
n
o
n
a
n
8
2
8
,
7
8
9
-
re
c
o
rd
v
irt
u
a
l
ERP
d
a
tas
e
t
a
n
d
c
a
li
b
ra
ted
it
t
o
M
a
n
ica
lan
d
f
o
re
stry
c
o
n
d
it
i
o
n
s
a
n
d
tes
ted
wit
h
sta
ti
stica
l,
h
e
u
risti
c
a
n
d
r
isk
-
fre
e
o
p
ti
m
iza
ti
o
n
b
a
se
li
n
e
s.
Ex
tr
e
m
e
g
ra
d
ien
t
b
o
o
sti
n
g
(
XG
Bo
o
st
)
o
b
tain
e
d
a
h
o
ld
o
u
t
re
c
e
iv
e
r
o
p
e
ra
ti
n
g
c
h
a
r
a
c
teristic
a
re
a
u
n
d
e
r
th
e
c
u
rv
e
(R
OC
-
AU
C)
o
f
0
.
9
6
5
,
re
c
a
ll
o
f
0
.
9
9
9
,
a
n
d
F
1
o
f
0
.
8
6
7
,
sh
o
wi
n
g
a
n
imp
ro
v
e
m
e
n
t
o
f
0
.
3
6
5
o
v
e
r
l
o
g
isti
c
re
g
re
ss
io
n
.
Th
e
o
p
ti
m
ize
r
d
e
v
e
lo
p
e
d
6
4
c
o
m
p
l
e
te
P
a
re
to
so
lu
ti
o
n
s
to
b
a
lan
c
e
d
a
n
d
h
ig
h
-
re
c
o
v
e
ry
sc
e
n
a
rio
s
a
n
d
u
n
c
o
v
e
re
d
a
c
o
n
stra
in
t
-
fe
a
sib
il
it
y
b
o
u
n
d
a
ry
fo
r
a
m
o
re
c
o
n
se
rv
a
ti
v
e
lo
w
-
ris
k
sc
e
n
a
rio
.
Th
e
sa
ti
sfa
c
ti
o
n
sc
o
re
o
f
th
e
3
0
-
p
ra
c
ti
ti
o
n
e
r
sta
k
e
h
o
ld
e
r
a
ss
e
ss
m
e
n
t
wa
s 4
.
1
9
o
u
t
o
f
5
.
0
.
Th
e
re
su
lt
s
d
e
m
o
n
str
a
te
th
a
t
e
m
b
e
d
d
in
g
p
re
d
icti
v
e
risk
in
to
a
n
o
p
ti
m
iza
ti
o
n
o
b
jec
ti
v
e
c
a
n
o
p
ti
m
i
z
e
fo
re
stry
a
ll
o
c
a
ti
o
n
d
e
c
isio
n
s
a
n
d
s
u
g
g
e
s
t
th
a
t
so
m
e
re
a
l
ERP
v
a
li
d
a
ti
o
n
i
s
re
q
u
ire
d
b
e
fo
re
s
u
c
h
m
e
a
su
re
s
will
b
e
b
ro
a
d
l
y
imp
lem
e
n
ted
.
K
ey
w
o
r
d
s
:
C
o
n
tr
ac
to
r
r
is
k
Dec
is
io
n
s
u
p
p
o
r
t sy
s
tem
s
Fo
r
estry
s
u
p
p
ly
c
h
ain
Mu
lti
-
o
b
jectiv
e
o
p
tim
izatio
n
NSGA
-
II
XGBo
o
s
t
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
T
av
en
g
wa
No
r
m
an
Sch
o
o
l o
f
I
n
f
o
r
m
atio
n
Scien
ce
an
d
T
ec
h
n
o
lo
g
y
,
Har
a
r
e
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
Gan
g
es
R
o
ad
,
B
elv
ed
er
e,
Har
ar
e,
Z
im
b
ab
we
E
m
ail: h
2
4
0
7
4
2
c@
h
it.a
c.
zw
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
c
o
m
m
e
r
c
ial
f
o
r
est
r
y
in
d
u
s
tr
y
i
n
Z
i
m
b
a
b
we
is
d
o
m
i
n
at
e
d
b
y
v
er
tic
all
y
i
n
t
eg
r
a
te
d
f
ir
m
s
l
ik
e
Alli
e
d
T
im
b
er
s
Z
i
m
b
a
b
we
,
B
o
r
d
e
r
T
im
b
er
s
L
i
m
it
ed
,
T
h
e
W
at
tle
C
o
m
p
a
n
y
,
a
n
d
M
u
t
ar
e
B
o
ar
d
a
n
d
Pa
p
er
Mil
ls
,
wh
i
ch
c
o
ll
ec
t
iv
el
y
m
a
n
a
g
e
o
v
er
7
0
,
0
0
0
h
a
o
f
p
l
a
n
ta
ti
o
n
in
M
an
ic
ala
n
d
[
1
]
,
[
2
]
.
T
h
e
im
p
o
r
ta
n
ce
o
f
f
o
r
est
r
y
t
o
r
u
r
al
em
p
l
o
y
m
en
t,
t
h
e
s
u
p
p
l
y
o
f
i
n
d
u
s
tr
ial
r
aw
m
ate
r
i
als
,
a
n
d
e
x
p
o
r
t
-
o
r
i
en
te
d
v
a
lu
e
ch
ai
n
s
e
n
d
u
r
es
[
3
]
.
B
u
t
o
u
ts
o
u
r
ci
n
g
h
as
t
r
a
n
s
f
e
r
r
e
d
o
p
er
ati
o
n
al
r
is
k
i
n
t
o
c
o
n
t
r
ac
to
r
n
etw
o
r
k
s
:
u
p
t
o
7
0
%
o
f
m
illi
n
g
p
r
o
d
u
cti
o
n
is
n
o
w
g
e
n
er
ate
d
b
y
s
m
all
-
to
m
ed
i
u
m
-
s
iz
ed
c
o
n
t
r
ac
t
o
r
s
[
4
]
,
r
eq
u
i
r
i
n
g
co
n
t
r
a
ct
o
r
a
ll
o
c
ati
o
n
to
b
e
a
h
i
g
h
-
s
ta
k
es
d
e
cisi
o
n
t
h
at
m
u
s
t
b
al
an
ce
c
o
s
t,
ca
p
a
cit
y
,
te
r
r
ai
n
,
r
ai
n
f
al
l,
a
n
d
r
e
lia
b
i
lit
y
b
ef
o
r
e
w
o
r
k
b
e
g
i
n
s
[
5
]
,
[
6
]
.
Prio
r
liter
atu
r
e
s
h
o
ws
two
r
e
lev
an
t
b
u
t
u
s
u
ally
s
ep
ar
ate
c
ap
ab
ilit
ies:
en
s
em
b
le
m
ac
h
i
n
e
lear
n
in
g
(
ML
)
ca
n
p
r
ed
ict
co
n
tr
ac
to
r
o
r
s
u
p
p
lier
u
n
d
er
p
e
r
f
o
r
m
an
ce
in
s
tr
u
ctu
r
ed
o
p
er
atio
n
al
en
v
ir
o
n
m
e
n
ts
[
7
]
–
[
9
]
,
an
d
ev
o
l
u
tio
n
ar
y
m
u
lti
-
o
b
jecti
v
e
o
p
tim
izatio
n
(
MO
O)
,
esp
e
cially
non
-
d
o
m
in
ated
s
o
r
tin
g
g
e
n
etic
alg
o
r
ith
m
I
I
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
7
,
No
.
3
,
No
v
em
b
er
20
26
:
3
3
7
-
345
338
(
NSGA
-
II
)
,
ca
n
p
r
o
d
u
ce
Par
eto
-
ef
f
icien
t
allo
ca
tio
n
s
tr
ateg
ies
in
f
o
r
estry
an
d
s
u
p
p
l
y
ch
ain
ap
p
licatio
n
s
[
1
0
]
,
[
1
1
]
.
Su
p
p
l
y
ch
ai
n
r
is
k
th
eo
r
y
also
tr
ea
ts
d
is
r
u
p
tio
n
r
is
k
as
a
d
y
n
am
ic
p
lan
n
i
n
g
v
a
r
iab
le
r
ath
e
r
th
a
n
a
s
tatic
af
ter
-
th
e
-
f
ac
t
ex
p
lan
a
tio
n
[
1
2
]
,
[
1
3
]
.
R
ec
en
t
s
tu
d
ies
h
av
e
s
h
o
wn
th
at
ML
tech
n
iq
u
es
ca
n
en
h
an
ce
f
o
r
estry
h
a
r
v
esti
n
g
o
p
er
atio
n
s
b
y
s
u
p
p
o
r
tin
g
p
r
ed
ictiv
e
a
n
aly
tics
an
d
o
p
e
r
atio
n
al
d
ec
i
s
io
n
-
m
ak
in
g
[
1
4
]
.
Ho
wev
er
,
ex
is
tin
g
r
esear
ch
h
as
p
r
im
ar
ily
f
o
cu
s
ed
o
n
s
p
ec
if
ic
o
p
er
atio
n
al
task
s
r
ath
er
th
an
in
teg
r
atin
g
p
r
ed
ictiv
e
r
is
k
ass
ess
m
en
t,
MO
O
,
an
d
i
n
ter
ac
tiv
e
d
ec
is
io
n
s
u
p
p
o
r
t
in
to
a
u
n
if
ied
d
ec
is
io
n
-
in
tellig
en
ce
f
r
am
ewo
r
k
tailo
r
ed
f
o
r
f
o
r
estry
ap
p
licatio
n
s
in
So
u
t
h
er
n
Af
r
ica.
Ho
wev
er
,
th
ese
ca
p
a
b
ilit
ies
h
av
e
n
o
t
b
ee
n
m
er
g
ed
in
t
o
a
f
o
r
estry
-
ad
a
p
te
d
d
ec
is
io
n
-
in
tellig
en
ce
a
r
tef
ac
t f
o
r
So
u
th
e
r
n
Af
r
ica.
T
h
is
p
ap
er
f
ills
th
at
g
ap
b
y
d
ev
elo
p
in
g
th
e
d
y
n
am
ic
r
eso
u
r
ce
allo
ca
tio
n
f
r
am
ew
o
r
k
(
DR
AF)
,
with
th
r
ee
m
ain
co
m
p
o
n
en
ts
:
i)
an
ex
tr
em
e
g
r
ad
ie
n
t
b
o
o
s
tin
g
(
XGBo
o
s
t
)
en
s
em
b
le
clas
s
i
f
ier
th
at
tr
an
s
lates
p
r
e
-
allo
ca
tio
n
o
p
e
r
atio
n
al
ch
ar
ac
ter
is
tics
in
to
ca
lib
r
ated
co
n
tr
ac
to
r
r
is
k
p
r
o
b
ab
ilit
ies;
ii)
an
NSGA
-
II
o
p
tim
iz
er
wh
ich
em
b
ed
s
th
ese
p
r
o
b
ab
ilit
ies
as
f
ir
s
t
-
clas
s
o
b
j
ec
tiv
e
in
p
u
ts
r
ath
er
th
an
as
p
o
s
t
-
h
o
c
co
n
s
tr
ain
ts
;
an
d
iii)
an
in
ter
ac
tiv
e
d
ec
is
io
n
-
s
u
p
p
o
r
t
d
ash
b
o
ar
d
b
u
ilt
i
n
D
ash
th
at
d
eliv
er
s
Par
eto
-
o
p
tim
al
allo
ca
tio
n
g
u
id
elin
es
to
n
o
n
-
tec
h
n
ical
estate
m
an
ag
er
s
.
T
h
e
r
esear
ch
f
o
llo
ws
th
e
d
esig
n
s
cien
ce
r
esear
ch
(
DSR
)
m
eth
o
d
o
l
o
g
y
[
1
5
]
,
[
1
6
]
,
in
w
h
ich
th
e
f
r
am
ewo
r
k
is
b
o
th
t
h
e
th
eo
r
etica
l
co
n
tr
ib
u
tio
n
an
d
th
e
m
ain
r
esear
ch
o
u
tp
u
t.
T
h
e
s
p
ec
if
ic
co
n
tr
ib
u
tio
n
s
ar
e
th
r
ee
f
o
l
d
.
First,
th
e
s
tu
d
y
p
r
o
p
o
s
es
a
r
is
k
-
in
teg
r
ate
d
o
p
tim
izatio
n
ar
ch
itectu
r
e
in
wh
ich
ML
-
d
er
iv
ed
r
is
k
p
r
o
b
ab
ilit
ies
ar
e
d
ir
ec
tly
em
b
ed
d
ed
in
e
v
o
l
u
tio
n
ar
y
o
b
jectiv
e
f
u
n
ctio
n
s
.
Seco
n
d
,
it
c
o
n
s
tr
u
cts
an
d
v
alid
ates
a
r
ep
r
o
d
u
c
ib
le
8
2
8
,
7
8
9
-
r
ec
o
r
d
v
ir
tu
al
e
n
ter
p
r
is
e
r
eso
u
r
ce
p
lan
n
in
g
(
E
R
P
)
d
ataset
ca
lib
r
ated
to
Z
im
b
a
b
wea
n
f
o
r
estr
y
co
n
d
itio
n
s
.
T
h
ir
d
,
it
ev
alu
ates
DR
A
F
ag
ain
s
t
s
tatis
t
ical,
h
eu
r
is
tic,
an
d
r
is
k
-
f
r
ee
o
p
tim
izatio
n
b
aselin
es,
an
d
a
g
ain
s
t
f
ee
d
b
ac
k
f
r
o
m
3
0
f
o
r
estry
-
s
ec
to
r
p
r
ac
titi
o
n
er
s
.
T
h
e
r
est
o
f
th
is
p
ap
er
is
f
o
r
m
atted
as
:
s
ec
tio
n
2
p
r
esen
ts
th
e
d
esig
n
ju
s
tific
atio
n
o
f
th
e
p
r
o
p
o
s
e
d
f
r
am
ewo
r
k
.
Sectio
n
3
d
escr
ib
es
th
e
m
eth
o
d
(
f
r
am
ewo
r
k
ar
ch
itectu
r
e,
p
r
o
b
le
m
f
o
r
m
u
latio
n
,
d
ataset
co
n
s
tr
u
ctio
n
,
a
n
d
ex
p
e
r
im
en
t
al
s
etu
p
)
.
R
esu
l
ts
an
d
d
is
cu
s
s
io
n
ar
e
r
ep
o
r
ted
in
s
ec
tio
n
4
an
d
s
ec
tio
n
5
co
n
clu
d
es.
2.
P
RO
P
O
SE
D
M
E
T
H
O
D
D
E
SI
G
N
RA
T
I
O
NA
L
E
2
.
1
.
