Indonesi
an
Journa
l
of El
ect
ri
cal Engineer
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
9
, No
.
1
,
J
an
ua
ry
201
8
, p
p.
157
~
163
IS
S
N:
25
02
-
4752
,
DOI: 10
.11
591/
ijeecs
.
v9.i
1
.
pp
157
-
163
157
Journ
al h
om
e
page
:
http:
//
ia
es
core.c
om/j
ourn
als/i
ndex.
ph
p/ij
eecs
Investig
ating Th
er
m
al C
omfort f
or the Cl
assroom
Envir
onment
usin
g I
oT
Nu
rsh
ah
ri
l
y
I
dura R
amli
*
, Mo
hd
Iz
an
i
Moham
ed
R
aw
i, Ah
m
ad
Z
ahi
d Hij
az
i,
Abdul
lah
Hay
yan
Kunj
i
Moh
am
med
Univer
siti
Te
kno
logi
MA
RA Mal
a
y
si
a
,
40450
Sha
h
Alam,
Se
la
ngo
r,
Mal
a
y
s
ia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
A
ug
30
, 201
7
Re
vised
N
ov
2
4
, 2
01
7
Accepte
d
Dec
11
, 201
7
In
thi
s
m
oder
n
ce
ntur
y
wh
ere
fine
comfort
is
a
nec
essit
y
e
spec
iall
y
i
n
buil
dings
and
o
cc
upi
ed
spac
e,
the
stud
y
to
sa
t
isf
y
on
e
aspe
ct
of
hum
an
comfort
is
a
m
ust.
Th
is
stud
y
e
ncompass
es
of
expl
oring
the
p
h
y
siolog
ic
a
l
and
envi
ronm
en
ta
l
fa
ct
ors
in
a
c
hie
ving
th
ermal
comfort
which
spec
ifica
l
l
y
conside
ring
th
e
cl
othi
ng
insula
t
i
on
and
m
et
aboli
c
ra
t
e
of
students
as
well
as
the
dep
lo
y
m
en
t of dr
y
-
bu
lb
t
empera
tur
e,
m
e
an
ra
dia
nt te
m
per
a
tur
e,
hum
idit
y
,
and
ai
r
m
ovement
in
orde
r
to
obta
in
the
le
v
el
of
comfort
student
s
ar
e
expe
ri
enc
ing
in
cl
ass.
Th
e
l
eve
l
of
comfort
are
d
et
e
ct
ed
b
y
using
AS
HRA
E
55
to
determ
ine
the
av
era
g
e
the
r
m
al
sensati
on
re
spons
e
through
the
Predicte
d
Mea
n
Vote
(PM
V)
val
ue.
An
andr
oid
appl
i
cation
were
dev
eloped
to
re
a
d
input
of
r
ec
ogn
iz
ing
cl
oth
ing
l
eve
l
(thickne
ss
of
cl
oth
ing)
an
d
ca
p
turi
ng
m
et
abol
i
c
ra
t
e
t
o
ca
t
er
the
inp
uts
for
ph
y
siolo
gic
a
l
fa
c
tors,
while
ra
d
ia
n
t
te
m
per
at
ur
e,
hu
m
idi
t
y
and
a
ir
m
ovement
are
ca
pture
d
through
stat
ic
sensors
set
up
in
the
cl
as
sroom
spac
e.
Th
is
pape
r
ana
l
y
ses
both
the
ph
y
siol
ogic
a
l
and
envi
ronm
ent
a
l
fa
ct
ors
in
aff
ec
t
i
ng
student
s
in
cl
ass
and
further
det
ermine
the
ir
comfort
levels
that
is
a
m
aj
or
inf
lue
n
ci
ng
fa
ct
or
of
foc
us
in
learni
ng
.
Through
cro
ss
r
efe
re
n
ci
ng
colle
ct
ed
data
from
I
oT
en
abl
ed
node
s,
it
is
found
tha
t
both
ph
y
si
o
logi
c
al
and
env
i
ronm
ent
al
fa
ct
or
s,
and
the
combinat
ion
of
the
m
gre
at
l
y
influence
in
g
et
t
ing
the
m
ost
comfor
ta
bl
e
stat
e
wi
th
PM
V
val
ue
of
0
.
Ke
yw
or
d
s
:
En
vironm
ental
facto
rs
Hu
m
an
physi
ol
og
y
facto
rs
In
te
r
net
of Th
i
ng
s
(IoT
)
Ther
m
al
co
m
f
or
t
Copyright
©
201
8
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Nurs
hahril
y
Idur
a
Ra
m
li
,
Un
i
ver
sit
i Te
knol
og
i M
ARA
Ma
la
ysi
a
,
4045
0 S
hah A
l
a
m
, S
el
ango
r,
Ma
la
ysi
a
.
Em
a
il
:
idu
ra@
t
m
sk
.u
it
m
.ed
u.m
y
1.
INTROD
U
CTION
Ph
ysi
ol
og
ic
al
and
en
vir
onm
e
ntal
facto
rs
pla
ys
a
vital
aspec
t
in
al
lowi
ng
hum
ans
to
achie
ve
c
om
fo
rt.
Pr
ope
r
li
gh
t,
ai
r,
the
rm
al
,
and
acou
sti
c
m
akes
up
m
ajo
r
bas
ic
par
ts
of
hu
m
an
com
fo
rt.
Com
fo
rt
is
defi
ned
as
the
abse
nce
of
discom
fo
rt.
T
her
e
are
five
m
ai
n
factor
s
that
i
nf
lue
nces
hu
m
an
com
f
or
t,
w
hic
h
are
visu
al
com
fo
rt,
aco
ust
ic
al
co
m
fo
rt,
therm
al
co
m
fo
rt,
in
door
ai
r
com
fo
rt,
and
sp
at
ia
l
co
m
fo
rt
[1
]
,
[2
]
.
