Proceedings of Conference: Adapting to Change: New Thinking on Comfort

Cumberland Lodge, Windsor, UK, 9-11 April 2010. London: Network for Comfort
and Energy Use in Buildings, http://nceub.org.uk

Thermal Comfort of Non-Malaysian Residents at Different Levels of

a Multi-Storey Residential Building in Malaysia
Samira Zare Mohazabieh, Elias Salleh, Mohammad Fakri Zaky Jaafar, Nur
Mariah Adam
Abstract:
Creating a thermally comfortable environment is one of the most important criteria to be
considered when designing multi-storey residential buildings in Malaysia. These types of
buildings are increasingly being occupied by non-Malaysian (Iranian) residents.
However, there is a lack of information as to the actual thermal comfort conditions
endured by their residents under natural ventilation.
This study seeks to establish the indoor climatic conditions at different levels of a
naturally-ventilated multi-storey residential building during the month of April 2009, by
comparing measured conditions to the thermal comfort sensation (Actual Thermal
Sensation-ATS) of international occupants. Additionally, the compatibility of PMV
application with ATS is also reviewed
The field measurement results indicate that residential units at higher levels tend to be
more thermally comfortable compared to the lower units. However, these results differ
from the ATS results in that the lower units were deemed more comfortable.
Furthermore, it was found that the PMV scales cannot be directly applied to thermal
comfort evaluation in countries with hot and humid conditions, such as Malaysia, even
for non-locals.
Keywords:
Multi-storey Residential Building, Microclimate, Thermal Comfort, Predicted Mean Vote,
Actual Mean Vote.
1. Introduction
As a developing country, Malaysia is facing problems to accommodate the urban
population. The migration of rural population to cities caused the government and private
agencies to build different kind of houses (Mohamad Ali 2003). Creating a thermally
comfortable environment is one of the most important parameters, in addition to the
residential type, to be considered when designing buildings,. However, in most cases
thermal comfort has not been given due consideration.
Due to this lack of optimal design of buildings according to the energy efficient standards
in different climates, especially tropical regions, there is a need for architects to find
optimal ways for designing buildings (Feriadi et al. 2004). In fact, recent trends have
shown a heavy usage of mechanical ventilation devices for effective distribution of air in
securing thermal comfort in tropical countries. These days, because of the increase in
usage of mechanical ventilation devices, it has been observed that energy consumption
with the aim of achieving comfortable environment is getting higher and therefore
costlier. As a result, most of the occupants prefer to use natural ventilation to reduce the
running cost in buildings (Wong et al. 2002). In addition, in maritime countries such as
Malaysia which are near the equator and have low evaporation rate, long hours of
sunshine, high relative humidity and high overcast cloud cover, buildings should be
1