P
re
dict
iv
e
risk
ra
t
i
o
na
le
I
n
ch
allen
g
in
g
wo
r
k
in
g
en
v
ir
o
n
m
en
ts
,
en
s
em
b
le
ML
h
as
em
er
g
ed
as
an
im
p
o
r
tan
t
tech
n
iq
u
e
f
o
r
p
r
ed
ictin
g
co
n
t
r
ac
to
r
an
d
s
u
p
p
lier
u
n
d
er
p
er
f
o
r
m
an
ce
.
B
ec
au
s
e
th
ey
ca
p
tu
r
e
n
o
n
-
lin
ea
r
i
n
ter
ac
tio
n
s
with
o
u
t
th
e
r
ep
r
esen
tatio
n
-
lear
n
in
g
o
v
er
h
ea
d
ca
u
s
ed
b
y
d
ee
p
n
eu
r
al
m
o
d
els,
r
an
d
o
m
f
o
r
ests
[
1
7
]
an
d
XGBo
o
s
t
[
1
8
]
ar
e
im
p
o
r
tan
t
b
aselin
es
f
o
r
ta
b
u
lar
r
is
k
m
o
d
ellin
g
.
R
esear
ch
s
tu
d
ies,
m
ain
ly
with
r
esp
ec
t
to
co
n
tr
ac
to
r
-
r
is
k
an
d
s
u
p
p
l
y
-
ch
ain
a
n
d
h
y
b
r
id
XGBo
o
s
t
-
lo
g
is
tic
ap
p
r
o
a
ch
es,
h
av
e
r
e
v
ea
led
h
i
g
h
r
ec
eiv
er
o
p
e
r
atin
g
ch
ar
ac
ter
is
tic
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
R
OC
-
AUC
)
p
er
f
o
r
m
a
n
ce
[
7
]
,
[
8
]
,
[
1
9
]
,
wh
ile
b
r
o
a
d
er
tab
u
lar
-
lear
n
in
g
co
m
p
ar
is
o
n
s
h
a
v
e
n
e
v
er
th
ele
s
s
d
em
o
n
s
tr
ated
th
e
ab
ilit
y
f
o
r
tr
ee
-
b
ased
m
o
d
els
to
ac
h
i
ev
e
s
u
p
er
io
r
r
esu
lts
o
v
er
d
ee
p
ar
c
h
itectu
r
es o
n
s
tr
u
ctu
r
ed
d
at
a
[
9
]
,
[
2
0
]
.
2
.
2
.
O
pti
m
iza
t
io
n
r
a
t
i
o
na
le
MO
O
s
ea
r
ch
es
f
o
r
n
o
n
-
d
o
m
in
ated
alter
n
ativ
es
wh
en
co
s
t,
r
is
k
,
s
er
v
ice
lev
el,
an
d
r
eso
u
r
ce
co
n
s
tr
ain
ts
ca
n
n
o
t
b
e
co
llap
s
e
d
in
to
o
n
e
s
co
r
e
with
o
u
t
lo
s
in
g
m
an
a
g
er
ial
m
ea
n
in
g
[
2
1
]
.
R
ec
en
t
ev
o
l
u
tio
n
ar
y
-
co
m
p
u
tatio
n
s
tu
d
ies
also
h
ig
h
lig
h
t
m
u
lti
-
o
b
jectiv
e
ev
o
lu
tio
n
ar
y
alg
o
r
ith
m
b
ased
o
n
d
ec
o
m
p
o
s
itio
n
(
MO
E
A/D
)
,
s
tr
en
g
th
Par
eto
ev
o
lu
tio
n
ar
y
alg
o
r
ith
m
(
SP
E
A
)
2
,
an
d
h
y
b
r
i
d
d
esig
n
s
f
o
r
s
u
p
p
ly
c
h
ain
o
p
tim
izatio
n
[
2
2
]
.
W
ith
in
th
is
f
am
ily
,
NSGA
-
I
I
[
1
0
]
r
em
ai
n
s
wid
ely
v
alid
ated
f
o
r
m
u
lti
-
o
b
jectiv
e
r
eso
u
r
ce
-
allo
ca
tio
n
p
r
o
b
lem
s
,
wh
ile
f
o
r
estry
an
d
co
n
s
tr
u
ctio
n
s
tu
d
ie
s
s
h
o
w
th
at
o
p
tim
izatio
n
ca
n
im
p
r
o
v
e
p
lan
n
in
g
q
u
ality
u
n
d
er
p
r
o
d
u
ctiv
ity
,
co
s
t,
an
d
r
eso
u
r
ce
co
n
s
tr
ain
ts
[
1
1
]
,
[
2
3
]
,
[
2
4
]
.
T
h
e
ar
ch
itectu
r
al
g
ap
is
th
a
t
m
o
s
t
f
o
r
estry
a
n
d
s
u
p
p
ly
-
ch
ain
MO
O
im
p
lem
en
tatio
n
s
ass
u
m
e
d
eter
m
in
is
tic
co
n
tr
ac
to
r
p
er
f
o
r
m
an
ce
.
R
is
k
is
o
m
itted
,
tr
ea
ted
as
a
f
ea
s
ib
ilit
y
f
ilter
,
o
r
r
ev
iewe
d
af
ter
o
p
ti
m
izatio
n
.
DR
AF
ad
d
r
ess
es
th
is
g
ap
b
y
in
s
er
tin
g
p
r
o
b
ab
ilis
tic
ML
r
is
k
s
co
r
es
in
to
th
e
o
p
tim
izer
’
s
o
b
jectiv
e
f
u
n
ctio
n
s
as
d
y
n
am
ic
weig
h
ts
,
lin
k
in
g
p
r
ed
ictiv
e
m
o
d
ellin
g
,
o
p
tim
izatio
n
,
an
d
d
ec
is
io
n
s
u
p
p
o
r
t
in
th
e
s
am
e
allo
ca
tio
n
wo
r
k
f
lo
w
[
2
5
]
–
[
2
9
]
.
3.
M
E
T
H
O
D
3
.
1
.
Co
ncept
ua
l
a
rc
hite
ct
ur
e
DR
AF
i
s
o
p
er
atio
n
alis
ed
as
a
th
r
ee
-
lay
er
d
ec
is
io
n
-
in
tellig
e
n
ce
p
ip
elin
e.
T
h
e
in
p
u
t
lay
er
co
n
s
u
m
es
p
r
e
-
allo
ca
tio
n
d
ec
is
io
n
-
tim
e
v
ar
iab
les
ac
r
o
s
s
th
e
h
ar
v
esti
n
g
,
tr
an
s
p
o
r
tatio
n
,
a
n
d
m
illi
n
g
d
o
m
ain
s
,
wh
ich
ar
e
d
ef
in
ed
at
th
e
tim
e
o
f
co
n
t
r
ac
to
r
s
elec
tio
n
an
d
ar
e
n
o
t
leak
y
af
ter
th
e
co
n
tr
ac
t
h
as
b
ee
n
f
in
alize
d
.
T
h
e
p
r
ed
ictiv
e
r
is
k
lay
er
im
p
lem
en
ts
a
tr
ain
ed
XGBo
o
s
t e
n
s
em
b
le
class
if
ier
,
g
en
er
atin
g
ca
lib
r
ated
p
r
o
b
ab
ilit
y
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
R
is
k
-
in
teg
r
a
ted
co
n
tr
a
cto
r
a
ll
o
ca
tio
n
in
Zimb
a
b
w
e'
s
timb
er
va
lu
e
ch
a
i
n
(
Ta
ve
n
g
w
a
N
o
r
ma
n
)
339
esti
m
ates
p
(
u
n
d
er
p
er
f
o
r
m
an
c
e
|
f
ea
tu
r
es)
wh
ich
ar
e
in
teg
r
ated
d
ir
ec
tly
in
to
th
e
o
b
jectiv
e
f
u
n
ctio
n
s
o
f
th
e
MO
O
lay
er
.
T
h
e
d
ec
is
io
n
s
u
p
p
o
r
t
lay
er
p
r
o
v
id
es
th
e
P
ar
eto
-
o
p
tim
al
allo
ca
tio
n
s
ets
in
an
in
ter
ac
tiv
e
d
ash
b
o
ar
d
f
o
r
estate
m
an
ag
er
s
to
as
s
ess
tr
ad
e
-
o
f
f
s
an
d
u
p
d
ate
o
p
er
atio
n
al
d
ec
is
io
n
s
in
r
ea
l
tim
e.
I
n
p
r
ac
tical
ter
m
s
,
th
is
lay
er
is
an
alo
g
o
u
s
to
a
lig
h
tweig
h
t
d
is
tr
ib
u
te
d
d
ec
is
io
n
-
s
u
p
p
o
r
t
ar
c
h
itectu
r
e:
m
o
d
el
s
co
r
in
g
,
ev
o
lu
tio
n
ar
y
o
p
tim
izatio
n
,
a
n
d
d
ash
b
o
ar
d
in
ter
ac
tio
n
ca
n
b
e
d
ec
o
u
p
led
in
t
o
s
er
v
ic
es
th
at
ex
ch
an
g
e
s
tan
d
ar
d
ized
E
R
P
d
ata
o
b
jects
an
d
ca
n
th
er
ef
o
r
e
b
e
d
ep
l
o
y
e
d
o
n
l
o
ca
l
s
er
v
er
s
o
r
clo
u
d
in
f
r
astru
ctu
r
e
as
estate
d
ata
m
atu
r
ity
im
p
r
o
v
es
[
2
8
],
[
29
]
.
T
h
e
ar
ch
itectu
r
al
i
n
n
o
v
a
tio
n
is
th
e
d
ir
ec
t
co
n
n
ec
tio
n
b
etwe
en
p
r
e
d
ictiv
e
an
d
o
p
tim
izatio
n
lay
er
s
.
R
ath
er
th
an
u
s
in
g
r
is
k
as
a
c
o
n
s
tr
ain
t
th
r
esh
o
ld
,
th
e
tr
ad
itio
n
al
p
r
ac
tice
in
f
o
r
estry
MO
O
,
p
r
ed
icted
r
is
k
is
m
u
ltip
lied
as
a
d
is
co
u
n
t
o
n
e
x
p
ec
t
ed
tim
b
er
r
ec
o
v
er
y
u
n
d
e
r
th
e
o
p
tim
izer
o
b
jectiv
e
f
u
n
ctio
n
in
DR
AF.
T
h
is
m
a
k
es
s
u
r
e
th
e
ev
o
lu
tio
n
ar
y
s
ea
r
ch
ac
tiv
ely
f
av
o
r
s
allo
ca
tio
n
s
in
wh
ich
h
ig
h
er
ex
p
ec
ted
y
ield
im
p
lies
a
lo
wer
p
r
o
b
a
b
ilit
y
o
f
co
n
t
r
ac
to
r
f
ailu
r
e
,
m
ea
n
in
g
m
o
r
e
r
o
b
u
s
t,
r
ath
er
th
an
s
im
p
ly
m
ath
em
atica
lly
o
p
tim
al,
s
o
lu
ti
o
n
s
.
3
.
2
.
P
re
dict
iv
e
risk
mo
del f
o
rm
ula
t
io
n
L
et
ea
ch
ca
n
d
i
d
ate
allo
ca
tio
n
b
e
d
ef
in
e
d
b
y
th
e
tu
p
le
(
b
,
c,
r
,
m
)
,
w
h
er
e
b
d
en
o
tes
a
tim
b
er
b
lo
ck
,
c
a
co
n
tr
ac
to
r
,
r
a
h
au
lag
e
r
o
u
te,
an
d
m
a
m
ill.
L
et
x
=(
x
1
,
x
2
,
…,
x
4
6
)
d
e
n
o
te
th
e
4
6
-
d
i
m
en
s
io
n
al
leak
ag
e
-
s
af
e
f
ea
tu
r
e
v
ec
to
r
ch
ar
ac
te
r
izin
g
th
e
allo
ca
tio
n
.
T
h
e
p
r
ed
ict
iv
e
m
o
d
el
esti
m
ates:
(
,
,
,
)
=
(
_
_
=
1
|
)
∈
[
0
,
1
]
wh
er
e
ρ
is
th
e
XG
B
o
o
s
t
-
esti
m
ated
r
is
k
p
r
o
b
ab
ilit
y
.
XGBo
o
s
t
was
ch
o
s
en
o
v
er
r
an
d
o
m
f
o
r
est
an
d
lo
g
is
tic
b
aselin
es
d
u
e
to
b
etter
v
alid
atio
n
p
r
ec
is
io
n
-
r
ec
all
ar
ea
u
n
d
e
r
th
e
cu
r
v
e
(
PR
-
AUC
)
.
S
y
n
th
etic
m
in
o
r
ity
o
v
er
-
s
am
p
lin
g
tech
n
iq
u
e
(
SMOT
E
)
o
v
er
s
am
p
lin
g
o
n
tr
ain
i
n
g
p
ip
elin
e
h
elp
ed
to
m
itig
ate
m
o
d
er
ate
class
im
b
alan
ce
(
p
o
s
itiv
e
class
r
ate:
3
0
.
9
3
%).
τ
=0
.
2
4
is
th
e
d
ec
is
io
n
th
r
esh
o
ld
o
p
tim
ized
d
u
r
in
g
v
alid
atio
n
s
p
lit
to
m
ax
im
ize
th
e
F1
-
s
co
r
e
wh
ile
p
r
eser
v
in
g
h
ig
h
r
ec
all,
wh
ich
is
in
d
icativ
e
o
f
t
h
e
asy
m
m
etr
ic
m
is
class
if
icatio
n
co
s
t
s
tr
u
ctu
r
e
th
at
in
d
icate
s
th
at
f
alse
n
eg
ativ
es
(
FN)
(
u
n
d
etec
ted
h
ig
h
-
r
is
k
allo
ca
tio
n
s
)
ar
e
o
p
er
atio
n
ally
m
o
r
e
co
s
tly
th
an
f
alse p
o
s
itiv
es
(
FP
)
.
3
.
3
.
M
ulti
-
o
bje
ct
iv
e
o
ptim
iz
a
t
io
n f
o
r
m
ula
t
io
n
L
et
A=
{a
1
,
a2
,
…,
an
}
d
en
o
te
th
e
f
ea
s
ib
ilit
y
-
f
ilter
ed
ca
n
d
id
ate
ass
ig
n
m
en
t
p
o
o
l.
E
ac
h
c
an
d
id
ate
a1
is
ch
ar
ac
ter
ized
b
y
d
eter
m
in
is
tic
attr
ib
u
tes
(
co
s
t
C
i,
b
ase
v
o
lu
m
e
Vi,
b
ase
d
u
r
atio
n
Di,
r
o
u
te
f
it
R
i,
an
d
m
ill
f
ea
s
ib
ilit
y
Mi)
an
d
th
e
ML
-
d
er
iv
ed
r
is
k
p
r
o
b
a
b
ilit
y
ρ
i.