T
he
rm
a
l
com
fo
rt is
on
e
of the
view
in a
chievin
g h
uma
n
c
om
fo
rt [3]
-
[5
]
.
In
order
t
o
ach
ie
ve
therm
al
hu
m
an
com
fo
rt
for
s
tu
den
ts
in
cl
assroom
s,
reg
ulati
ons
s
uch
as
Stand
a
r
d
55
-
2004,
T
he
r
m
al
Env
ir
on
m
ental
Co
nd
it
io
ns
for
Hu
m
an
Occ
up
a
ncy,
by
Am
erican
So
ci
et
y
of
He
at
ing
,
Re
fr
ige
rati
ng,
and
Air
-
Co
nd
it
io
ning
E
nginee
rs,
ASH
RAE
Am
erica
n
Nati
onal
Stand
a
r
ds
I
nst
it
ute,
ANSI
dicta
te
s
th
at
prop
e
r
the
rm
al
co
m
fo
rt
in
occ
up
ie
d
sp
a
ce
m
us
t
be
adhered
[
3].
H
owe
ver,
curre
nt
therm
al
regulat
or
i
ns
tr
um
ents
do
es
not
f
ully
incorp
or
at
es
al
l
aspe
ct
s
that
has
to
be
co
ns
i
der
e
d
wh
ic
h
co
ntri
bute
to
therm
al
co
m
fort.
As
a
res
ult,
therm
al
co
m
fort f
or all
o
cc
up
ants in
the clas
sroom
is n
ot sa
ti
sfied.
This
st
ud
y
is
co
nducte
d
in
a
cl
assr
oom
of
Fac
ulty
of
Com
pu
te
r
a
nd
Ma
them
at
ical
Scie
nces
,
Un
i
ver
sit
i
Te
knol
og
i
MARA
(
UiTM)
Sh
a
h
Alam
,
a
sm
all
le
arn
in
g
en
vi
ronm
ent
with
ai
r
co
ndit
ion
i
ng
as
a
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vol
.
9
,
No.
1
,
Jan
ua
ry
201
8
:
157
–
163
158
te
st
bed
.
To
re
al
iz
e
the
ai
m
of
this
researc
h,
an
im
ple
m
ent
at
ion
of
t
he
I
oT
fr
am
ework
wh
ic
h
c
on
sist
s
of
sta
ti
c
node
t
o
be
ins
ta
ll
ed
in
the
te
sti
ng
sit
e.
The
node
will
be
equ
i
pp
e
d
with
m
ult
iple
sens
or
s
to
ac
quire
al
l
the
values
of
t
he
e
nv
i
ronm
ental
factor
s.
T
his
f
r
a
m
ewo
r
k
will
al
so
b
e
est
a
blishe
d
to
f
un
ct
i
on
an
d
c
omm
un
ic
at
e
with
m
ob
il
e
node
f
ram
ewo
r
k
w
hic
h
in
fer
s
the
val
ues
of
ph
ysi
ologica
l
facto
rs
ob
ta
in
ed.
Fi
nally
,
al
l
these
values
a
re
use
d
in
cal
culat
ing
the
Pr
e
dicte
d
Me
an
V
ote
(PM
V)
to
s
how
t
he
le
vel
of
c
om
fo
rt
in
the
cl
assroom
,
h
ence
f
or
th
intr
oducin
g
a
n
e
w and en
ha
nced
way of
regulat
ing t
em
per
at
ur
e
in
the
classr
oo
m
u
tili
zi
ng
the
Io
T
.
2.
RESEA
R
CH MET
HO
D
This
stu
dy
em
br
aces
t
he
ad
va
ncem
ent
in
IoT
in
achievi
ng
the
accum
ulatio
n
of
data
as
in
m
any
o
ther
researc
h
areas
su
c
h
as
[6
]
-
[8
]
.
Partic
ularly
in
this
stud
y,
th
e
accum
ulati
on
of
data
f
ro
m
e
m
bed
ded
se
nsor
ar
e
of
the
purpose
to
cal
culat
e
the
PMV
value
.
The
rati
onal
e
of
usi
ng
IoT
is
to
i
m
ple
m
ent
netw
ork
te
ch
nolo
gies
wh
ic
h
inclu
des
the
us
e
of
hi
gh
te
ch
no
l
og
y
s
ens
or
s
t
o
aut
om
at
e
the
pr
oce
ss
of
data
c
ollec
ti
on
in
a
l
oc
at
ion
,
con
te
xtu
al
ly
a
s
m
al
l
le
arn
in
g
en
vir
on
m
ent
and
aut
om
at
i
cal
ly
gen
erate
the
PMV
val
ue
us
i
ng
al
gor
it
h
m
s
pro
gr
am
m
ed
in the fram
ewo
r
k [
8].
This im
ple
m
entat
ion
w
il
l al
lo
w
a
faster
d
at
a
colle
ct
ion
,
proc
essing, a
nd g
e
ner
at
io
n
of r
es
ults.
2.1.
Predi
cte
d
M
e
an
V
ot
e
(
P
MV)
The
PMV
is
a
value
on
a
Fanger
Se
ve
n
Po
i
nt
Scal
e
[11]
,
[
10
]
or
cal
le
d
therm
al
sensatio
n
scal
e
[
3]
that
co
ns
ist
s
of
seve
n
val
ues
f
ro
m
Cold
(
-
3)
t
o
H
ot
(
3)
[
12
]
.
Figure
1
de
pict
s
the
the
rm
al
s
ensati
on
scal
e
[10].
ASHRAE
55
reco
m
m
end
s
that
the
acce
pt
able
therm
al
com
fo
rt
ra
ng
e
for
interi
or
oc
cup
ie
d
spa
ce
is
in
betwee
n
-
0.5
a
nd
+
0.5
(
-
0.5
¡
PMV
¡
+
0.5)
[3
]
.
T
he
PMV
is
ob
ta
ine
d
by
equ
at
io
ns
pro
du
ce
d
by
Fang
er
in
1970
[11].