designed to enable natural ventilation. Therefore, the efficient use of energy and the
promotion of energy saving in designing new residential buildings and renovation of old
buildings in tropical countries are crucial (Hatamipour et al, 2007).
In the past 50 years, some thermal comfort studies have been performed in humid tropical
regions (Wong et al. 2002). Unsurprisingly, thermal responses from occupants recorded
in different countries displayed disagreement with predictions by the PMV model
(Becker et al. 2009). The PMV model seems to overestimate the sensation of warmth in
non-air-conditioned buildings located in warm climates. Also, through most of the
previous studies researchers proved that the predicted mean vote (PMV) model of
thermal comfort has not agreed with the results of field studies in non air-conditioned
buildings in different climates, including naturally ventilated buildings located in the
tropical regions (Fanger and Toftum 2002).
Many researchers have compared the results of field surveys with the PMV model in
different areas to find out whether the actual mean votes agree with the predicted mean
votes (PMV) model in different climates (Wong et al. 2002).
Currently, there has been little discussion in the earlier studies on establishing a sufficient
thermal comfort assessment for the domestic sectors, especially multi-storey domestic
buildings in tropical countries like Malaysia. According to the literature review, most of
the previous studies in relation to the thermal comfort in Malaysia have been conducted
in low-cost housing, traditional Malay houses, terrace houses, and air-conditioned houses,
however, only the one study by Dahlan (2008) involved multi-storey hostels in Malaysia.
Correspondingly, bioclimatic design in Malaysian multi-storey domestic buildings has
not been fully explored according to the occupants comfort needs. Therefore, it is the
intention of this study to fill this gap, by providing information of thermal comfort in this
type of dwellings where the use of non-air conditioning is employed, focusing on nonMalaysian residents.
Hence, this could be possible by finding out the existing level of thermal sensation and
thermal comfort of residents in such buildings. The need to establish comfort conditions
necessary for the occupants then becomes the first line of inquiry of the current study
which is intended to investigate the thermal comfort conditions and also to be the first to
examine thermal comfort in a Malaysian naturally ventilated multi-storey domestic
building occupied by non-Malaysians.
The first objective of this study is to evaluate the indoor thermal comfort conditions at
different levels of a naturally-ventilated multi-storey residential building in Malaysia
using the PMV model, through the measurement of microclimatic factors.
The second objective is to establish the thermal comfort sensation (actual thermal
sensation) of the non-Malaysian residents (in this case, Iranians) at different levels of the
same building. The final objective is to investigate the compatibility of the PMV scale of
thermal comfort with the thermal sensation of the occupants.
2. Research Methodology
Since the focus of the study is on indoor climate and thermal comfort in multi-storey
residential buildings in Malaysia, an empirical research method has been used. It involves
two steps, namely objective and subjective measurements. These measurements were
conducted in a multi-storey residential building located in the East Lake complex from
the15th to 20th of April, 2009. The three units under study were located in the third, sixth,
and tenth floors of this multi-storey residential building. Moreover, all the units had the
same orientation and were all naturally - ventilated (Figure 1).

Figure 1: Locations of the three units at different levels

2.1 Objective Measurements

The first method is field work measurements of all the relevant thermal comfort
variables. The thermal comfort level of the indoor environment was measured inside
three units, each from a different level, from the 15th April to 20th of April, 2009. The
month of April was chosen because according to the Malaysian Meteorological
Department (MMD) report (2009), generally, most places in the country had higher than
average temperature in the months of April and May (www.met.gov.my).
For the purpose of the measurements, a physical quantities measurement instrument,
namely the Universiti Kebangsaan Malaysia (UKM) Thermal Comfort Multi-station
(TCM) logger was used. This logger is able to measure all four environmental variables
(air velocity, relative humidity, dry bulb temperature and globe bulb temperature). In
addition, HOBO loggers were used during the field measurement for the purpose of
double checking the climate data. For every residential unit, one HOBO logger was
placed outside and inside the living room starting from the 12th until the 26th of April,
2009. The measurements were taken by the HOBO loggers at every 10 minute interval to
record the respective air temperatures and relative humidity of both the outdoor and the
indoor environments.
The UKM TCM logger was installed in the living room of each apartment. This is due to
the fact that the majority of the residents spend most of their time in the areas of their
living and dining rooms for their daily activities (watching TV, eating, socializing with
mates and maybe studying). In contrast to the many HOBO loggers, only one UKM TCM
data logger was available throughout the field measurement, therefore each unit had the
logger for the time period of two days only during the 15th until the 20th of April, 2009.
The measurements ran every 10 minutes from 6.00 pm till 10:00 pm. This timing was
chosen because all the residents were present in their units at that period. At the end of
each measurement in a unit, the data were saved accordingly, and the UKM TCM logger
was transferred to the next unit for installation (Figure 2).

In this study, the metabolic rate is set to be 60 W/m2 (Seated relaxed) whereas the Clovalue (thermal resistance) is set to be 0.4 clo where the residents were wearing
underwear, normal trousers, and T-shirt.
At the end of each monitoring by the UKM TCM logger the data were downloaded by the
PC software for the purpose of data acquisition. In order to analyze them, the output file
was then exported to a Microsoft Excel 2007 spreadsheet. The measurement assessment
is mainly concerned with the variation of PMV during the evening period. To calculate
the PMV equation, an online program PMV 2008 ver 1.0, Ingvar Holmer was used
(www.Ingvar Holmer.com).