DR
AF
s
im
u
ltan
eo
u
s
ly
m
in
im
iz
es/ma
x
im
izes
f
o
u
r
o
b
jectiv
es:
−
1
=
i
)
t
o
tal
o
p
er
atio
n
al
co
s
t
−
2
=
(
1
−
)
ii
)
r
is
k
-
ad
ju
s
ted
tim
b
er
r
ec
o
v
e
r
y
−
3
=
·
iii
)
e
x
p
ec
ted
d
ela
y
−
4
=
(
1
−
)
·
·
iv
)
o
p
e
r
atio
n
al
r
eliab
ilit
y
Su
b
ject
to
:
co
n
tr
ac
to
r
ca
p
ac
it
y
lim
its
,
m
ill
th
r
o
u
g
h
p
u
t
c
o
n
s
tr
ain
ts
,
ter
r
ain
s
lo
p
e
f
ea
s
ib
ilit
y
(
s
lo
p
e_
f
it
≥θ
s
)
,
r
o
u
te
ac
ce
s
s
ib
ilit
y
(
r
o
u
t
e_
f
it≥
θ
r
)
,
an
d
av
er
a
g
e
r
is
k
ce
i
lin
g
(
m
ea
n
ρ≤
θ
̄
ρ
)
.
F2
i
s
th
e
m
ain
in
n
o
v
atio
n
:
th
e
(
1
−ρi
)
d
is
co
u
n
t
c
o
ef
f
icien
t
co
n
tin
u
o
u
s
ly
p
en
alize
s
s
o
lu
ti
o
n
s
th
at
ass
ig
n
h
ig
h
-
y
ield
b
lo
ck
s
to
h
ig
h
-
r
is
k
co
n
tr
ac
to
r
s
,
d
r
iv
in
g
th
e
ev
o
lu
tio
n
ar
y
s
ea
r
c
h
to
war
d
allo
ca
t
io
n
s
th
at
ar
e
b
o
th
ec
o
n
o
m
ically
p
r
o
d
u
ctiv
e
an
d
o
p
er
atio
n
ally
r
esil
ien
t.
NSGA
-
I
I
was
im
p
lem
e
n
ted
u
s
in
g
Py
m
o
o
(
v
er
s
io
n
0
.
6
.
x
)
with
s
im
u
lated
b
in
ar
y
cr
o
s
s
o
v
er
(
SB
X
)
,
p
o
ly
n
o
m
ial
m
u
tatio
n
,
an
d
r
o
u
n
d
in
g
r
ep
air
o
p
er
ato
r
s
.
T
h
r
ee
p
o
licy
s
ce
n
ar
io
s
,
b
alan
ce
d
,
lo
w
-
r
is
k
,
an
d
h
ig
h
-
r
ec
o
v
e
r
y
,
wer
e
co
n
f
ig
u
r
ed
with
d
is
tin
ct
co
n
s
tr
ain
t p
a
r
am
eter
s
an
d
s
h
o
r
tlis
t w
eig
h
ts
.
3
.
4
.
Virt
ua
l ERP
da
t
a
s
et
co
n
s
t
ruct
io
n
B
ec
au
s
e
Z
im
b
ab
wea
n
f
o
r
estr
y
E
R
P
r
ec
o
r
d
s
wer
e
u
n
av
aila
b
le,
th
e
s
tu
d
y
b
u
ilt
a
v
ir
tu
al
E
R
P
d
atase
t
f
r
o
m
th
e
Un
ited
States
Dep
ar
tm
en
t
o
f
Ag
r
icu
ltu
r
e
(
USDA
)
Fo
r
est
Ser
v
ice
tim
b
er
h
ar
v
ests
f
ea
tu
r
e
lay
er
(
8
2
8
,
7
8
9
s
o
u
r
ce
r
ec
o
r
d
s
,
7
0
attr
ib
u
tes).
A
Nu
m
Py
-
b
ased
p
ip
elin
e
(
R
ANDO
M_
SEE
D=
4
2
)
lo
ca
lized
th
e
d
ata
to
s
ix
Ma
n
icala
n
d
f
o
r
estry
r
e
g
io
n
s
,
th
r
ee
co
m
m
e
r
cial
s
p
ec
ies
(
p
in
e
5
5
%,
eu
ca
ly
p
tu
s
3
5
%,
an
d
wattle
1
0
%),
f
o
u
r
c
o
n
tr
ac
to
r
s
ca
le
s
,
an
d
r
e
g
io
n
-
s
p
ec
if
ic
estate
m
ap
p
in
g
s
.
E
ac
h
r
ec
o
r
d
was
en
g
in
ee
r
ed
as
a
d
ec
is
io
n
-
tim
e
allo
ca
tio
n
o
b
s
er
v
atio
n
with
c
o
n
tr
ac
to
r
attr
ib
u
tes,
s
ea
s
o
n
al
co
n
d
itio
n
s
,
d
is
tan
ce
an
d
r
o
u
t
e
in
d
icato
r
s
,
ter
r
ain
s
lo
p
e,
esti
m
ated
ex
t
r
ac
tio
n
v
o
lu
m
e,
c
o
s
t
p
r
o
x
ies,
m
ill
co
m
p
atib
ilit
y
v
ar
iab
le
s
,
an
d
th
e
ta
r
g
et
late_
d
eliv
er
y
_
r
is
k
.
T
h
e
r
esu
lti
n
g
4
6
-
f
ea
tu
r
e
s
ch
em
a
was
d
es
ig
n
ed
to
ex
clu
d
e
p
o
s
t
-
co
n
tr
ac
t
leak
ag
e
wh
ile
s
til
l
r
ef
lectin
g
th
e
o
p
er
atio
n
al
in
f
o
r
m
atio
n
av
ailab
le
to
estate
m
an
ag
er
s
b
ef
o
r
e
co
n
tr
ac
t
o
r
co
m
m
i
tm
en
t.
3
.
5
.
Da
t
a
s
et
s
pli
t
t
ing
a
nd
cla
s
s
dis
t
ri
bu
t
io
n
T
h
e
d
ataset
was
d
i
v
id
ed
in
to
tr
ain
in
g
(
6
0
%,
n
=4
9
7
,
2
7
3
)
,
v
a
lid
atio
n
(
2
0
%,
n
=
1
6
5
,
7
5
8
)
,
an
d
h
o
ld
o
u
t
test
(
2
0
%,
n
=
1
6
5
,
7
5
8
)
s
ets
u
s
in
g
s
tr
atif
ied
t
r
ain
_
test
_
s
p
lit
to
m
ain
tain
class
p
r
o
p
o
r
tio
n
s
.
T
h
e
th
r
ee
-
way
s
p
lit
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
7
,
No
.
3
,
No
v
em
b
er
20
26
:
3
3
7
-
345
340
is
a
m
eth
o
d
o
lo
g
ical
r
eq
u
ir
em
en
t:
th
e
i
n
d
ep
e
n
d
en
t
v
alid
ati
o
n
s
et
allo
ws
t
h
r
esh
o
ld
o
p
ti
m
izatio
n
an
d
m
o
d
el
s
elec
tio
n
with
o
u
t
co
n
tam
in
atin
g
th
e
h
o
ld
o
u
t
s
et,
wh
ich
wa
s
ev
alu
ated
ex
ac
tly
o
n
ce
af
ter
all
d
esig
n
ch
o
ices
wer
e
f
in
alize
d
[
3
0
]
.
T
h
e
p
o
s
itiv
e
class
r
ate
(
late_
d
eliv
er
y
_
r
is
k
=1
)
was 3
0
.
9
3
% a
cr
o
s
s
all
p
ar
titi
o
n
s
,
in
d
icatin
g
m
o
d
er
ate
class
im
b
alan
ce
,
wh
i
ch
was a
d
d
r
ess
ed
u
s
in
g
SMO
T
E
with
in
th
e
tr
ain
in
g
p
ip
elin
e.
3
.
6
.
E
x
perim
ent
a
l
s
et
up
a
nd
ba
s
elines
Fo
u
r
class
if
ier
s
wer
e
tr
ain
ed
an
d
test
ed
i
n
o
r
d
er
t
o
allo
w
r
o
b
u
s
t
c
o
m
p
ar
ativ
e
an
aly
s
is
:
lo
g
is
tic
r
eg
r
ess
io
n
(
lin
ea
r
b
aselin
e)
,
d
ec
is
io
n
tr
e
e
(
n
o
n
-
en
s
em
b
le
b
a
s
elin
e)
,
r
an
d
o
m
f
o
r
est
(
en
s
em
b
le
b
e
n
ch
m
ar
k
)
,
an
d
XGBo
o
s
t
(
p
r
im
ar
y
m
o
d
el)
.
All
wer
e
b
u
ilt
o
n
t
h
e
s
am
e
Scik
it
-
L
ea
r
n
p
ip
eli
n
es
u
s
in
g
C
o
lu
m
n
T
r
an
s
f
o
r
m
er
p
r
ep
r
o
ce
s
s
in
g
(
m
ed
ia
n
im
p
u
ta
tio
n
+Stan
d
ar
d
Scaler
f
o
r
n
u
m
e
r
ics;
On
eHo
tEn
co
d
e
r
f
o
r
ca
te
g
o
r
icals),
th
e
s
am
e
s
tr
atif
ied
f
o
ld
s
,
an
d
ev
alu
ated
o
n
th
e
s
am
e
s
et
o
f
m
etr
ics:
R
OC
-
AU
C
,
P
R
-
AU
C
,
p
r
ec
is
i
o
n
,
r
ec
all,
F1
,
an
d
B
r
ier
s
co
r
e.
T
h
is
d
esig
n
en
s
u
r
es
th
at
p
er
f
o
r
m
an
ce
d
if
f
er
en
ce
s
ar
e
attr
ib
u
tab
le
to
m
o
d
el
ar
ch
itectu
r
e
r
ath
er
th
an
d
ata
p
r
ep
r
o
ce
s
s
in
g
.
T
o
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
DR
AF
as
a
wh
o
le,
it
was
co
m
p
ar
ed
with
two
im
p
licit
b
aselin
es:
i
)
r
an
d
o
m
ass
ig
n
m
en
t,
i.e
.
,
u
n
i
f
o
r
m
r
an
d
o
m
ass
ig
n
m
en
t
o
f
co
n
tr
ac
to
r
-
b
lo
ck
co
m
b
in
atio
n
s
f
r
o
m
t
h
e
f
ea
s
ib
le
p
o
o
l,
r
ep
r
e
s
en
tin
g
th
e
wo
r
s
t
-
ca
s
e
p
r
ac
tical
b
aselin
e
an
d
ii
)
a
c
o
s
t
-
o
n
ly
g
r
ee
d
y
h
eu
r
is
tic
(
s
eq
u
en
tial
allo
ca
tio
n
o
f
th
e
l
o
west
-
co
s
t
f
ea
s
ib
le
co
n
tr
ac
t
o
r
to
ea
c
h
b
lo
ck
b
ased
o
n
E
R
P
-
cen
tr
ic
allo
ca
tio
n
lo
g
ic
)
.
T
h
e
o
p
tim
izatio
n
was
im
p
lem
en
ted
with
p
o
p
u
lati
o
n
s
ize
2
0
0
,
3
0
0
g
en
er
atio
n
s
,
an
d
to
u
r
n
am
en
t
s
elec
tio
n
p
r
ess
u
r
e
o
f
3
,
with
th
e
r
esu
lts
r
ep
licated
ac
r
o
s
s
5
i
n
d
ep
en
d
en
t
NSGA
-
I
I
s
ee
d
s
to
ch
ec
k
th
e
s
tab
ilit
y
o
f
th
e
s
o
lu
tio
n
.
T
h
e
ex
p
e
r
im
e
n
t
d
id
n
o
t
in
clu
d
e
MO
E
A/D,
SP
E
A2
,
o
r
d
ee
p
n
eu
r
al
r
is
k
m
o
d
els
b
ec
au
s
e
t
h
e
r
ev
is
io
n
s
co
p
e
was
to
v
alid
ate
th
e
p
r
o
p
o
s
ed
a
r
ch
itectu
r
e
u
n
d
er
a
co
n
tr
o
lled
b
aselin
e
s
et;
th
ese
alg
o
r
ith
m
s
ar
e
ex
p
licitly
id
en
tifie
d
as th
e
n
ex
t c
o
m
p
ar
ativ
e
la
y
er
f
o
r
f
u
t
u
r
e
wo
r
k
.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
4
.
1
.
P
re
dict
iv
e
m
o
del per
f
o
r
m
a
nce
T
h
e
co
m
p
lete
ev
al
u
atio
n
r
esu
lts
f
o
r
all
f
o
u
r
class
if
ier
s
o
n
b
o
th
v
alid
atio
n
an
d
h
o
l
d
o
u
t
test
s
ets
ar
e
p
r
esen
ted
in
T
ab
le
1
.
T
h
e
h
o
ld
o
u
t
s
et
was
ass
e
s
s
ed
o
n
ce
af
ter
m
o
d
el
s
elec
tio
n
in
o
r
d
er
to
esti
m
ate
g
en
er
aliza
tio
n
p
e
r
f
o
r
m
an
ce
i
n
an
u
n
b
iased
m
an
n
er
.
B
ased
o
n
th
e
h
ig
h
est
v
alid
atio
n
PR
-
AUC
(
0
.
8
6
9
)
,
XGBo
o
s
t
was
ch
o
s
en
as
th
e
f
in
al
m
o
d
el.
T
h
e
h
o
ld
o
u
t
c
o
n
f
u
s
io
n
m
atr
ix
was:
tr
u
e
n
eg
a
tiv
e
(
TN
)
=9
8
,
8
9
6
,
FP
=1
5
,
5
9
8
,
FN=6
2
,
tr
u
e
p
o
s
itiv
e
(
TP
)
=5
1
,
2
0
2
.
T
h
e
n
ea
r
-
z
er
o
f
alse
n
eg
ativ
e
co
u
n
t
(
6
2
o
f
5
1
,
2
6
4
p
o
s
itiv
e
ca
s
es)
is
o
p
er
atio
n
ally
cr
itical:
ea
ch
f
alse
n
eg
ativ
e
co
r
r
es
p
o
n
d
s
to
an
u
n
d
etec
te
d
h
ig
h
-
r
is
k
allo
ca
tio
n
th
at
wo
u
ld
p
lace
a
h
ig
h
-
v
alu
e
tim
b
er
b
lo
ck
with
an
in
ca
p
ac
itated
co
n
tr
ac
to
r
.
W
ith
a
B
r
ier
s
co
r
e
o
f
0
.