Figure
1. Se
ve
n
P
oin
t
Scal
e /
Ther
m
al
Sen
sa
ti
on
Scale
Wh
e
re,
M
Me
ta
bo
li
c rate,
W
/m
2
W
Exter
nal wo
rk,
W
/m
2
(ass
ume
d
ze
ro for
a
lm
os
t al
l o
f
th
e ac
ti
viti
es)
Pa
Partia
l water
vapou
r
pr
ess
ure,
Pa f
cl
Ra
ti
o
of clot
he
d
a
nd nu
de
s
urface are
a
of
t
he
bod
y
ta
Air
te
m
per
at
ure,
◦
C
tr
Me
an
Ra
dia
nt
tem
per
at
ur
e
,
◦
C
hc
Coef
fici
ent of
convecti
ve
h
ea
t t
ran
s
fer
,
W
/ (
m
2
*
K
)
t
cl
Cl
oth
in
g
s
urfa
ce tem
per
at
ure
,
◦
C
Icl
Cl
oth
in
g
the
rm
al
insu
la
ti
on, m
2
*
K
/
W
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
In
vest
ig
atin
g
T
hermal C
om
f
ort
for
the
Cl
as
sr
oom Envir
onm
ent usi
ng IoT
(N
ur
sha
hr
il
y I
dura R
am
li
)
159
IS
O
77
30
(
2005)
dicta
te
s
th
at
wh
il
e
this
m
od
el
is
der
iv
ed
f
or
a
m
or
e
sta
bili
zed
c
onditi
on
s
[
13
]
,
the
desta
bili
zat
ion
of
one
or
m
or
e
var
ia
bles
can
be
fi
xed
with
a
n
ap
pro
xim
atio
n
us
in
g
aver
a
ge
values
of
t
he
pr
e
vious
hour
[12]
[
2]
em
phasi
ze
that
PM
V
value
cl
os
es
t
to
zer
o
(
neu
t
ral)
in
dicat
e
be
tt
er
therm
al
com
fo
rt
wh
il
e
zer
o
PM
V
val
ue
sho
ws
that
the
m
os
t
t
her
m
al
co
m
fo
r
t
is
fu
ll
y
achieved
.
Fig
ure
2
[
2]
dep
ic
ts
the
f
act
or
s
involve
d
in
obt
ai
nin
g t
he
rm
al
com
fo
rt and t
he
r
el
at
ion
s
hi
p of PMV
on t
he
ther
m
al
sen
sat
ion
scal
e.
Figure
2. PMV
on th
e
the
rm
al
sensati
on
scal
e [
2]
3.
ENVIRO
N
M
ENTAL F
A
C
TORS INF
O
RMATIO
N G
ATHERI
NG (
STATIC
NO
DES)
A
te
st
bed
will
be
set
up
in
a
s
m
al
l
cl
assro
om
of
abo
ut
10
x
10
m
et
ers.
Each
sens
orp
acked
node
s
will
colle
ct
qu
antit
at
ive
data
in
resp
ect
of
each
sens
ors
and
will
send
the
colle
ct
ed
da
ta
to
the
m
ain
node
(sink)
us
i
ng
wireless
c
omm
un
ic
at
ion
w
her
e
the
sin
k
w
il
l
cal
culat
e
the
PMV
va
lue
us
in
g
pro
gr
am
m
ed
al
gorithm
and
disp
la
y
the
va
lue
on
the
di
sp
la
y.
Stat
ic
nodes
a
re
ass
e
m
bled
with
Ardu
i
no
Na
no
V3.1,
SH
T
21
/
HT
U21
te
m
per
at
ur
e
and
hum
idity
sens
or
,
LM
35
tem
per
at
ur
e
se
ns
or,
HC
-
06
with
JY
-
MC
U
carri
e
r
Bl
uetoo
t
h
a
da
pte
r,
16
x2
1602A
V
2.0
LCD
,
3.3
V
5V
Brea
dboa
rd
P
ow
e
r
Supp
ly
,
an
d
1K
resist
or
wh
i
ch
a
re
at
ta
ched
to
a
br
ea
dboa
rd
f
or
de
velo
pm
ent
pur
po
ses
.
T
he
hard
war
e
pro
gram
m
ing
is
do
ne
th
rou
gh
A
r
du
i
no
so
ft
war
e
v1.
0.6
-
a
softwa
re
for
c
onfig
uri
ng
A
rduin
o
base
d
hard
war
e
s
uc
h
t
he
Ard
uino
Nano.
All
t
he
s
ens
ors
are
cal
ibr
at
ed
us
in
g
se
rial
co
nn
ect
io
n
on
9600
ba
ud
f
or
a
sta
ble
se
rial
transm
issi
on
and
hi
gh
cal
ib
rati
on
accuracy.
Fig
ure
3
il
lustrate
s
(a)
the
de
ploym
ent
of
sta
ti
c
node
s
(
oran
ge
half
cy
li
nd
er)
on
the
cei
li
ng
a
nd
m
ai
n
nod
e/
sin
k (
or
a
ng
e
r
ect
a
ngle
)
i
n
the
test
bed an
d (
b)
shows t
he
actual
assem
bly of
a s
ta
ti
c n
ode.
Figure
3. (a
)
T
est
b
e
d
la
yo
ut
and (
b) stat
ic
node
co
m
ponen
t
s
In
e
nsuri
ng
th
e
accu
racy
of
sens
or
s
,
the
outp
uts
are
com
par
e
d
with
Wi
ndows
Weathe
r
ser
vice
as
they
pro
vid
e
a
highly
accu
rate
rea
ding
of
e
nvir
on
m
ent
temperat
ur
es
o
f
a
n
area.