Figure 2: View of the living room, TCM logger and the computer used for Data acquisition

2.2 Subjective Measurements

To complement the results obtained by environmental monitoring, the method of field
survey of the occupants responses to simple questionnaires was used. The questionnaires
along with the environmental monitoring process were simultaneously carried out among
the occupants of the three specified units.
The assessment of the thermal environment in the aforementioned naturally ventilated
multi-storey residential building was based on the occupants votes on thermal sensation
and comfort perception estimated by the Preference Scale (Mclntyre scale) and ASHRAE
Scale.
Each specified housing unit was occupied by about 3 to 4 persons and all of the residents
were international Iranian students. Residents must have been living in their respective
residential units for at least three months to be chosen as participants in the survey. At
least two individuals from each of the three units were selected to answer the
questionnaires. In the survey method, the occupants were randomly chosen from both
genders in the age range of approximately 20 to 45 years old.
The residents provided answers to the questionnaires two times during each of the two
monitoring days. The first one was at 6.00 pm and the second at 10.00 pm. This timing
was chosen because all the residents were present in their units at those periods. For the
sake of having more accurate responses (ISO 7730 Standard), all the respondents must
4

have been in seated positions for at least 30 minutes before the time of answering the
questions. In total, there were 32 residents, who responded to the questionnaires from the
15th to the 20thof April, 2009. Out of all the respondents, 11 of them were residents of the
three units that were monitored simultaneously. The remaining 21 respondents, did not
have their units monitored, however their units were in the same orientation as the
monitored units. Each of the low, middle and high levels had respondents from both
monitored and non-monitored units. All the respondents, from either monitored or nonmonitored units, answered the surveys two times a day, at the same time periods, during
the evening. Altogether, because each respondent provided answers four times (two times
during each of the two days), there were 128 responses collected.
To assess the thermal sensation, a seven-point ASHRAE scale was provided in the
questionnaire form. The scale consists of the following: cold (-3); cool (-2); slightly cool
(-1); neutral (0); slightly warm (+1); warm (+2) and hot (+3). Throughout this study,
subjects responses are referred to as Actual Thermal Sensation Vote. The Preference
scale is useful for subjective assessment of comfort and focuses on the issue of thermal
preference, introduced by McIntyre (1978). This scale identifies the thermal conditions
people prefer to be in rather than the exact sensation they are experiencing. The neutral
vote is categorised as no change whilst the other two are want cooler and want
warmer. At last, according to the statistical analysis of the mean comparison of paired
samples t-test, the assessment of the objectives was performed by the SPSS Program
version15.00.
3. Results and analysis
3.1 Evaluation of Thermal Comfort
In accordance to the first objective, the summary of the field measurement results
conducted during the 15th to20th of April, in the three different levels units of a naturallyventilated multi-storey residential building is presented in Table 1.

0.05

+1.46

29.10

66

Neutral
Temp.
(C)

Mean
Values

29.80

RH (%)

Temp. (C)

Air Velocity
(m/s)

RH (%)
59

Mean
Values

30.40

PMV

Outside Climatic
parameters (Mean
Values)

Globe
Temp.
(C)

3rd

Thermal Comfort parameters

(Mean Values)

Temp. (C)

Units position

Table 1: Summary of indoor Thermal Comfort measurements, outdoor climatic parameters and
Neutral Temperatures during the 15th-20th of April 2009