0
5
3
th
e
p
r
o
b
a
b
ilit
y
o
u
t
p
u
ts
ar
e
v
er
y
w
ell
tu
n
ed
an
d
ca
n
b
e
d
ir
ec
tly
m
ap
p
ed
o
n
t
o
th
e
o
p
tim
izatio
n
o
b
jectiv
e
f
u
n
ctio
n
s
with
o
u
t
d
o
i
n
g
a
n
y
r
etu
n
in
g
p
o
s
t
-
h
o
c.
T
h
e
R
OC
-
AUC
im
p
r
o
v
em
en
t
o
f
0
.
3
6
5
o
v
er
lo
g
is
tic
r
eg
r
ess
io
n
is
in
lin
e
with
b
en
ch
m
ar
k
s
r
ep
o
r
ted
i
n
s
im
ilar
co
n
tr
ac
to
r
r
is
k
co
n
te
x
ts
[
7
]
,
[
8
]
a
n
d
p
r
o
v
id
es
em
p
ir
i
ca
l
s
u
p
p
o
r
t
f
o
r
th
e
en
s
em
b
le
n
o
n
-
lin
ea
r
ar
ch
itectu
r
e
o
u
tp
er
f
o
r
m
i
n
g
lin
ea
r
alter
n
ativ
es
o
n
th
is
clas
s
o
f
p
r
o
b
lem
.
T
h
e
v
er
y
lo
w
PR
-
AU
C
o
f
lo
g
is
tic
r
eg
r
ess
io
n
(
0
.
3
1
0
)
co
n
f
ir
m
s
th
at
t
h
e
r
i
s
k
class
if
icatio
n
p
r
o
b
lem
is
n
o
t
lin
ea
r
ly
s
ep
ar
ab
le
an
d
th
at
lin
ea
r
d
is
cr
im
in
an
ts
a
r
e
p
r
ac
tically
in
s
u
f
f
icie
n
t f
o
r
o
p
er
atio
n
al
d
e
p
lo
y
m
e
n
t.
T
ab
le
1
.
C
lass
if
ier
p
er
f
o
r
m
a
n
c
e
co
m
p
ar
is
o
n
—
v
alid
atio
n
an
d
h
o
ld
o
u
t te
s
t sets
(
p
r
im
ar
y
tar
g
et:
late_
d
eliv
er
y
_
r
is
k
)
Mo
d
e
l
V
a
l
.
p
r
e
c
.
V
a
l
.
r
e
c
.
V
a
l
.
F
1
V
a
l
.
R
O
C
-
AUC
V
a
l
.
P
R
-
AUC
Te
st
F
1
Te
st
R
O
C
-
AUC
Te
st
B
r
i
e
r
X
G
B
o
o
st
✓
0
.
7
6
5
0
.
9
9
9
0
.
8
6
7
0
.
9
6
5
0
.
8
6
9
0
.
8
6
7
0
.
9
6
5
0
.
0
5
3
R
a
n
d
o
m f
o
r
e
s
t
0
.
7
6
3
0
.
9
9
9
0
.
8
6
5
0
.
9
6
3
0
.
8
6
6
0
.
8
6
4
0
.
9
6
3
0
.
0
5
5
D
e
c
i
s
i
o
n
t
r
e
e
0
.
7
6
5
0
.
9
9
9
0
.
8
6
6
0
.
9
6
4
0
.
8
6
7
0
.
8
6
5
0
.
9
6
2
0
.
0
5
6
Lo
g
i
s
t
i
c
r
e
g
r
e
ss
i
o
n
0
.
3
0
9
1
.
0
0
0
0
.
4
7
2
0
.
6
0
0
0
.
3
1
0
0
.
4
7
1
0
.
6
0
1
0
.
2
1
3
4
.
2
.
F
e
a
t
ure
im
po
rt
a
nce
a
nd
s
ha
pley
a
dd
it
iv
e
ex
pla
na
t
io
ns
(
SH
AP
)
a
na
ly
s
is
R
ain
f
all
-
s
ea
s
o
n
v
ar
iab
les
wer
e
id
en
tifie
d
as
th
e
lead
in
g
d
r
iv
er
s
o
f
r
is
k
b
o
th
th
r
o
u
g
h
Gi
n
i
f
ea
tu
r
e
im
p
o
r
tan
ce
a
n
d
SHAP
an
aly
s
is
.
is
_
wet_
s
ea
s
o
n
h
ad
t
h
e
h
ig
h
est
Gin
i
im
p
o
r
tan
c
e
(
0
.
6
7
6
)
wh
e
r
ea
s
m
o
n
th
ly
_
r
ain
f
all_
m
m
h
ad
t
h
e
h
ig
h
est
m
ea
n
a
b
s
o
lu
te
SHAP
v
alu
e
(
5
.
2
4
7
)
.
T
o
g
eth
er
,
wet
-
s
ea
s
o
n
,
r
ain
f
all
-
lev
el,
an
d
r
ain
f
all
-
v
o
l
u
m
e
v
ar
iab
les
ac
co
u
n
t
f
o
r
m
o
r
e
th
an
8
4
%
o
f
p
r
ed
ictiv
e
i
m
p
o
r
tan
ce
,
w
h
ich
is
co
n
s
is
ten
t
with
s
o
ciété
g
én
ér
a
le
d
e
s
u
r
v
eillan
ce
(
SGS
)
au
d
it
o
b
s
er
v
atio
n
s
th
at
wet
-
s
ea
s
o
n
r
o
ad
d
e
g
r
ad
atio
n
is
a
m
ajo
r
ca
u
s
e
o
f
co
n
t
r
ac
to
r
d
e
liv
er
y
f
ailu
r
e
i
n
Z
im
b
a
b
wea
n
f
o
r
estry
[
3
1
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
R
is
k
-
in
teg
r
a
ted
co
n
tr
a
cto
r
a
ll
o
ca
tio
n
in
Zimb
a
b
w
e'
s
timb
er
va
lu
e
ch
a
i
n
(
Ta
ve
n
g
w
a
N
o
r
ma
n
)
341
4
.
3
.
M
ulti
-
o
bje
ct
iv
e
o
ptim
iz
a
t
io
n r
esu
lt
s
Of
8
2
8
,
7
8
9
to
tal
ca
n
d
id
ate
as
s
ig
n
m
en
ts
,
6
3
9
,
7
7
2
(
7
7
.
2
7
%)
p
ass
ed
s
tr
ict
f
ea
s
ib
ilit
y
f
ilter
i
n
g
(
ter
r
ain
s
lo
p
e,
r
o
u
te
f
it,
m
ill
co
m
p
atib
ilit
y
,
an
d
r
ec
o
v
er
y
f
ea
s
ib
ilit
y
)
an
d
en
ter
ed
th
e
NSGA
-
I
I
o
p
tim
izatio
n
p
o
o
l.
T
ab
le
2
r
ep
o
r
ts
th
e
s
ce
n
ar
io
-
le
v
el
o
p
tim
izatio
n
o
u
tco
m
es.
So
lu
tio
n
d
iv
er
s
ity
was
co
n
f
ir
m
e
d
in
all
ca
s
es:
ea
ch
s
o
lu
tio
n
h
ad
a
d
is
tin
ct
co
n
tr
a
cto
r
-
b
lo
c
k
-
m
ill
ass
ig
n
m
en
t
s
et
(
m
ax
_
r
ep
ea
te
d
_
s
et_
co
u
n
t=1
)
,
wh
ich
in
d
icate
s
tr
u
e
d
is
co
v
er
y
o
f
th
e
Par
eto
f
r
o
n
t
an
d
n
o
t
p
r
em
atu
r
e
co
n
v
er
g
en
ce
.
T
h
e
b
alan
ce
d
an
d
h
ig
h
-
r
ec
o
v
er
y
s
ce
n
ar
io
s
p
r
o
d
u
ce
d
6
4
f
u
lly
f
ea
s
ib
le,
n
o
n
-
d
o
m
in
ated
s
o
lu
tio
n
s
ea
c
h
with
ze
r
o
co
n
s
tr
ain
t
v
io
lat
io
n
s
,
g
iv
in
g
estate
m
an
ag
er
s
a
r
ich
an
d
u
s
ab
l
e
d
ec
is
io
n
s
p
ac
e
alo
n
g
th
e
co
s
t
-
r
ec
o
v
er
y
-
d
elay
-
r
eliab
ilit
y
tr
ad
e
-
o
f
f
f
r
o
n
t.
T
h
e
b
alan
ce
d
s
ce
n
ar
i
o
is
th
er
ef
o
r
e
th
e
m
o
s
t
d
ep
lo
y
ab
le
p
o
l
icy
s
ettin
g
b
ec
a
u
s
e
it
k
ee
p
s
r
e
liab
ilit
y
co
n
s
tr
ain
ts
ac
tiv
e
with
o
u
t
s
u
p
p
r
ess
in
g
f
e
asib
le
y
ield
o
p
tio
n
s
,
wh
e
r
ea
s
th
e
h
ig
h
-
r
ec
o
v
er
y
s
ce
n
ar
i
o
i
s
m
o
s
t
u
s
ef
u
l
wh
en
m
ills
f
ac
e
th
r
o
u
g
h
p
u
t
p
r
ess
u
r
e
an
d
ca
n
to
ler
ate
m
o
d
er
ate
r
is
k
.
Su
ch
a
lo
w
-
r
is
k
s
ce
n
ar
io
in
f
ea
s
ib
ilit
y
(
m
ea
n
v
io
latio
n
=0
.
0
3
6
)
is
an
aly
tically
in
s
tr
u
ctiv
e.
T
h
e
o
v
er
ly
s
tr
in
g
en
t
r
is
k
ce
ilin
g
o
f
0
.
4
5
is
v
io
lated
b
ec
au
s
e
th
e
p
r
ed
icted
m
ea
n
r
is
k
in
th
e
c
an
d
id
ate
p
o
o
l
is
0
.
2
0
1
;
co
m
b
in
ed
with
tig
h
t
lo
ad
lim
its
(
0
.
9
5
)
a
n
d
a
lim
ited
p
lan
n
in
g
h
o
r
izo
n
(
5
b
lo
ck
s
)
,
f
ewe
r
th
an
th
e
m
in
im
u
m
n
u
m
b
er
o
f
co
n
s
tr
ain
t
-
s
atis
f
y
in
g
ass
ig
n
m
en
t
co
m
b
in
atio
n
s
ar
e
p
o
s
s
ib
le
f
o
r
a
f
u
lly
f
ea
s
ib
le
Par
eto
s
et.
T
h
is
is
a
co
n
s
tr
ain
t
-
f
ea
s
ib
ilit
y
b
o
u
n
d
ar
y
c
o
n
d
itio
n
:
th
e
f
r
am
ewo
r
k
s
u
r
f
ac
es
th
is
co
n
d
itio
n
ex
p
licitly
t
h
r
o
u
g
h
th
e
d
ash
b
o
ar
d
’
s
s
ce
n
ar
io
co
m
p
a
r
is
o
n
p
an
el
in
s
tead
o
f
s
ilen
tly
r
etu
r
n
in
g
a
s
u
b
-
o
p
tim
al
s
o
lu
tio
n
.
E
s
tate
m
an
ag
er
s
th
er
ef
o
r
e
r
ec
eiv
e
clea
r
o
p
er
at
io
n
al
g
u
id
an
ce
th
at
a
r
is
k
ce
ilin
g
b
elo
w
0
.
4
5
r
e
q
u
ir
es
eith
er
ex
p
an
d
i
n
g
th
e
co
n
tr
ac
to
r
r
eg
is
tr
y
,
r
ela
x
in
g
l
o
a
d
lim
its
,
o
r
s
h
if
tin
g
wet
-
s
ea
s
o
n
b
lo
ck
s
to
a
later
p
l
an
n
in
g
win
d
o
w.
T
ab
le
2
.
NSGA
-
I
I
s
ce
n
ar
io
o
p
tim
izatio
n
s
u
m
m
ar
y
S
c
e
n
a
r
i
o
B
l
o
c
k
s
O
p
t
i
o
n
s/
b
l
o
c
k
P
a
r
e
t
o
s
o
l
u
t
i
o
n
s
F
e
a
si
b
l
e
(
%)
M
e
a
n
c
o
n
s
t
r
a
i
n
t
v
i
o
l
a
t
i
o
n
B
a
l
a
n
c
e
d
6
14
64
1
0
0
.
0
0
.
0
0
0
H
i
g
h
-
r
e
c
o
v
e
r
y
6
16
64
1
0
0
.
0
0
.
0
0
0
Lo
w
-
r
i
s
k
5
12
32
0
.
0
0
.
0
3
6
4
.
4
.
Co
m
pa
riso
n
a
g
a
ins
t
ba
s
elines
T
ab
le
3
p
r
esen
ts
a
s
tr
u
ctu
r
ed
co
m
p
a
r
is
o
n
o
f
DR
AF
ag
ain
s
t
th
e
b
aselin
e
m
eth
o
d
s
ev
alu
ated
.
T
h
e
s
alien
t
r
ea
s
o
n
wh
y
DR
AF
is
b
etter
th
an
th
e
co
s
t
-
g
r
ee
d
y
h
eu
r
is
tic
is
n
o
t
s
im
p
ly
th
e
d
ash
b
o
ar
d
;
we
in
s
er
t
r
is
k
in
to
th
e
o
p
tim
izatio
n
o
b
je
ctiv
e
its
elf
.
Ho
wev
er
,
a
r
is
k
-
f
r
ee
NSGA
-
I
I
ca
n
s
till
u
n
c
o
v
er
a
Par
eto
f
r
o
n
t
a
n
d
m
ay
in
s
tead
f
av
o
u
r
c
h
ea
p
o
r
h
ig
h
-
y
iel
d
ass
ig
n
m
e
n
ts
th
at
ar
e
f
r
a
g
ile
in
p
r
ac
tice.
DR
AF
’
s
r
is
k
-
d
is
co
u
n
ted
r
ec
o
v
er
y
tar
g
et
p
u
s
h
es th
ese
o
p
tio
n
s
to
th
e
r
ea
r
,
a
n
d
allo
ws m
an
ag
er
s
to
allo
ca
te
s
ets o
f
r
e
s
o
u
r
ce
s
th
at
ar
e
s
till
co
s
t
-
ef
f
ec
tiv
e
s
in
ce
th
ey
c
o
n
tr
o
l f
o
r
d
eliv
er
y
r
is
k
[
3
2
]
.
T
ab
le
3
.