I
n
c
om
par
iso
n,
t
he
DBT
an
d
RH
rea
d
by
S
HT
21
/
HTU2
1
and
LM3
5
are
si
m
il
ar
with
w
hat
rea
d
by
Wi
ndows
Weathe
r
se
rv
ic
e
a
nd
he
nce,
sh
ows
acc
ur
at
e
cal
ibrati
on
a
nd
will
pro
vid
e
accurate
value
s
to
be
us
ed
i
n
cal
culat
ing
PM
V
val
ue.
I
n
a
ddit
ion
,
the
cal
ibra
ti
on
process
is
c
on
du
ct
e
d
in
a
su
i
ta
ble
en
vironme
nt
that
di
d
not
blo
c
k
or
inte
rfere
with
a
ny
na
tural
ou
t
door tem
per
at
ur
e s
uc
h
as
close
d
(
no air
f
lo
w)
or air
-
co
ndit
ion
ed
ro
om
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vol
.
9
,
No.
1
,
Jan
ua
ry
201
8
:
157
–
163
160
4.
PHY
SIOLO
G
ICA
L
FACT
ORS IN
FO
R
MA
TI
ON G
A
THE
RING
(
MOBILE
NODES)
In
t
he
physi
ol
ogic
al
facto
rs
se
tup
,
t
his
proj
ec
t
requires
t
he
de
te
ct
ion
of
the
act
ivit
y
su
ch
a
s
sta
nd
i
ng
,
walkin
g,
jum
pin
g,
r
unning
a
nd
sit
ti
ng
a
nd
instance
of
the
m
e
ta
bo
li
c
rat
e
of
the
us
e
r
or
t
he
s
ubj
ect
in
this
exp
e
rim
ent,
the
stud
e
nts.
I
n
order
to
ac
hie
ve
an
accu
rate
and
preci
se
P
MV
value
,
an
acce
le
ro
m
et
er
sens
or
from
the
m
ob
il
e
ph
one
is
to
infer
the
m
etab
olic
rate
of
the
us
er
a
nd
a
n
NF
C
ta
g
is
e
m
bed
de
d
wit
hin
the
cl
oth
in
g
to
def
i
ne
the
clot
hing
thick
ness w
orn by t
he use
r.
Cl
oth
in
g
insu
l
at
ion
is
ano
t
he
r
pr
im
ary
determ
inant
in
achieving
the
rm
al
com
fo
rt.
W
hi
le
the
m
ai
n
pur
po
se
of
nor
m
al
cl
oth
ing
is
to
protect
fro
m
the
cold
,
oth
er
ty
pe
s
of
cl
oth
in
g
s
uc
h
as
protect
ive
cl
oth
in
g
i
s
us
e
d
to
protec
t
fr
om
heat
as
we
ll
.
Cl
oth
in
g
ins
ulati
on
is
a
sta
nd
ar
dize
d
unit
by
ASHRAE
55
to
m
easur
e
therm
al
insu
la
ti
on
on
var
i
ous
cl
oth
in
g
ty
pes
[3
]
.
A
qua
ntit
at
ive
m
easur
e
will
be
w
her
e
1
cl
o
=
0.155
m
2
◦
C
.
Eq
ually
,
0
cl
o
is
wh
e
n
a
per
s
on
is
not
wea
ring
a
nythi
ng
(full
y
nak
e
d)
w
hi
le
1
cl
o
is
w
he
n
a
per
s
on
is
w
eari
ng
a
regular
tw
o
-
piece
busines
s
su
it
[1
]
.
Ta
ble
1
prese
nts
the
cl
oth
in
g
le
vel
base
d
on
sta
nd
ard
(I
S
O
7730)
that
is
us
e
d
as
t
he
bas
is
of
re
fer
e
nce
in
this
e
xp
e
rim
ent
[
13
]
.
The
m
et
abo
li
c
rate
t
ha
t
is
base
d
on
t
he
c
urren
t
act
i
vity
of
the
us
e
r
is
ob
ta
ine
d
us
in
g
the
acce
le
r
om
et
er
sensor
and
t
he
val
ue
of
m
et
abo
li
c
rate
is
ref
er
ri
ng
t
he
ANSI/A
SH
R
A
E
Stand
a
r
d
55
wh
ic
h
is
al
so
inferred
t
hrough
t
he
m
ob
il
e
app
li
cat
ion.
T
able
2
prese
nt
s
the
m
et
abo
li
c rate
of sele
ct
ed
act
i
viti
es r
el
at
ing t
o
t
he
pr
oj
ect
.
Table
1.
T
he
Cl
oth
in
g
Le
vel
ba
sed o
n
sta
nda
rd
(I
S
O 7
730)
Table
2.
Met
a
boli
c Rat
e f
or
S
om
e A
ct
ivit
y
(Met
abo
li
c Rat
e)
Ty
p
e of
Clo
th
es
Valu
e of
Clo
th
in
g
Sh
irt
0
.14
Sweate
r
0
.26
Jack
et
0
.22
Dress and
Skirt
0
.23
Activ
ity
Metabo
lic Rate
Sleep
in
g
0
.8
Sittin
g
1
.0
Stan
d
in
g
1
.2
W
alk
in
g
1
.9
Ru
n
n
in
g
8
.5
The
set
up
is
te
ste
d
to
perf
orm
con
necti
vity
between
sta
ti
c
and
m
ob
il
e
node
an
d
able
to
perform
data
transm
issi
on
.
This
ex
per
im
e
nt
is
do
ne
in
c
ollaborat
io
n
wi
th
m
ob
il
e
node
fo
r
data
trans
m
issi
on
and
g
e
ner
at
io
n
of
PMV
value
by
m
ob
il
e
node
.
U
pon
receiv
ing
t
he
val
ues
from
sta
t
ic
node
,
the
m
ob
il
e
node
rec
ord
the
value
in
their
respec
ti
ve
fiel
d
for
use
rs
co
nv
e
nienc
e
apar
t
to
be
use
d
to
ge
ner
at
e
PMV
value.
Figure
4
s
how
s
the
return
tra
ns
m
is
sion
s
of
t
wo
va
lues
of
physi
ol
og
ic
al
facto
rs
tog
et
he
r
wit
h
cal
culat
ed
PMV
v
al
ue
ba
ck
to
sta
ti
c
node
in
wh
ic
h
the
sta
ti
c
no
de
disp
la
y
the
rec
ei
ved
data
on
t
he
LCD
.
The
proces
s
of
te
sti
ng
are
re
peate
d
us
in
g
si
m
il
ar
ste
ps
at
diff
e
ren
t
locat
ion
s
wit
h
par
a
m
et
er
var
ia
ti
ons
to
ensure
f
ull
fu
nc
ti
on
al
it
y
of
the
set
up.
Fi
gure
4
sh
ows
the s
et
up
flo
w
in
cali
brat
ion
of the sta
ti
c and
m
ob
il
e
nodes
r
ea
dings
in
cal
culat
in
g t
he
PMV
.
Figure
4.
Wo
r
ki
ng
flo
w
a
nd c
al
ibrati
on of st
at
ic
an
d
m
ob
il
e nod
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
In
vest
ig
atin
g
T
hermal C
om
f
ort
for
the
Cl
as
sr
oom Envir
onm
ent usi
ng IoT
(N
ur
sha
hr
il
y I
dura R
am
li
)
161
To
eval
uate
th
e
fu
ll
functi
on
al
it
y
and
accur
acy
,
this
fr
am
e
work
is
im
ple
m
ented
as
in
scop
e
to
s
h
ow
the
therm
al
com
fo
rt
le
vel
in
a
ty
pical
cl
assro
om
of
10
x10
m
et
ers
du
e
to
var
i
ou
s
c
ontrib
uting
facto
rs
m
ai
nl
y
DBT
ove
r
a
s
m
al
l
per
iod
of
tim
e.
This
area
is
about
the
si
ze
of
a
sm
al
l
c
la
ssroom
that
c
ou
l
d
fit
20
-
30
per
s
on
and
fitt
ed
with
an
ai
r
co
ndit
ion
e
r.
Tim
e
slot
us
ed
is
in
t
he
m
or
ning
in
w
hich
it
s
hould
hav
e
a
lo
wer
e
xter
nal
therm
al
i
m
pact
on
the
te
st
be
d.
T
he
ch
os
e
n
scenari
o
is
to
m
anipu
la
te
DB
T
as
ai
r
tem
per
at
ur
e
has
the
highest
sign
ific
a
nce
on
therm
al
com
fo
rt,
hen
ce
t
o
achie
ve
bett
er
eff
e
ct
s,
thi
s
ex
pe
rim
ent
is
te
ste
d
with
ou
t
the
util
iz
at
ion
of
t
he
ai
r
-
co
ndit
io
ner
as
sli
ght
c
hanges
in
ai
r
t
e
m
per
at
ure
will
aff
ect
hum
an
com
fo
rtabil
it
y.
T
he
sta
ti
c
node
is p
la
ced
in
the
m
i
dd
le
of
the
cl
as
s
as
a
n
i
dea o
f
a
base
re
fer
e
nc
e
that
the
data colle
ct
ed
are
si
m
il
ar
thr
oughout t
he
whole cla
ss
roo
m
o
n
ave
rag
e
for t
he d
ur
at
io
n of 1
20 m
inu
te
s.
5.
RESU
LT
A
N
D DIS
CUSSI
ON
Table
3
sho
ws
the
colle
ct
ed
values
of
both
con
tri
bu
ti
ng
e
nv
i
ronm
ental
and
ph
ysi
ologica
l
factor
s
with
cal
culat
ed
PMV b
ei
ng
r
un
in
t
he
durati
on
of 1
20
m
inu
te
s,
w
hich
is
t
he
ty
pical
durati
on
of
a
le
ss
on
i
n
th
e
cl
assroom
at Un
ive
rsity
Tekn
ologi M
AR
A.
The
ai
r
t
em
per
at
ur
e
inc
reases
in
the
cl
assroom
after
the
air
-
c
onditi
on
e
r
is
turn
e
d
off.
T
he
increm
ent
of
te
m
per
at
ur
e
is
delibe
ratel
y
slow
over
ti
m
e
due
t
o
the
co
nt
ai
ned
ai
r
in
th
e
cl
assr
oo
m
that
trap
t
he
c
old
ai
r
t
o
sta
y
in
the
roo
m
.
Si
m
i
la
rly
,
t
her
e
was
al
so
a
ste
ady
cl
i
m
b
of
the
MR
T
va
lues
over
t
he
ti
m
e
per
io
d.
Whi
le
the
increm
ent
of
MR
T
is
alm
os
t
propor
ti
onal
to
the
increm
ent
of
DBT,
the
i
ncr
em
ent
of
MR
T
is
sli
gh
tl
y
slow
e
r
du
e
to
it
ta
kes
tim
e
fo
r
MR
T
to
adap
t
to
new
te
m
per
at
ure
cha
ng
es
as
it
con
sider
t
he
tem
per
a
tures
of
it
s
nearby s
urrou
ndin
gs
i
nto
calc
ulati
on
s
(F
i
gur
e 5)
.
Figure
5. (a
)
T
ran
sm
issi
on
of
DBT a
nd RH
val
ues
(
b) T
ran
s
m
issi
on
of MR
T and
Vel
valu
es (
c
)
Transm
issi
on
Cl
o
an
d
Me
t
va
lues (d
)
T
ran
s
m
issi
on
of
PMV
value
Table
3.