DR
AF v
er
s
u
s
b
aselin
e
m
eth
o
d
s
: p
er
f
o
r
m
a
n
ce
ac
r
o
s
s
all
ev
alu
atio
n
d
im
en
s
io
n
s
D
i
me
n
si
o
n
R
a
n
d
o
m
a
ssi
g
n
me
n
t
C
o
s
t
g
r
e
e
d
y
h
e
u
r
i
s
t
i
c
Lo
g
i
s
t
i
c
r
e
g
r
e
ssi
o
n
+
M
O
O
D
R
A
F
(
X
G
B
o
o
st
+
N
S
G
A
-
II)
P
r
e
d
i
c
t
i
o
n
R
O
C
-
AUC
—
—
0
.
6
0
0
0
.
9
6
5
R
e
c
a
l
l
(
r
i
s
k
d
e
t
e
c
t
i
o
n
)
—
—
1
.
0
0
(
p
o
o
r
p
r
e
c
i
s
i
o
n
)
0
.
9
9
9
(
w
i
t
h
p
r
e
c
i
si
o
n
)
R
i
s
k
i
n
o
b
j
e
c
t
i
v
e
f
u
n
c
t
i
o
n
No
No
Th
r
e
s
h
o
l
d
f
i
l
t
e
r
o
n
l
y
Emb
e
d
d
e
d
m
u
l
t
i
p
l
i
c
a
t
i
v
e
l
y
P
a
r
e
t
o
s
o
l
u
t
i
o
n
s
(
b
a
l
a
n
c
e
d
)
N
/
A
1
(
s
i
n
g
l
e
h
e
u
r
i
s
t
i
c
)
~8
–
1
2
(
r
i
s
k
-
f
r
e
e
f
r
o
n
t
)
6
4
(
r
i
sk
-
i
n
t
e
g
r
a
t
e
d
)
M
e
a
n
c
o
n
s
t
r
a
i
n
t
v
i
o
l
a
t
i
o
n
s
H
i
g
h
(
~
2
3
%
i
n
f
e
a
s
i
b
l
e
)
M
o
d
e
r
a
t
e
Lo
w
Ze
r
o
(
b
a
l
a
n
c
e
d
,
h
i
g
h
-
r
e
c
.
)
D
e
c
i
s
i
o
n
s
u
p
p
o
r
t
N
o
n
e
S
t
a
t
i
c
ER
P
t
a
b
l
e
O
f
f
l
i
n
e
r
e
p
o
r
t
R
e
a
l
-
t
i
me
i
n
t
e
r
a
c
t
i
v
e
D
S
S
S
t
a
k
e
h
o
l
d
e
r
sa
t
i
sf
a
c
t
i
o
n
—
—
—
4
.
1
9
/
5
.
0
4
.
5
.
St
a
k
eho
lder
ev
a
lua
t
io
n
re
s
ults
Af
ter
a
s
tr
u
ctu
r
ed
d
em
o
n
s
tr
atio
n
,
th
e
DR
AF
ar
tef
ac
t
was
ev
alu
ated
b
y
t
h
ir
ty
s
ec
to
r
r
ep
r
es
en
tativ
es
:
eig
h
t
estate
m
an
ag
er
s
,
eig
h
t
c
o
n
tr
ac
to
r
s
,
s
ev
en
I
T
/ERP
s
taf
f
,
an
d
s
ev
e
n
p
o
licy
r
ep
r
esen
ta
tiv
es.
T
h
e
s
ec
tio
n
-
lev
el
m
ea
n
L
ik
er
t
s
co
r
es
a
r
e
s
h
o
wn
i
n
T
a
b
le
4
.
Me
th
o
d
o
l
o
g
ical
tr
an
s
p
ar
en
c
y
:
th
is
was
r
ated
(
4
.
6
/5
.
0
)
h
ig
h
est
f
o
r
I
T
/ERP
s
taf
f
with
th
e
f
ix
ed
r
an
d
o
m
s
ee
d
(
R
ANDO
M_
SEE
D=
4
2
)
,
s
tr
atif
ied
s
p
lits
a
n
d
s
er
ialized
m
o
d
el
ar
tef
ac
ts
b
ein
g
f
o
u
n
d
to
d
is
tin
g
u
is
h
th
em
f
r
o
m
a
d
h
o
c
a
n
aly
t
ica
l
s
cr
ip
ts
.
E
s
tate
m
an
ag
er
s
r
ated
th
e
f
r
am
ewo
r
k
h
ig
h
est
o
v
er
all
(
4
.
2
6
/
5
.
0
)
,
with
q
u
alitativ
e
f
ee
d
b
ac
k
h
ig
h
li
g
h
tin
g
th
e
u
tili
ty
o
f
b
ei
n
g
ab
le
to
test
a
h
ea
v
y
-
r
ain
f
all
s
ce
n
ar
io
b
ef
o
r
e
s
ea
s
o
n
al
co
m
m
itm
e
n
t.
C
o
n
tr
ac
to
r
s
ca
m
e
in
with
th
e
lo
west
to
tal
s
co
r
e
(
3
.
9
8
/
5
.
0
)
,
b
u
t
m
eth
o
d
o
l
o
g
ical
tr
a
n
s
p
ar
en
cy
ap
p
ea
r
ed
to
b
e
a
m
ajo
r
is
s
u
e
(
3
.
8
)
i
n
p
a
r
t
b
ec
a
u
s
e
o
f
th
e
d
is
p
ar
ity
b
etwe
en
s
tatis
t
ical
r
ig
o
u
r
an
d
ac
ce
s
s
ib
ilit
y
f
o
r
f
ield
o
p
e
r
ato
r
s
,
s
u
g
g
esti
n
g
th
at
a
s
im
p
lific
atio
n
o
n
th
e
r
is
k
-
s
co
r
e
ex
p
lan
atio
n
lay
er
h
as
g
iv
en
th
e
p
r
o
ject
its
s
h
o
r
t
-
ter
m
f
o
cu
s
o
f
d
ev
elo
p
m
en
t.
T
h
e
p
ap
e
r
p
r
esen
ts
an
in
teg
r
ated
r
is
k
-
in
teg
r
ated
MO
O
-
b
ased
r
is
k
m
an
ag
e
m
en
t
f
r
am
ewo
r
k
f
o
r
o
p
er
atio
n
al
r
eso
u
r
ce
allo
ca
tio
n
,
f
o
r
m
alize
d
as
a
m
ath
em
atica
l p
atter
n
o
n
a
c
o
n
ce
p
tu
al
lev
el
as a
n
a
r
ch
itectu
r
al
d
esig
n
p
at
ter
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
7
,
No
.
3
,
No
v
em
b
er
20
26
:
3
3
7
-
345
342
4
.
6
.
T
heo
re
t
ica
l c
o
ntr
ibu
t
io
n a
nd
dis
c
us
s
io
n
T
h
e
p
a
p
er
’
s
k
ey
co
n
tr
i
b
u
tio
n
to
th
e
th
eo
r
etica
l
co
m
m
u
n
ity
was
th
at
it
d
ef
in
es
an
ar
ch
itectu
r
al
ap
p
r
o
ac
h
f
o
r
a
f
o
r
m
alize
d
s
tr
u
ctu
r
e
f
o
r
o
p
er
atio
n
al
r
eso
u
r
ce
o
p
tim
i
s
atio
n
f
r
o
m
r
is
k
-
in
teg
r
ated
MO
O
.
Alth
o
u
g
h
p
r
ev
io
u
s
wo
r
k
s
h
a
v
e
in
teg
r
ated
ML
a
n
d
MO
O
i
n
s
u
p
p
ly
c
h
ain
s
ett
in
g
s
[
2
9
]
,
[
3
2
]
,
in
tr
ad
itio
n
al
ar
ch
itectu
r
e
ML
r
esu
lts
ar
e
c
o
n
s
id
er
ed
as
f
ea
s
ib
ilit
y
f
ilter
s
o
r
d
iag
n
o
s
tics
af
ter
o
p
tim
izatio
n
.
DR
AF
r
ev
er
s
es
th
is
r
elatio
n
:
th
e
ML
-
g
en
er
ate
d
r
is
k
p
r
o
b
a
b
ilit
y
ca
n
b
e
in
co
r
p
o
r
ated
in
to
th
e
r
e
co
v
er
y
o
b
je
ctiv
e
as
a
d
is
co
u
n
t
f
ac
to
r
–
s
o
th
e
ev
o
l
u
tio
n
ar
y
s
ea
r
ch
will
b
e
d
ictated
b
y
th
e
p
r
o
b
a
b
ilit
y
in
th
e
f
o
r
m
o
f
th
e
r
is
k
its
elf
.
T
h
is
is
s
im
ilar
to
r
is
k
ad
ju
s
ted
r
etu
r
n
o
p
tim
izatio
n
in
p
o
r
t
f
o
lio
th
eo
r
y
,
b
u
t im
p
lem
en
te
d
h
er
e
to
an
in
teg
er
ass
ig
n
m
en
t
p
r
o
b
lem
with
ter
r
ain
an
d
ca
p
ac
ity
co
n
s
tr
ain
ts
.
T
h
e
c
o
n
tr
ib
u
tio
n
is
also
n
estl
ed
in
t
o
d
ec
is
io
n
-
s
u
p
p
o
r
t
ar
ch
itectu
r
e
th
eo
r
y
:
p
r
ed
ictiv
e
an
aly
tics
,
o
p
tim
izatio
n
,
a
n
d
h
u
m
an
s
ce
n
a
r
io
s
elec
tio
n
ar
e
tr
ea
ted
as
co
u
p
led
s
er
v
ices,
n
o
t
s
ilo
-
b
ased
r
e
p
o
r
ts
[
2
6
]
,
[
2
8
]
.
I
n
ad
d
itio
n
,
th
e
f
r
am
ewo
r
k
is
d
ir
ec
tly
a
p
p
lic
ab
le
to
d
is
tr
ib
u
ted
s
y
s
tem
s
an
d
clo
u
d
-
b
ased
E
R
P
ex
p
an
s
io
n
,
as
ea
ch
lay
er
c
an
b
e
ex
p
o
s
ed
as
a
m
o
d
u
lar
s
er
v
ice
o
n
ce
r
ea
l
co
n
tr
ac
to
r
,
r
ai
n
f
all
an
d
m
ill
-
th
r
o
u
g
h
p
u
t
d
ata
s
tr
ea
m
s
ar
e
co
llected
.
T
h
ese
r
esu
lt
in
s
tr
u
ctu
r
ally
d
if
f
e
r
en
t
Par
eto
f
r
o
n
ts
co
m
p
ar
e
d
with
r
is
k
-
f
r
ee
f
r
o
n
ts
:
th
ey
s
y
s
tem
at
ically
ex
clu
d
e
s
tr
ateg
ies
th
at
ar
e
ec
o
n
o
m
ical
b
u
t
o
p
er
atio
n
ally
u
n
s
o
u
n
d
,
g
iv
in
g
d
ec
is
io
n
m
ak
er
s
a
Par
eto
-
o
p
ti
m
al
s
et
o
f
s
o
lu
tio
n
s
in
th
e
p
r
ac
tically
r
elev
an
t
o
b
jectiv
e
s
p
ac
e
an
d
n
o
t a
n
ar
r
o
w
d
eter
m
in
is
tic
ef
f
icien
cy
s
p
ac
e.
Seco
n
d
,
th
e
m
o
d
el
estab
lis
h
es
th
at
th
e
d
o
m
in
an
t
d
r
iv
e
r
o
f
o
p
er
atin
g
r
is
k
in
th
e
Z
i
m
b
ab
wea
n
co
m
m
er
cial
f
o
r
estry
s
ec
to
r
is
s
ea
s
o
n
al
r
ain
f
all.
T
h
e
SHAP
an
aly
s
is
,
co
m
b
in
ed
with
ex
p
lo
r
ato
r
y
d
ata
an
al
y
s
is
(
E
DA)
cr
o
s
s
-
p
h
ase
co
r
r
elatio
n
r
esu
lts
,
in
d
icate
s
th
at
is
_
wet_
s
ea
s
o
n
an
d
m
o
n
t
h
ly
_
r
ai
n
f
all_
m
m
is
jo
in
tly
s
ig
n
if
ican
t
p
r
ed
icto
r
s
ac
co
u
n
t
in
g
f
o
r
m
o
r
e
th
an
8
4
%
o
f
f
e
atu
r
e
im
p
o
r
ta
n
ce
m
ass
.
T
h
is
co
n
clu
s
io
n
,
d
e
r
iv
ed
f
r
o
m
a
r
e
p
r
o
d
u
cib
le
co
m
p
u
tati
o
n
al
p
ip
elin
e
r
at
h
er
th
an
q
u
ali
tativ
e
ex
p
er
t
o
p
i
n
io
n
,
g
iv
es
em
p
ir
ical
s
u
p
p
o
r
t f
o
r
s
ea
s
o
n
al
allo
ca
tio
n
s
ch
ed
u
lin
g
,
an
in
s
ig
h
t
th
at
estate
m
a
n
a
g
er
s
ca
n
a
d
o
p
t
with
o
u
t
d
e
p
lo
y
in
g
th
e
f
u
ll
DR
AF
p
ip
elin
e.
Fro
m
a
p
o
licy
p
e
r
s
p
e
ctiv
e,
th
e
r
esu
lt
s
u
g
g
est
s
th
at
n
atio
n
al
f
o
r
estry
r
es
o
u
r
ce
-
m
a
n
ag
em
e
n
t
g
u
id
elin
es
s
h
o
u
ld
r
eq
u
ir
e
co
n
tr
ac
to
r
s
to
m
ain
tain
wet
-
s
ea
s
o
n
r
ea
d
i
n
ess
ev
id
en
ce
,
s
u
ch
as
r
o
ad
-
ac
ce
s
s
p
lan
s
,
eq
u
ip
m
en
t
r
e
d
u
n
d
an
cy
,
an
d
v
er
if
iab
le
d
eliv
er
y
h
is
to
r
ies,
b
ef
o
r
e
h
ig
h
-
r
is
k
b
lo
c
k
s
ar
e
allo
ca
ted
d
u
r
in
g
r
ain
f
all
-
s
en
s
itiv
e
p
er
io
d
s
.
T
ab
le
4
.
Stak
eh
o
ld
er
e
v
alu
atio
n
s
co
r
es b
y
s
ec
tio
n
a
n
d
g
r
o
u
p
(
1
–
5
L
i
k
er
t scale
)
Ev
a
l
u
a
t
i
o
n
sec
t
i
o
n
Est
a
t
e
ma
n
a
g
e
r
s
C
o
n
t
r
a
c
t
o
r
s
I
T/
ER
P
st
a
f
f
P
o
l
i
c
y
m
a
k
e
r
s
O
v
e
r
a
l
l
O
r
i
g
i
n
a
l
i
t
y
a
n
d
n
o
v
e
l
t
y
4
.
3
4
.
0
4
.
5
4
.
3
4
.
3
P
r
a
c
t
i
c
a
l
a
p
p
l
i
c
a
b
i
l
i
t
y
4
.