Rec
orded Stat
ic
and
Mob
il
e
Nodes R
ead
in
gs
with
Ca
lc
ulate
d
PM
V
Ti
m
e
(
m
in
)
DBT (
◦
C)
MRT
(
◦
C)
RH (%)
Vel
Met
Clo
PMV
10
2
0
.69
1
9
.28
7
8
.06
0
.1
1
.0
0
.7
-
1
.4
20
2
1
.54
1
9
.47
7
7
.98
0
.2
1
.1
0
.7
-
1
.1
30
2
2
.03
1
9
.95
7
7
.68
0
.2
1
.3
0
.7
-
0
.4
40
2
2
.91
2
0
.67
7
7
.13
0
.1
1
.0
0
.7
-
0
.8
50
2
3
.86
2
0
.83
7
6
.85
0
.1
1
.5
0
.7
0
.4
60
2
4
.37
2
1
.42
7
6
.31
0
.1
1
.7
1
.0
1
.0
70
2
5
.72
2
2
.39
7
6
.02
0
.1
2
.2
0
.7
1
.4
80
2
6
.03
2
3
.08
7
5
.62
0
.2
1
.0
1
.0
0
.5
90
2
6
.67
2
3
.81
7
5
.52
0
.2
2
.3
1
.0
1
.8
100
2
6
.99
2
4
.55
7
5
.12
0
.3
1
.6
0
.7
1
.1
110
2
7
.01
2
5
.23
7
4
.88
0
.1
1
.0
0
.7
0
.6
120
2
7
.32
2
5
.64
7
4
.47
0
.3
1
.2
0
.4
0
.2
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vol
.
9
,
No.
1
,
Jan
ua
ry
201
8
:
157
–
163
162
Hu
m
idit
y
oth
erw
ise
rec
ords
a
dec
reasin
g
r
at
e
in
the
cl
assroom
inv
ersel
y
to
the
i
ncr
e
m
ent
of
ai
r
tem
per
at
ur
e
[
5]
.
This
is
a
n
exp
ect
e
d
beh
a
viour
a
s
the
a
m
ou
nt
of
wa
te
r
va
pour
wi
ll
decr
ease
t
hro
ugh
evapo
rati
on
w
hen
t
he
ai
r
te
m
per
at
ur
e
incr
eases.
The
c
ha
ng
e
s
of
ai
r
ve
locit
y
in
the
cl
assroom
is
ho
weve
r
rand
om
at
time.
Wh
il
e
there
are
cha
ng
e
s
in
ai
r
vel
ocity
,
the
cha
ng
es
a
re
m
ino
r
a
nd
bar
el
y
noti
ceable
by
hu
m
an
accor
di
ng
to
it
s
st
an
da
rd
iz
e
d
ta
ble
by
AS
HR
AE
[
3].
This
occ
urrenc
e
m
ay
be
caused
by
the
m
ove
m
ent
of
h
um
an
in
th
e
room
or
any
oth
e
r
possible fact
ors
that
cau
sed
ai
r
m
ov
em
ent
inclu
ding
the
desi
gn
o
f
buil
ding
that p
ro
m
otes
ai
r
flo
w
a
nd
not ai
r
-
ti
gh
t
[8]
,
[
9]
,
[
14]
.
On
t
he
physi
ol
og
i
cal
fact
or
s
data
colle
ct
ion,
there
a
re
co
nst
ant
changes
ob
s
er
ved
i
n
th
e
read
in
g
of
m
et
abo
li
c
rate.
This
is
du
e
to
th
e
r
an
dom
act
ivit
y
the
subj
ect
pe
rfor
m
s
in
t
he
cl
ass
room
as
to
ob
se
rv
e
t
he
i
m
pact
of
m
etab
olic
rate
on
therm
al
co
m
fo
rt.
As
for
the
cl
oth
ing
l
e
vel,
the
cl
oth
in
g
values
worn
by
the
su
bject
in
the
cl
assroom
du
ri
ng
the
ti
m
e
of
te
sti
ng
.
Se
ve
ra
l
ty
pes
of
cl
ot
hing
ty
pes
were
te
ste
d
in
t
he
entire
durati
on
of
the
exp
e
rim
ent.
T
her
e
a
re
three
var
ia
ti
ons
of
cl
oth
in
g
us
ed
w
hich
are
s
hirt
with
shorts
(0.
4),
sh
irt
with
tr
ou
se
rs
(
0.7),
an
d
li
ght
business
s
uit
(1.0).
Cl
ot
hes
worn
will
ha
ve
an
ef
fect
on
therm
al
co
m
fo
rt
as
cl
oth
es
prov
i
de
s
insu
la
ti
on
f
or
hum
an
body
.
Figure
6
pr
esents
the
PM
V
val
ues
f
or
the
du
rati
on
of
120
m
inu
te
s
base
d
on
six
co
ntri
bu
ti
ng
value
s
o
f
t
her
m
al
com
fo
rt.
The
a
c
ceptable
PM
V
values
f
or
th
erm
a
l
com
fo
rt
in
sm
a
ll
occu
pie
d
sp
a
ce
is
fr
om
-
0.5 to
+0.5
wit
h
0
bein
g
m
os
t
com
fo
rtable.
A
fter
tur
ning o
f
f
the
ai
r
-
conditi
on
e
r
f
or
te
n
m
inu
te
s,
t
he
cl
ass
room
i
s
sli
gh
tl
y
c
oo
l
for
a
sede
ntary
pe
rson
wea
rin
g
a
sh
irt
an
d
tr
ou
s
er
s
(P
MV:
-
1.4
).
In
t
he
el
apse
of
20
m
inu
te
s,
the
pe
rson
wi
th
sam
e
cl
oth
i
ng
a
nd
m
et
abo
li
c
act
ivit
y
is
a
bit
m
or
e
com
fo
rtable
as
the
tem
per
at
ur
e
of
the
room
is
increasin
g
(
PMV:
-
1.1
).
A
fter
30
m
inu
te
s,
the
perso
n
is
do
i
ng
so
m
e
l
igh
t
act
i
v
it
y
su
ch
as
ar
rangin
g
pa
pers
wh
il
e
sti
ll
s
it
t
ing
do
wn
wh
i
ch
induce
the
r
m
al
fr
om
bo
dy
and
al
lowing
the
pe
rson
to
be
m
or
e
com
fo
rta
ble
in
the
co
ol
en
vir
on
m
ent
(PM
V:
-
0.4
).
At
m
inu
te
40,
the
per
s
on
feels
a
bit
m
o
re
unc
om
fo
rta
ble
due
t
o
the
increase
te
m
p
eratur
e
ove
r
ti
m
e
al
tho
ugh
not
doin
g
a
ny
a
ct
ivit
y
(P
MV:
-
0.8)
.
At
m
inu
te
50,
the
pe
rs
on
feel
s
wa
rm
er
as
do
ing
m
or
e
act
i
vity
su
ch
as
wal
king
in
the
cl
assroom
on to
p of t
he
i
nc
reasin
g
ai
r
tem
per
at
ur
e (
PMV: +0
.4).
Af
te
r
the
du
rati
on
of
60
a
nd
70
m
inu
te
s,
th
e
pe
rs
on
is
do
i
ng
m
or
e
act
ivi
ty
in
a
war
m
er
place
t
hat
con
t
rib
ute
to
a
highe
r
PM
V
va
lue
of
+1
.0
a
nd
+1
.4
re
sp
ec
ti
vely
.
A
nea
re
r
PM
V
value
of
+
0.5
is
ac
hie
ved
at
m
inu
te
80
wh
e
n
the
pe
rson
goes
back
to
sit
ti
ng
(se
den
ta
ry)
al
though
weari
ng
a
li
gh
t
business
s
uit
an
d
a
m
or
e
increase
d
ai
r
tem
per
at
ure.
T
he
rest
of
the
durati
on
of
this
exp
e
rim
ent
i
nd
ic
at
e
the
the
rm
al
co
m
fo
rt
of
the
per
s
on
in
the
exp
e
rim
ent
with
va
rio
us
m
etab
olic
act
ivit
ie
s
an
d
cl
oth
es
worn
in
an
inc
reasin
g
te
m
per
at
ur
es
and d
ec
reasin
g h
um
idity.
Figure
6
.
G
e
ne
rated
PMV
val
ues base
d on e
nv
i
ronm
ental
an
d p
h
ysi
ol
og
ic
al
f
act
ors (St
at
ic
and
Mob
il
e
Nodes
)
The
com
bin
at
ion
of
co
ntri
bu
ti
ng
fact
or
s
aff
ect
therm
al
co
m
fo
rt
a
nd
s
hows
th
at
m
ulti
ple
consi
der
at
io
ns
h
ave
t
o
be
ta
ke
n
int
o
acco
unts
to
achie
ve
the m
os
t
co
m
fo
rtable
sta
te
wh
i
c
h
denote
d
with PM
V
value
of
0.
T
his
exp
e
rim
ent
had
pro
ve
n
the
f
ull
functi
onal
it
y
and
acc
ur
acy
of
the
fr
am
ewo
r
k
not
just
th
r
ough
the
log
ic
al
inf
eren
ce
by
hu
m
an
brai
n
but
th
rou
gh
cr
os
s
re
fer
e
ncin
g
colle
ct
ed
data
with
sta
nd
ar
dize
d
ta
bles
an
d st
udie
s to
confirm
the int
egr
it
y o
f gene
r
at
ed
PM
V valu
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
In
vest
ig
atin
g
T
hermal C
om
f
ort
for
the
Cl
as
sr
oom Envir
onm
ent usi
ng IoT
(N
ur
sha
hr
il
y I
dura R
am
li
)
163
6.
CONCL
US
I
O
N
This
obse
rv
at
ory
stud
y
f
ound
that
Io
T
e
nab
l
es
autom
at
ion
of
c
ollec
ti
on
of
quantit
at
ive
va
lues
of
the
env
i
ronm
ent
that
aff
ect
s
hum
a
n
c
om
fo
rt
the
r
m
al
l
y,
an
d
e
na
bles
the
ge
ne
ra
ti
on
of
the
rm
al
com
fo
rt
value
to
be
i
m
ple
m
ented
on
the
rm
al
reg
ul
at
or
dev
ic
es,
s
pecifica
ll
y
ai
r
-
conditi
on
e
rs
i
n
sm
a
ll
le
arn
ing
en
vironm
ents.
W
it
h
the
us
e
of
this
te
chn
ol
og
y,
t
he
achievem
en
t
of
therm
al
c
om
fo
rt
will
be
easi
er
and
in
dep
e
ndent
of
hu
m
an
interve
ntio
n.
I
t
is
ho
we
ver
reco
m
m
end
ed
that
therm
al
regulat
or
de
vices
especial
ly
ai
r
-
co
ndit
ion
e
rs
f
or
autom
at
ic
therm
al
reg
ulati
ons
that
are
bette
r
su
it
ed
for
the
rm
al
co
m
fo
rt
are
util
iz
ed
to
w
ard
s
regulat
in
g
PMV
of zer
o value i
n fu
t
ur
e
resea
r
ch.
ACKN
OWLE
DGE
MENT
The
aut
hors
gr
at
efu
ll
y
ackno
wled
ge
the
hel
p
of
In
sti
tute
of
Re
searc
h
M
anag
em
ent
and
In
no
vatio
n
(I
RM
I
),
U
nive
rsiti
Teknolo
gi
MARA
(
U
iTM
)
in
provi
din
g
the
Aca
dem
ic
and
Re
search
Assim
il
at
ion
(P
r
oject
N
um
ber
:
600
-
IRMI/
DAN
A
5/3
/
A
RAS
(
0187/2
016)
resea
rc
h
gr
ant.
The
a
ut
ho
rs
are
al
so
t
ha
nkf
ul
to
the
Dep
a
rtm
ent
of
Com
pu
te
r
Tech
no
l
og
y
an
d
Netw
orki
ng
for
pro
vid
i
ng
t
he
sp
ace
for
th
e
exp
e
rim
ent
t
o
be
cond
ucted
i
n
t
his stu
dy.
REFERE
NCE
S
[1]
Rawi,
M.