2
4
.
1
4
.
0
3
.
9
4
.
1
M
e
t
h
o
d
o
l
o
g
i
c
a
l
t
r
a
n
sp
a
r
e
n
c
y
4
.
1
3
.
8
4
.
6
4
.
2
4
.
2
C
l
a
r
i
t
y
a
n
d
p
r
e
s
e
n
t
a
t
i
o
n
4
.
3
3
.
9
4
.
2
4
.
1
4
.
1
Ev
a
l
u
a
t
i
o
n
v
s
.
b
a
se
l
i
n
e
s
4
.
4
4
.
1
4
.
5
4
.
3
4
.
3
O
v
e
r
a
l
l
mea
n
4
.
2
6
3
.
9
8
4
.
3
6
4
.
1
6
4
.
1
9
4
.
7
.
M
et
ho
do
lo
g
ic
a
l
t
ra
ns
pa
re
ncy
a
nd
re
pro
du
cibi
lity
DR
AF
is
r
ea
lized
as
a
s
in
g
le
s
elf
-
co
n
tain
e
d
J
u
p
y
ter
No
teb
o
o
k
with
a
f
ix
ed
r
an
d
o
m
s
ee
d
(
R
ANDO
M_
SEE
D=
4
2
)
,
d
eter
m
in
is
tic
s
tr
atif
ied
s
p
lits
,
s
e
r
ialized
m
o
d
el
ar
tef
ac
ts
(
XGBo
o
s
t
p
ip
elin
e
at
o
u
tp
u
ts
/m
o
d
els/
b
est_
m
o
d
el
_
p
i
p
elin
e.
jo
b
lib
)
,
a
n
d
s
tr
u
ctu
r
e
d
o
u
tp
u
t
d
ir
ec
to
r
ies.
All
r
ep
o
r
ted
m
etr
ics
ar
e
p
r
o
d
u
ce
d
b
y
th
is
No
teb
o
o
k
an
d
ca
n
b
e
r
ec
r
ea
te
d
b
y
r
u
n
n
in
g
th
e
ce
lls
in
tu
r
n
in
a
f
r
esh
Py
th
o
n
3
.
x
en
v
ir
o
n
m
en
t w
ith
th
e
ap
p
r
o
p
r
i
ate
lib
r
ar
y
v
er
s
io
n
s
.
T
h
is
r
ep
r
o
d
u
cib
ilit
y
ar
ch
i
tectu
r
e
is
a
d
elib
er
ate
r
esp
o
n
s
e
to
th
e
r
ep
licatio
n
c
r
is
is
in
ap
p
li
ed
ML
,
in
wh
ic
h
r
e
p
o
r
ted
p
e
r
f
o
r
m
a
n
ce
m
etr
ics
ar
e
o
f
ten
n
o
n
-
r
e
p
r
o
d
u
cib
le
b
y
d
ef
au
lt
o
win
g
to
s
to
ch
asti
c
tr
ain
in
g
,
u
n
d
o
cu
m
e
n
ted
p
r
e
p
r
o
ce
s
s
in
g
,
o
r
u
n
c
o
n
tr
o
lled
d
ata
leak
ag
e
[
3
3
]
.
T
h
e
leak
a
g
e
-
s
af
e
f
ea
tu
r
e
en
g
i
n
ee
r
in
g
p
r
o
to
co
l
,
wh
ich
ex
clu
d
es
p
o
s
t
-
o
u
tco
m
e
v
ar
iab
les
(
a
ctu
al
r
ec
o
v
er
y
r
ate,
ac
tu
al
co
s
t,
an
d
d
eliv
e
r
y
s
tatu
s
)
f
r
o
m
th
e
p
r
ed
icto
r
s
et
,
is
p
ar
ticu
lar
ly
im
p
o
r
tan
t.
I
n
o
p
er
a
tio
n
al
ML
m
o
d
els,
leak
ag
e
in
f
latio
n
r
esu
lts
in
o
v
er
-
o
p
tim
is
tic
m
o
d
el
p
er
f
o
r
m
an
ce
esti
m
ates
th
at
co
llap
s
e
in
p
r
o
d
u
ctio
n
.
T
h
e
leak
ag
e
v
alid
atio
n
lay
e
r
(
c
ell
1
1
o
f
t
h
e
No
te
b
o
o
k
)
s
h
o
ws
wh
ich
co
lu
m
n
s
we
r
e
ex
clu
d
ed
,
with
ex
p
lan
atio
n
a
n
d
ju
s
tific
atio
n
f
o
r
ea
ch
e
x
clu
s
io
n
,
e
n
ab
lin
g
c
o
m
p
lete
au
d
itab
ilit
y
.
4
.
8
.
Appl
ica
bil
it
y
,
deplo
y
m
e
nt,
a
nd
lim
it
a
t
io
ns
Dep
lo
y
m
en
t
h
in
g
es
o
n
th
r
ee
p
ar
am
eter
s
:
s
tr
u
ctu
r
e
d
co
n
tr
ac
to
r
r
ec
o
r
d
s
,
alig
n
m
en
t
to
E
R
P
s
ch
em
as,
an
d
m
an
ag
e
r
ial
r
ea
d
in
ess
to
ac
t
o
n
alg
o
r
ith
m
ic
r
ec
o
m
m
en
d
atio
n
s
.
Stak
eh
o
ld
er
f
ee
d
b
ac
k
i
n
d
icate
s
s
h
o
r
t
-
ter
m
f
ea
s
ib
ilit
y
,
g
iv
en
th
at
I
T
/ERP
s
taf
f
r
ate
in
teg
r
atio
n
f
ea
s
ib
ilit
y
as
4
.
0
/5
.
0
an
d
estate
m
an
ag
er
s
s
u
p
p
o
r
te
d
s
ce
n
ar
io
-
s
en
s
itiv
ity
to
o
ls
.
Ho
wev
er
,
liv
e
u
s
e
wo
u
ld
s
till
d
em
an
d
r
ea
l
-
tim
e
E
R
P
in
teg
r
atio
n
,
m
o
b
ile
f
ield
ca
p
tu
r
e,
r
o
le
-
b
ased
ac
ce
s
s
c
o
n
tr
o
ls
,
p
er
i
o
d
ic
m
o
d
el
r
ec
al
ib
r
atio
n
,
an
d
a
g
o
v
er
n
an
ce
p
r
o
ce
s
s
to
o
v
e
r
r
id
e
r
ec
o
m
m
en
d
atio
n
s
wh
en
lo
c
al
k
n
o
wled
g
e
co
n
tr
ad
icts
t
h
e
m
o
d
el.
Fo
u
r
ca
v
ea
ts
m
ay
b
e
id
en
tifie
d
.
T
h
e
s
y
n
th
etic
d
ataset
m
ea
n
s
t
h
e
h
o
l
d
o
u
t
m
etr
ics
ar
e
an
u
p
p
er
b
o
u
n
d
;
liv
e
E
R
P
d
ata
will
h
av
e
m
is
s
in
g
n
ess
,
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
R
is
k
-
in
teg
r
a
ted
co
n
tr
a
cto
r
a
ll
o
ca
tio
n
in
Zimb
a
b
w
e'
s
timb
er
va
lu
e
ch
a
i
n
(
Ta
ve
n
g
w
a
N
o
r
ma
n
)
343
b
ias,
o
r
g
an
izatio
n
al
id
i
o
s
y
n
cr
asies
,
an
d
n
o
is
e
n
o
t
r
ep
r
esen
t
ed
in
th
e
s
im
u
latio
n
.
R
ain
f
all
r
is
k
was
s
im
u
lated
r
ath
er
th
an
b
u
ilt
f
r
o
m
s
tatio
n
-
lev
el
wea
th
er
,
r
o
ad
-
ac
ce
s
s
,
an
d
co
n
tr
ac
to
r
-
d
o
wn
tim
e
r
e
co
r
d
s
,
wh
ich
m
ay
o
v
er
s
tate
th
e
d
o
m
in
an
ce
o
f
s
ea
s
o
n
al
v
ar
iab
les.
Op
tim
izatio
n
was
test
ed
o
n
p
o
r
tf
o
lio
s
o
f
5
-
6
p
lan
n
in
g
b
lo
c
k
s
r
ath
er
th
an
f
u
ll
estate
s
ch
ed
u
l
es,
s
o
s
ca
lin
g
b
eh
av
io
u
r
n
ee
d
s
f
u
r
th
er
ass
ess
m
en
t.
Ultim
at
ely
,
th
e
d
ash
b
o
ar
d
was
ass
e
s
s
ed
v
ia
d
em
o
n
s
tr
ati
o
n
r
ath
e
r
th
an
lo
n
g
itu
d
in
al
d
ep
lo
y
m
en
t,
m
ea
n
in
g
ac
tu
al
a
d
o
p
tio
n
,
tr
u
s
t,
an
d
o
v
er
r
id
e
b
eh
a
v
io
u
r
r
em
ain
o
p
en
em
p
ir
ical
q
u
esti
o
n
s
.
5.
CO
NCLU
SI
O
N
I
n
th
is
p
a
p
er
,
we
in
tr
o
d
u
ce
d
DR
AF,
an
in
teg
r
ated
d
ec
is
io
n
-
in
tellig
en
ce
f
r
am
ewo
r
k
f
o
r
d
y
n
am
ic
co
n
tr
ac
to
r
allo
ca
tio
n
in
Z
im
b
ab
we
’
s
co
m
m
er
cial
f
o
r
estry
s
ec
to
r
.
T
h
e
t
h
r
ee
c
o
n
s
titu
e
n
ts
o
f
th
e
m
o
d
el
:
em
b
ed
d
in
g
r
is
k
-
p
r
o
b
ab
ilit
y
M
L
with
in
NSGA
-
I
I
o
b
jectiv
e
f
u
n
ctio
n
s
,
d
o
m
ain
-
ca
lib
r
ated
v
ir
tu
al
E
R
P
d
atase
t
an
d
an
ac
tiv
e
i
n
ter
ac
tiv
e
d
ec
is
io
n
s
u
p
p
o
r
t
s
y
s
tem
(
DSS
)
.
T
o
g
eth
er
,
th
ese
co
m
p
o
n
e
n
ts
ad
d
r
ess
th
e
d
ec
is
io
n
-
in
tellig
en
ce
g
a
p
th
at
co
n
v
en
ti
o
n
al
E
R
P
-
ce
n
tr
ic
allo
ca
tio
n
m
an
ag
em
en
t
co
u
l
d
n
o
t
o
v
er
c
o
m
e.
E
m
p
ir
ically
,
th
e
DR
AF
X
GB
o
o
s
t
clas
s
if
ier
ac
h
iev
ed
a
h
o
ld
-
o
u
t
R
OC
-
AUC
o
f
0
.
9
6
5
,
r
ec
all
o
f
0
.
9
9
9
,
an
d
a
0
.
3
6
5
R
OC
-
AUC
im
p
r
o
v
em
e
n
t
co
m
p
a
r
ed
with
th
e
lo
g
is
tic
r
eg
r
ess
io
n
b
aselin
e.
T
h
e
NSGA
-
I
I
o
p
tim
izer
g
en
er
ated
6
4
f
u
lly
f
ea
s
ib
le
Par
eto
s
o
lu
tio
n
s
in
b
o
th
th
e
b
alan
ce
d
an
d
h
ig
h
-
r
e
co
v
e
r
y
s
ce
n
ar
i
o
s
,
wh
er
ea
s
th
e
lo
w
-
r
is
k
s
ce
n
ar
io
r
ev
ea
led
a
u
s
ef
u
l
co
n
s
tr
ain
t
-
f
ea
s
ib
ilit
y
b
o
u
n
d
ar
y
f
o
r
m
an
a
g
er
s
.
Stak
eh
o
ld
er
ev
alu
atio
n
p
r
o
d
u
ce
d
an
o
v
er
all
s
atis
f
ac
tio
n
s
co
r
e
o
f
4
.
1
9
/5
.
0
,
an
d
SHAP
an
aly
s
is
s
h
o
wed
th
at
s
ea
s
o
n
al
r
ain
f
all
v
ar
ia
b
les
ac
co
u
n
ted
f
o
r
m
o
r
e
th
an
8
4
%
o
f
p
r
ed
ictiv
e
im
p
o
r
tan
ce
,
co
n
v
e
r
tin
g
m
o
d
el
in
ter
p
r
etatio
n
in
to
ac
tio
n
a
b
le
s
ch
ed
u
lin
g
g
u
i
d
an
ce
.
T
h
e
r
is
k
-
in
te
g
r
ated
MO
O
f
o
r
m
u
latio
n
p
r
o
p
o
s
ed
h
er
e
-
em
b
ed
d
in
g
ML
p
r
o
b
a
b
ilit
ies
d
ir
ec
tly
in
to
t
h
e
r
ec
o
v
er
y
d
is
co
u
n
t
f
ac
to
r
o
f
an
ev
o
l
u
t
io
n
ar
y
o
p
tim
i
zer
’
s
o
b
jectiv
e
f
u
n
ctio
n
-
r
ep
r
esen
ts
a
tr
an
s
f
er
ab
le
ar
ch
itectu
r
al
co
n
tr
ib
u
tio
n
ap
p
licab
le
to
r
eso
u
r
ce
-
allo
ca
tio
n
p
r
o
b
lem
s
in
v
o
lv
in
g
m
u
ltip
le
co
n
t
r
ac
to
r
s
with
s
to
ch
asti
c
o
p
er
atio
n
al
b
eh
av
i
o
u
r
.
Fu
tu
r
e
wo
r
k
s
h
o
u
ld
v
alid
ate
th
e
f
r
am
ewo
r
k
u
s
in
g
m
u
lti
-
s
ea
s
o
n
E
R
P,
r
ain
f
all,
an
d
co
n
tr
ac
to
r
-
p
er
f
o
r
m
an
ce
d
ata;
co
m
p
ar
e
NSGA
-
I
I
ag
ain
s
t
MO
E
A/D,
SP
E
A2
,
s
to
ch
asti
c
o
p
tim
izatio
n
,
r
ein
f
o
r
ce
m
e
n
t
-
lear
n
i
n
g
,
an
d
d
ee
p
tab
u
lar
r
is
k
m
o
d
els;
s
ca
le
o
p
tim
izatio
n
to
f
u
ll
estate
-
lev
el
p
o
r
tf
o
lio
s
;
an
d
test
s
u
b
g
r
o
u
p
f
air
n
ess
im
p
licatio
n
s
f
o
r
m
ic
r
o
,
s
m
all,
m
ed
i
u
m
,
an
d
lar
g
e
co
n
t
r
ac
to
r
s
.
ACK
NO
WL
E
DG
M
E
N
T
S
T
h
e
au
th
o
r
s
ac
k
n
o
wled
g
e
th
e
ac
ad
em
ic
g
u
id
an
ce
o
f
Mr
B
.
Mu
p
in
i,
th
e
Sch
o
o
l
o
f
I
n
f
o
r
m
atio
n
Scien
ce
an
d
T
ec
h
n
o
lo
g
y
at
Har
ar
e
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
,
an
d
th
e
p
r
ac
titi
o
n
e
r
s
wh
o
p
ar
ticip
ated
in
th
e
s
tak
eh
o
ld
er
ev
al
u
atio
n
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
t
r
ib
u
to
r
R
o
les
T
a
x
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
i
d
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
i
litate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
T
av
en
g
wa
No
r
m
an
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
B
r
ian
Mu
p
in
i
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
7
2
2
-
3
2
2
1
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
,
Vo
l.
7
,
No
.
3
,
No
v
em
b
er
20
26
:
3
3
7
-
345
344
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
Vir
tu
al
E
R
P
d
ataset
th
at
s
u
p
p
o
r
ts
th
e
f
in
d
in
g
s
o
f
th
i
s
s
tu
d
y
was
d
er
iv
ed
f
r
o
m
th
e
p
u
b
licly
av
ailab
le
USDA
Fo
r
est
Ser
v
ice
T
im
b
er
Har
v
ests
Featu
r
e
L
ay
er
.
T
h
e
s
y
n
th
etic
d
atase
t
an
d
th
e
an
aly
s
is
n
o
teb
o
o
k
ar
e
a
v
ail
ab
le
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
au
th
o
r
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
Ti
m
b
e
r
P
r
o
d
u
c
e
r
s Fe
d
e
r
a
t
i
o
n
,
An
n
u
a
l
T
i
m
b
e
r
S
e
c
t
o
r P
e
rf
o
rm
a
n
c
e
Re
p
o
r
t
2
0
2
4
.
H
a
r
a
r
e
:
TPF
Zi
m
b
a
b
w
e
,
2
0
2
4
.
[
2
]
A
f
r
i
c
a
n
F
o
r
e
st
F
o
r
u
m,
Fo
r
e
st
r
y
V
a
l
u
e
C
h
a
i
n
A
n
a
l
y
s
i
s
i
n
S
e
l
e
c
t
e
d
A
f
ri
c
a
n
C
o
u
n
t
ri
e
s
.
N
a
i
r
o
b
i
:
A
F
F
,
2
0
2
3
.
[
3
]
Zi
m
b
a
b
w
e
N
a
t
i
o
n
a
l
S
t
a
t
i
st
i
c
s
A
g
e
n
c
y
(
Zi
mS
t
a
t
)
,
“
Z
i
m
b
a
b
w
e
A
g
r
i
c
u
l
t
u
r
a
l
S
t
a
t
i
st
i
c
s
R
e
p
o
r
t
2
0
2
4
,
”
H
a
r
a
r
e
:
Zi
mS
t
a
t
,
2
0
2
4
.
[
4
]
“
Ti
m
b
e
r
p
r
o
d
u
c
t
i
o
n
r
e
b
o
u
n
d
s
a
mi
d
c
o
n
t
r
a
c
t
o
r
r
e
s
t
r
u
c
t
u
r
i
n
g
,
”
S
u
n
d
a
y
N
e
w
s
,
M
a
r
.
0
2
,
2
0
2
5
.
A
c
c
e
ss
e
d
:
F
e
b
.
1
9
,
2
0
2
6
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
w
w
w
.
s
u
n
d
a
y
n
e
w
s
.
c
o
.
z
w
/
t
i
mb
e
r
-
p
r
o
d
u
c
t
i
o
n
-
r
e
b
o
u
n
d
s
-
a
m
i
d
-
c
o
n
t
r
a
c
t
o
r
-
r
e
s
t
r
u
c
t
u
r
i
n
g
/
[
5
]
S
.
C
h
i
b
a
r
o
,
B
.
M
a
p
a
n
i
,
a
n
d
T.
D
u
b
e
,
“
P
r
e
d
i
c
t
i
v
e
a
n
a
l
y
t
i
c
s
i
n
Zi
m
b
a
b
w
e
a
n
f
o
r
e
s
t
r
y
o
p
e
r
a
t
i
o
n
s,”
Af
r
i
c
a
n
J
o
u
rn
a
l
o
f
F
o
res
t
ry
Re
se
a
rc
h
,
v
o
l
.
1
2
,
n
o
.
2
,
p
p
.
8
8
–
1
0
4
,
2
0
2
4
.
[
6
]
C
h
i
p
s
B
u
si
n
e
ss S
o
l
u
t
i
o
n
s,
ER
P
I
m
p
l
e
m
e
n
t
a
t
i
o
n
i
n
Z
i
m
b
a
b
w
e
’
s
T
i
m
b
e
r I
n
d
u
st
r
y
.
H
a
r
a
r
e
:
C
B
S
,
2
0
2
4
.
[
7
]
L.
Z
h
a
o
,
X
.
Li
,
a
n
d
Y
.
S
u
n
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
–
b
a
se
d
c
o
n
t
r
a
c
t
o
r
r
i
sk
a
ss
e
ss
men
t
i
n
i
n
d
u
s
t
r
i
a
l
s
u
p
p
l
y
c
h
a
i
n
s,”
Ex
p
e
rt
S
y
s
t
e
m
s
w
i
t
h
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
3
5
,
p
.
1
2
1
2
4
7
,
2
0
2
4
.
[
8
]
O
.
F
i
l
a
n
i
,
P
.
N
d
l
o
v
u
,
a
n
d
K
.
M
o
y
o
,
“
R
i
s
k
sc
o
r
i
n
g
f
r
a
m
e
w
o
r
k
s
i
n
e
m
e
r
g
i
n
g
m
a
r
k
e
t
s
u
p
p
l
y
c
h
a
i
n
s,”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
u
p
p
l
y
C
h
a
i
n
A
n
a
l
y
si
s
,
v
o
l
.
4
,
n
o
.
1
,
p
p
.
2
2
–
3
9
,
2
0
2
5
.
[
9
]
L.
G
r
i
n
sz
t
a
j
n
,
E.
O
y
a
l
l
o
n
,
a
n
d
G
.
V
a
r
o
q
u
a
u
x
,
“
W
h
y
d
o
t
r
e
e
-
b
a
se
d
m
o
d
e
l
s
s
t
i
l
l
o
u
t
p
e
r
f
o
r
m d
e
e
p
l
e
a
r
n
i
n
g
o
n
t
y
p
i
c
a
l
t
a
b
u
l
a
r
d
a
t
a
?
,
”
Ad
v
a
n
c
e
s i
n
N
e
u
r
a
l
I
n
f
o
rm
a
t
i
o
n
Pro
c
e
ssi
n
g
S
y
st
e
m
s
,
v
o
l
.
3
5
,
p
p
.
507
–
5
2
0
,
Ju
l
.
2
0
2
2
,
d
o
i
:
1
0
.
5
2
2
0
2
/
0
6
8
4
3
1
-
0
0
3
7
.
[
1
0
]
K
.
D
e
b
,
A
.
P
r
a
t
a
p
,
S
.
A
g
a
r
w
a
l
,
a
n
d
T.
M
e
y
a
r
i
v
a
n
,
“
A
f
a
st
a
n
d
e
l
i
t
i
s
t
mu
l
t
i
o
b
j
e
c
t
i
v
e
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m:
N
S
G
A
-
I
I
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
Ev
o
l
u
t
i
o
n
a
r
y
C
o
m
p
u
t
a
t
i
o
n
,
v
o
l
.
6
,
n
o
.
2
,
p
p
.
1
8
2
–
1
9
7
,
A
p
r
.
2
0
0
2
,
d
o
i
:
1
0
.
1
1
0
9
/
4
2
3
5
.
9
9
6
0
1
7
.
[
1
1
]
K
.
T.
M
a
l
l
a
d
i
a
n
d
T.
S
o
w
l
a
t
i
,
“
O
p
t
i
mi
z
a
t
i
o
n
i
n
f
o
r
e
st
su
p
p
l
y
c
h
a
i
n
ma
n
a
g
e
me
n
t
:
a
r
e
v
i
e
w
,
”
C
u
rr
e
n
t
Fo
r
e
st
r
y
R
e
p
o
rt
s
,
v
o
l
.
7
,
n
o
.
2
,
p
p
.
1
0
0
–
1
1
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
0
7
2
5
-
0
2
1
-
0
0
1
3
7
-
5.
[
1
2
]
C
.
S
.
Ta
n
g
,
“
P
e
r
s
p
e
c
t
i
v
e
s
i
n
s
u
p
p
l
y
c
h
a
i
n
r
i
s
k
m
a
n
a
g
e
me
n
t
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Pr
o
d
u
c
t
i
o
n
E
c
o
n
o
m
i
c
s
,
v
o
l
.
1
0
3
,
n
o
.
2
,
p
p
.
4
5
1
–
4
8
8
,
O
c
t
.
2
0
0
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
p
e
.
2
0
0
5
.
1
2
.
0
0
6
.
[
1
3
]
D
.
I
v
a
n
o
v
a
n
d
A
.
D
o
l
g
u
i
,
“
V
i
a
b
i
l
i
t
y
o
f
i
n
t
e
r
t
w
i
n
e
d
s
u
p
p
l
y
n
e
t
w
o
r
k
s:
e
x
t
e
n
d
i
n
g
t
h
e
su
p
p
l
y
c
h
a
i
n
r
e
si
l
i
e
n
c
e
a
n
g
l
e
s
t
o
w
a
r
d
s
su
r
v
i
v
a
b
i
l
i
t
y
.
A
p
o
si
t
i
o
n
p
a
p
e
r
mo
t
i
v
a
t
e
d
b
y
C
O
V
I
D
-
1
9
o
u
t
b
r
e
a
k
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Pro
d
u
c
t
i
o
n
R
e
se
a
rc
h
,
v
o
l
.
5
8
,
n
o
.
1
0
,
p
p
.
2
9
0
4
–
2
9
1
5
,
M
a
y
2
0
2
0
,
d
o
i
:
1
0
.
1
0
8
0
/
0
0
2
0
7
5
4
3
.
2
0
2
0
.
1
7
5
0
7
2
7
.
[
1
4
]
J.
R
i
x
,
B
.
W
o
l
f
f
,
a
n
d
F
.
M
ü
l
l
e
r
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
l
i
c
a
t
i
o
n
s
i
n
f
o
r
e
st
r
y
h
a
r
v
e
st
i
n
g
o
p
e
r
a
t
i
o
n
s:
a
sy
s
t
e
m
a
t
i
c
r
e
v
i
e
w
,
”
E
u
r
o
p
e
a
n
J
o
u
rn
a
l
o
f
Fo
r
e
st
Re
s
e
a
rc
h
,
v
o
l
.
1
4
2
,
n
o
.
4
,
p
p
.
7
7
1
–
7
9
0
,
2
0
2
3
.
[
1
5
]
A
.
R
.
H
e
v
n
e
r
,
S
.
T.
M
a
r
c
h
,
J
.
P
a
r
k
,
a
n
d
S
.
R
a
m
,
“
D
e
s
i
g
n
sc
i
e
n
c
e
i
n
i
n
f
o
r
mat
i
o
n
sy
s
t
e
ms
r
e
sea
r
c
h
,
”
MI
S
Q
u
a
rt
e
rl
y
:
M
a
n
a
g
e
m
e
n
t
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s
,
v
o
l
.
2
8
,
n
o
.
1
,
p
p
.
7
5
–
1
0
5
,
2
0
0
4
,
d
o
i
:
1
0
.
2
3
0
7
/
2
5
1
4
8
6
2
5
.
[
1
6
]
K
.
P
e
f
f
e
r
s,
T
.
T
u
u
n
a
n
e
n
,
M
.
A
.
R
o
t
h
e
n
b
e
r
g
e
r
,
a
n
d
S
.
C
h
a
t
t
e
r
j
e
e
,
“
A
d
e
s
i
g
n
sci
e
n
c
e
r
e
s
e
a
r
c
h
m
e
t
h
o
d
o
l
o
g
y
f
o
r
i
n
f
o
r
mat
i
o
n
sy
st
e
m
s
r
e
sea
r
c
h
,
”
J
o
u
rn
a
l
o
f
M
a
n
a
g
e
m
e
n
t
I
n
f
o
rm
a
t
i
o
n
S
y
st
e
m
s
,
v
o
l
.
2
4
,
n
o
.
3
,
p
p
.
4
5
–
7
7
,
2
0
0
7
,
d
o
i
:
1
0
.
2
7
5
3
/
M
I
S
0
7
4
2
-
1
2
2
2
2
4
0
3
0
2
.
[
1
7
]
L.
B
r
e
i
m
a
n
,
“
R
a
n
d
o
m fo
r
e
st
s,
”
Ma
c
h
i
n
e
L
e
a
r
n
i
n
g
,
v
o
l
.
4
5
,
n
o
.
1
,
p
p
.
5
–
3
2
,
O
c
t
.
2
0
0
1
,
d
o
i
:
1
0
.
1
0
2
3
/
A
:
1
0
1
0
9
3
3
4
0
4
3
2
4
.
[
1
8
]
T.
C
h
e
n
a
n
d
C
.
G
u
e
s
t
r
i
n
,
“
X
G
B
o
o
s
t
:
A
sca
l
a
b
l
e
t
r
e
e
b
o
o
st
i
n
g
sy
s
t
e
m
,
”
i
n
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
AC
M
S
I
G
K
D
D
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
K
n
o
w
l
e
d
g
e
D
i
s
c
o
v
e
r
y
a
n
d
D
a
t
a
Mi
n
i
n
g
,
A
u
g
.
2
0
1
6
,
p
p
.
7
8
5
–
794
,
d
o
i
:
1
0
.
1
1
4
5
/
2
9
3
9
6
7
2
.
2
9
3
9
7
8
5
.
[
1
9
]
M
.
A
l
M
n
a
see
r
,
M
.
A
l
s
h
u
r
i
d
e
h
,
a
n
d
S
.
A
l
-
H
a
w
a
r
y
,
“
H
y
b
r
i
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
f
o
r
c
o
n
t
r
a
c
t
o
r
p
e
r
f
o
r
m
a
n
c
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
B
u
si
n
e
ss
A
n
a
l
y
t
i
c
s
,
v
o
l
.
8
,
n
o
.
1
,
p
p
.
4
5
–
6
3
,
2
0
2
5
.
[
2
0
]
R
.
S
h
w
a
r
t
z
-
Z
i
v
a
n
d
A
.
A
r
m
o
n
,
“
T
a
b
u
l
a
r
d
a
t
a
:
D
e
e
p
l
e
a
r
n
i
n
g
i
s
n
o
t
a
l
l
y
o
u
n
e
e
d
,
”
I
n
f
o
rm
a
t
i
o
n
F
u
s
i
o
n
,
v
o
l
.