I.
M
.
,
”Sensor
Netw
ork
Embedde
d
Inte
lligen
ce
:
Hu
m
an
Com
fort
Am
bie
nt
Inte
lli
genc
e
,
”
Au
ckla
nd
Unive
rs
it
y
of
Te
chnol
ogy
,
2013
.
[2]
Rawi,
M.
I.
M.
,
&
Al
-
Anbuk
y
,
A.,
”De
v
el
opm
e
nt
of
Inte
l
li
gen
t
W
ire
le
ss
Sensor
Networks
for
Hum
an
Com
fort
Inde
x
Mea
sur
ement,”
Proce
dia
C
omputer
Scien
c
e
,
5
,
232
-
239
.
doi
:
10.
1016/
j.
proc
s.2011.
07.
031
6
2
,
2011.
[3]
Am
eri
ca
n
Soci
e
t
y
of
Hea
ti
ng
,
Refr
igerat
ing
and
Air
-
Condit
i
oning
Engi
n
ee
r
s,
Inc
.
,
”
The
rm
a
l
Envi
ronm
enta
l
Condit
ions f
or
Hum
an
Occ
upan
c
y
”
A
tl
anta
,
N
ebraska
,
2004
.
[4]
Boduch
M.,
&
Finche
r,
W
.
,
”Sta
ndar
ds
of
Hum
an
Com
fort:
R
elati
v
e
and
Abs
olut
e,”
The
Uni
ver
sity
of
Te
xas
a
t
Austi
n
,
2009.
[5]
Roset,
J.,
e
t.
al.,”L
e
arn
ing
about
Hum
idi
t
y
in
Th
ermal
Com
fort,
”
Unive
rs
it
at
Pol
i
te
cn
ic
a
de
Catal
uyna
,
Ret
r
ie
v
e
d
from
htt
p://ww
w
-
fa
.
upc
.
es/p
ersona
ls/j
roset
/e
sun
-
058.
html, 20
09.
[6]
Chan,
L
eon,
Mo
hd
Eqwa
n
Bin
Mohd
Roslan,
a
nd
Hass
an
Bin
Moham
ed.
”In
ve
stiga
ti
on
of
the
Optimal
Pos
it
io
n
of
W
ind
Sensors
and
W
ind
Turbi
n
es
on
a
Buil
d
ing
:
A
Com
puta
ti
on
al
Fluid
D
y
nami
cs
Stud
y
.
”
Indon
esian
Journal
of
El
e
ct
rica
l
Eng
in
ee
ring a
nd
Computer
Sc
i
ence
8
.
3
,
2017
.
[7]
Gunawan,
T
edd
y
Sur
y
a
,
et
al
.
”
Deve
lopment
of
Face
R
ec
ogn
it
i
on
on
Raspber
r
y
Pi
for
Secur
ity
Enha
n
ce
m
ent
of
Sm
art
Hom
e
S
y
stem.”
Indone
sia
n
Journal
o
f El
e
ct
rical E
ng
ineering
and
Informati
cs
(
IJ
EE
I)
5.
4
,
2
017.
[8]
Ramesh,
P.,
and
V.
Mathi
van
an.
”War
eWise:
Business
Deve
lopment
Mana
gemen
t
Fram
ework
bas
ed
on
Devic
e
-
to
-
Devic
e
Industr
ial
Inte
rn
et
of
Things
.
”
Indone
sia
n
Journal
of
E
l
ec
tri
cal
Eng
inee
ring
and
Computer
Scienc
e
8.
2
,
2017.
[9]
Noh,
S.
-
K.,
e
t
a
l.
,
”De
sign
of
a
Room
Monitor
ing
S
y
stem
for
W
ire
le
ss
Sen
sor
Networks,”
De
sign
of
a
Room
Monit
oring
S
yst
em
for
Wire
le
ss
Sensor
Net
works
,
Int
ern
a
ti
ona
l
Journal
of
Distribut
ed
Sensor
Networks,
2013,
1
-
7.
doi:
10
.
1155/20
13/189840,
2013
.
[10]
D.
Overb
y
,
”Eval
ua
ti
ng
Hum
an
The
rm
al
Com
fort,
”
IDO
Inc
orporat
ed
[
on
li
ne
]
Av
ail
ab
le
at:
htt
p://idoinc
orp
orated.
com/
ev
a
l
uati
ng
-
human
-
th
ermal
-
comfort/
,
2013.
[11]
Fanger
,
P.
O.
,
”
T
her
m
al
Com
fort:
Anal
y
sis
and
Applicati
ons
in
E
nvironmenta
l
En
gine
er
ing,
”
New
Y
ork:
Mc
Gr
aw
Hill
,
1970.
[12]
Holopai
nen
,
R
.
,
”
A hum
an
the
rm
a
l
m
odel
fo
r
impr
oved
th
ermal
co
m
f
ort,
”
Fi
nland:
Aalto Uni
v
ersit
y
,
2012
.
[13]
Inte
rna
ti
ona
l
Or
gani
z
at
ion
for
Standa
rdi
za
t
ion
(2005).
ISO
7
730:
Ergonomi
cs
of
the
Therm
al
Environm
ent
Anal
ytica
l
Dete
r
minati
on
and
Int
erpretat
ion
of
T
hermal
Comfort
Us
ing
Calc
ulat
i
on
of
the
PMV
a
nd
PP
D
Indic
es
and
Local
Ther
mal
Comfort Cri
te
ria.
,
Gen
eva,
S
witz
er
la
nd,
2005
.
[14]
Inte
rna
ti
ona
l
Organi
z
at
ion
for
Standa
rdi
za
t
ion
(2006).
ISO16813:
ISO
16813
Bui
ldi
ng
env
ironmen
t
design
Indoor
Env
ironment
Ge
neral
princ
ipl
es,
Gene
va
,
Sw
itze
rla
nd,
2006.
Evaluation Warning : The document was created with Spire.PDF for Python.