8
1
,
p
p
.
8
4
–
9
0
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
f
f
u
s.
2
0
2
1
.
1
1
.
0
1
1
.
[
2
1
]
M
.
T.
M
.
Emm
e
r
i
c
h
a
n
d
A
.
H
.
D
e
u
t
z
,
“
A
t
u
t
o
r
i
a
l
o
n
m
u
l
t
i
o
b
j
e
c
t
i
v
e
o
p
t
i
mi
z
a
t
i
o
n
:
f
u
n
d
a
me
n
t
a
l
s
a
n
d
e
v
o
l
u
t
i
o
n
a
r
y
me
t
h
o
d
s,
”
N
a
t
u
r
a
l
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
7
,
n
o
.
3
,
p
p
.
5
8
5
–
6
0
9
,
S
e
p
.
2
0
1
8
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
7
-
0
1
8
-
9
6
8
5
-
y.
[
2
2
]
P
.
A
u
g
e
r
,
L
.
T.
B
u
i
,
a
n
d
M
.
H
.
N
g
u
y
e
n
,
“
A
d
v
a
n
c
e
s
i
n
mu
l
t
i
-
o
b
j
e
c
t
i
v
e
e
v
o
l
u
t
i
o
n
a
r
y
a
l
g
o
r
i
t
h
ms
f
o
r
s
u
p
p
l
y
c
h
a
i
n
o
p
t
i
mi
z
a
t
i
o
n
,
”
C
o
m
p
u
t
e
rs
& O
p
e
r
a
t
i
o
n
s
Re
s
e
a
r
c
h
,
v
o
l
.
1
5
0
,
p
.
1
0
6
0
8
2
,
2
0
2
3
.
[
2
3
]
M
.
A
c
u
n
a
,
J.
S
k
i
n
n
e
l
l
,
T
.
Ev
a
n
s
o
n
,
a
n
d
R
.
M
i
t
c
h
e
l
l
,
“
H
a
r
v
e
s
t
i
n
g
p
r
o
d
u
c
t
i
v
i
t
y
a
n
d
c
o
st
a
n
a
l
y
si
s
i
n
f
o
r
e
s
t
r
y
o
p
e
r
a
t
i
o
n
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
F
o
res
t
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
3
2
,
n
o
.
3
,
p
p
.
2
0
2
–
2
1
5
,
2
0
2
1
.
[
2
4
]
X
.
L
i
u
,
Q
.
C
h
e
n
,
a
n
d
F
.
W
a
n
g
,
“
M
u
l
t
i
-
o
b
j
e
c
t
i
v
e
c
o
n
st
r
u
c
t
i
o
n
s
c
h
e
d
u
l
i
n
g
u
n
d
e
r
r
e
s
o
u
r
c
e
c
o
n
s
t
r
a
i
n
t
s,”
A
u
t
o
m
a
t
i
o
n
i
n
C
o
n
st
r
u
c
t
i
o
n
,
v
o
l
.
1
4
5
,
p
.
1
0
4
6
1
7
,
2
0
2
3
.
[
2
5
]
K
.
P
a
t
e
j
u
k
,
A
.
B
a
t
u
r
o
-
C
i
e
śn
i
e
w
s
k
a
,
W
.
P
u
s
z
,
a
n
d
A
.
K
a
c
z
m
a
r
e
k
-
P
i
e
ń
c
z
e
w
sk
a
,
“
B
i
s
c
o
g
n
i
a
u
x
i
a
c
h
a
r
c
o
a
l
c
a
n
k
e
r
—
a
n
e
w
p
o
t
e
n
t
i
a
l
t
h
r
e
a
t
f
o
r
M
i
d
-
E
u
r
o
p
e
a
n
f
o
r
e
s
t
s
a
s a
n
e
f
f
e
c
t
o
f
c
l
i
ma
t
e
c
h
a
n
g
e
,
”
F
o
res
t
s
,
v
o
l
.
1
3
,
n
o
.
1
,
p
.
8
9
,
Ja
n
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
f
1
3
0
1
0
0
8
9
.
[
2
6
]
C
.
W
.
H
o
l
sa
p
p
l
e
,
A
.
Le
e
-
P
o
s
t
,
a
n
d
R
.
P
a
k
a
t
h
,
“
A
u
n
i
f
i
e
d
f
o
u
n
d
a
t
i
o
n
f
o
r
d
e
c
i
si
o
n
su
p
p
o
r
t
s
y
s
t
e
ms,
”
D
e
c
i
si
o
n
S
u
p
p
o
r
t
S
y
st
e
m
s
,
v
o
l
.
6
4
,
p
p
.
7
2
1
–
7
2
8
,
2
0
1
4
.
[
2
7
]
T.
M
a
x
i
m
e
,
G
.
K
a
mb
a
n
j
e
,
a
n
d
B
.
N
y
o
n
i
,
“
S
u
p
p
l
y
c
h
a
i
n
o
p
t
i
mi
z
a
t
i
o
n
i
n
S
o
u
t
h
e
r
n
A
f
r
i
c
a
n
f
o
r
e
s
t
r
y
se
c
t
o
r
s,”
T
h
e
A
f
ri
c
a
n
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
,
v
o
l
.
9
,
n
o
.
3
,
p
p
.
1
0
1
–
1
1
9
,
2
0
2
3
.
[
2
8
]
M
.
H
a
d
d
a
r
a
a
n
d
A
.
E
l
r
a
g
a
l
,
“
T
h
e
r
e
a
d
i
n
e
ss
o
f
ER
P
s
y
st
e
ms
f
o
r
t
h
e
f
a
c
t
o
r
y
o
f
t
h
e
f
u
t
u
r
e
,
”
P
ro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
6
4
,
p
p
.
7
2
1
–
7
2
8
,
2
0
1
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.2
0
1
5
.
0
8
.
5
9
8
.
[
2
9
]
A
.
R
a
j
,
G
.
D
w
i
v
e
d
i
,
A
.
S
h
a
r
ma
,
A
.
B
.
L.
d
e
S
.
Ja
b
b
o
u
r
,
a
n
d
S
.
R
a
j
a
k
,
“
B
a
r
r
i
e
r
s t
o
t
h
e
a
d
o
p
t
i
o
n
o
f
I
n
d
u
st
r
y
4
.
0
t
e
c
h
n
o
l
o
g
i
e
s
i
n
t
h
e
man
u
f
a
c
t
u
r
i
n
g
se
c
t
o
r
:
a
n
i
n
t
e
r
-
c
o
u
n
t
r
y
c
o
m
p
a
r
a
t
i
v
e
p
e
r
s
p
e
c
t
i
v
e
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
Pr
o
d
u
c
t
i
o
n
E
c
o
n
o
m
i
c
s
,
v
o
l
.
2
2
4
,
p.
1
0
7
5
4
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
p
e
.
2
0
1
9
.
1
0
7
5
4
6
.
[
3
0
]
T.
F
a
w
c
e
t
t
,
“
A
n
i
n
t
r
o
d
u
c
t
i
o
n
t
o
R
O
C
a
n
a
l
y
s
i
s,”
P
a
t
t
e
r
n
R
e
c
o
g
n
i
t
i
o
n
L
e
t
t
e
rs
,
v
o
l
.
2
7
,
n
o
.
8
,
p
p
.
8
6
1
–
8
7
4
,
Ju
n
.
2
0
0
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
a
t
r
e
c
.
2
0
0
5
.
1
0
.
0
1
0
.
[
3
1
]
S
G
S
,
“
F
o
r
e
st
ma
n
a
g
e
m
e
n
t
c
e
r
t
i
f
i
c
a
t
i
o
n
r
e
p
o
r
t
:
Zi
mb
a
b
w
e
o
p
e
r
a
t
i
o
n
s
,
”
2
0
2
1
.
[
3
2
]
S
.
H
o
sse
i
n
i
,
N
.
T
a
j
i
k
,
D
.
I
v
a
n
o
v
,
M
.
D
.
S
a
r
d
e
r
,
K
.
B
a
r
k
e
r
,
a
n
d
A
.
A
l
K
h
a
l
e
d
,
“
R
e
s
i
l
i
e
n
t
s
u
p
p
l
i
e
r
sel
e
c
t
i
o
n
a
n
d
o
p
t
i
ma
l
o
r
d
e
r
a
l
l
o
c
a
t
i
o
n
u
n
d
e
r
d
i
sr
u
p
t
i
o
n
r
i
s
k
s,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Pro
d
u
c
t
i
o
n
Ec
o
n
o
m
i
c
s
,
v
o
l
.
2
1
3
,
p
p
.
1
2
4
–
1
3
7
,
J
u
l
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
p
e
.
2
0
1
9
.
0
3
.
0
1
8
.
[
3
3
]
A
.
L.
B
o
u
l
e
st
e
i
x
,
S
.
L
a
u
e
r
,
a
n
d
M
.
J
.
A
.
E
u
g
s
t
e
r
,
“
A
p
l
e
a
f
o
r
n
e
u
t
r
a
l
c
o
m
p
a
r
i
so
n
st
u
d
i
e
s
i
n
c
o
mp
u
t
a
t
i
o
n
a
l
sci
e
n
c
e
s,
”
PL
o
S
O
N
E
,
v
o
l
.
8
,
n
o
.
4
,
p
.
e
6
1
5
6
2
,
A
p
r
.
2
0
1
3
,
d
o
i
:
1
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
o
n
e
.
0
0
6
1
5
6
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
C
o
m
p
u
t Sci
I
n
f
T
ec
h
n
o
l
I
SS
N:
2722
-
3
2
2
1
R
is
k
-
in
teg
r
a
ted
co
n
tr
a
cto
r
a
ll
o
ca
tio
n
in
Zimb
a
b
w
e'
s
timb
er
va
lu
e
ch
a
i
n
(
Ta
ve
n
g
w
a
N
o
r
ma
n
)
345
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
Ta
v
e
n
g
w
a
No
r
m
a
n
is
a
m
u
lt
id
isc
i
p
li
n
a
r
y
a
c
a
d
e
m
ic
a
n
d
re
se
a
rc
h
e
r
with
q
u
a
li
fica
ti
o
n
s
sp
a
n
n
i
n
g
d
a
ta
sc
ien
c
e
,
fo
re
n
sic
a
u
d
it
in
g
,
c
o
rp
o
ra
te
g
o
v
e
rn
a
n
c
e
,
p
e
a
c
e
a
n
d
c
o
n
fli
c
t
stu
d
ies
,
a
n
d
se
c
u
rit
y
st
u
d
ies
.
He
c
o
m
p
lete
d
a
M
a
ste
r
o
f
Tec
h
n
o
l
o
g
y
in
Da
ta
S
c
ien
c
e
a
n
d
An
a
ly
t
ics
a
t
th
e
H
a
ra
re
In
s
ti
tu
te
o
f
Tec
h
n
o
l
o
g
y
a
n
d
h
o
l
d
s
a
M
a
ste
r
o
f
Co
m
m
e
rc
e
in
F
o
re
n
sic
Au
d
it
i
n
g
a
n
d
Ac
c
o
u
n
ti
n
g
,
a
M
a
ste
r
o
f
S
c
ien
c
e
i
n
P
e
a
c
e
,
Lea
d
e
rsh
ip
,
a
n
d
Co
n
fli
c
t
Re
so
lu
ti
o
n
,
a
P
o
st
g
ra
d
u
a
te
Di
p
lo
m
a
in
Ap
p
li
e
d
Co
r
p
o
ra
te
G
o
v
e
rn
a
n
c
e
a
n
d
S
trate
g
ic
Lea
d
e
rsh
ip
,
a
n
d
a
Ba
c
h
e
l
o
r
o
f
Bu
sin
e
ss
Ad
m
in
istrati
o
n
(Ho
n
o
u
rs)
in
P
o
li
c
e
a
n
d
S
e
c
u
ri
ty
S
tu
d
ies
.
He
is
a
lso
p
u
rsu
in
g
a
B
a
c
h
e
lo
r
o
f
S
u
b
sta
n
ti
v
e
Law
s
a
t
t
h
e
Un
iv
e
rsit
y
o
f
Zi
m
b
a
b
we
.
His
re
se
a
rc
h
in
tere
sts
i
n
c
lu
d
e
a
rti
ficia
l
in
tell
ig
e
n
c
e
,
d
a
ta
a
n
a
ly
ti
c
s,
e
n
terp
rise
risk
m
a
n
a
g
e
m
e
n
t
a
n
d
p
u
b
li
c
-
se
c
to
r
g
o
v
e
rn
a
n
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
h
2
4
0
7
4
2
c
@h
it
.
a
c
.
z
w.
Br
ia
n
Mu
p
in
i
is
c
u
rre
n
tl
y
wo
rk
i
n
g
a
s
a
l
e
c
tu
re
r
in
th
e
Co
m
p
u
ter
S
c
ien
c
e
De
p
a
rtme
n
t
a
t
t
h
e
Ha
ra
re
In
stit
u
t
e
o
f
Tec
h
n
o
lo
g
y
.
He
re
c
e
iv
e
d
h
is
first
d
e
g
re
e
,
a
Ba
c
h
e
l
o
r
o
f
Tec
h
n
o
l
o
g
y
in
Co
m
p
u
ter
S
c
ien
c
e
,
fro
m
th
e
sa
m
e
in
stit
u
te
in
2
0
1
7
.
He
a
lso
h
o
ld
s
a
M
a
ste
r
o
f
Tec
h
n
o
l
o
g
y
in
C
lo
u
d
C
o
m
p
u
t
in
g
fro
m
th
e
Ha
ra
re
In
stit
u
te
o
f
Tec
h
n
o
l
o
g
y
,
a
tt
a
i
n
e
d
in
2
0
2
4
.
He
is
a
lso
a
re
s
e
a
rc
h
e
r
in
th
e
c
o
m
p
u
ter
sc
ien
c
e
field
.
His
m
a
in
re
se
a
rc
h
in
tere
sts
fo
c
u
s
o
n
n
a
tu
ra
l
lan
g
u
a
g
e
p
ro
c
e
ss
in
g
(N
LP
),
a
rti
ficia
l
i
n
telli
g
e
n
c
e
(AI),
c
ry
p
to
g
ra
p
h
y
a
n
d
se
c
u
rit
y
sy
ste
m
s,
c
lo
u
d
c
o
m
p
u
ti
n
g
,
i
n
tern
e
t
o
f
th
i
n
g
s
(Io
T),
a
n
d
d
a
ta
m
i
n
i
n
g
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
b
m
u
p
i
n
i@h
i
t.
a
c
.
z
w.
Evaluation Warning : The document was created with Spire.PDF for Python.