26.50

28.60
6

th

10th

29.80

62.5

0.13

29.40

30.60

61

0.38

30.40

+ 1.21

+ 1.43

30.30

67

26.80

66

27.00

3.2 Data Analysis Based on Simulation with Reference to the ISO 7730
The thermal comfort range was taken to be the conditions when the PMV has the
values between -1 and +1, based on ISO 7730-94. Thermal comfort study of the units
in different levels of the naturally-ventilated multi-storey building showed that in
most of conditions, PMV values for the specified units do not fall within this range.
By comparing the results, PMV range in the unit which is located in the 6th floor is the
lowest as its mean shows (+ 1.21), followed by the 10th floor (+ 1.43) and the 3rd floor
(+ 1.46). While the main climatic elements affecting building process and thermal
comfort level (PMV) are solar radiation, air temperature, humidity, wind, and rainfall
(Koenigsberger et al. 1980; Markus and Morris 1980; Givoni 1976), solar radiation is
the most important element among all, as it includes the amount of heat transferred to
the buildings and residents (Peng Chen 2002; Griffiths 1976; Battan 1974; Trewartha
1968).
From the data analysis, the 6th floor has the lowest mean temperature (29.80C)
followed by the 3rd floor (30.40C) which in turn is followed by 10th floor (30.60C).
Therefore, it is evident that the resulting values are not within the comfort range, as
stated by ASHRAE 55, during the evening hours (6.00 until 10.00 PM).
Air temperature variation values have a close relationship with the solar radiation
absorption from the sun. As faade design of the building is the same for all levels in
terms of shading, materials and glazing, findings revealed that the 6th floor unit is
more comfortable, because the unit is located in the middle level and thus is not
exposed to the solar radiation as much as the highest level (10th floor). Moreover, the
mean temperature value (29.8C) in the 6th floor is the lowest compared with the 10th
(30.6C) and the 3rd floor (30.4C) units mean temperature value.
Although the mean temperature value in the 3rd floor (30.4C) is less than that of the
10th floor (30.6C), its PMV is higher. The reason is that the 3rd floor unit, as the
lowest level, has less wind speed variations (0.05m/s) compared to the 10th floor
wind speed variations (0.38m/s).Thus, it renders the 10th floor to have a lower mean
value of PMV than the 3rd floor. Some studies (Rohles et al. 1982, Scheatzle et al.
1989, and Fountain 1991) have implied that the extent of comfort zone could be
reached in high air temperatures by increasing the air movement more than the
expectation of international standards (ASHRAE 55-2004). This shows that as the
height increases, the wind speed variations increase as well to affect thermal comfort.
3.3 Discussion With Respect to Neutral Temperature
In this study, the neutrality temperature for naturally-ventilated buildings in Malaysia
was predicted by using the Auliciems equation. According to the literature, the
studies of Daghigh et al. (2009) in a naturally-ventilated office and Dahlan et al.
(2008) in naturally-ventilated multi-storey hostels, both utilized the Auliciems
equation in the Malaysia region . Subjects thermal neutrality was predicted using the
calculation adopted from the optimum thermal comfort, Tn, model for naturally
ventilated buildings. The Auliciem model (1982) estimated Tn, using the following
equation: Tn = 17.6 + 0.31To where Tn is the comfort temperature and To is the mean
of the outdoor dry bulb temperature. Then the range of temperature around Tn
corresponding to 90% thermal acceptability is defined (Table 2).
This percentage of acceptability is applied as a function of operative temperature in
order to produce a 90% acceptable comfort zone.
Numerous climate chamber and field studies have been conducted in hot and humid
South East Asian countries and the surrounding regions since the 1930s. In all studies,
the proposed neutral temperatures are higher than the 24.5C recommended by
6

ASHRAE Standards 55. For the indoor design conditions, the comfort range for all
studies have higher maximum values, some at 4 degC higher than the recommended
range (23C to 26C) for air-conditioned buildings. For naturally-ventilated buildings,
the neutrality temperatures of 26.1C and 27.4C have been recommended for
Malaysia (Abdul Rahman and Kannan 1997, Sabarinah and Ahmad 2006). In this
study, the neutrality temperatures for the units of the three levels are between 26.5C
and 27.0 C. It is also possible to state that the three units located in different levels,
have almost acceptable amount of neutrality temperatures (26.5C for the 3rd floor,
26.8C for the 6th floor and 27.0C for the 10th floor unit) in terms of thermal
conditions.
Based on Auliciems equation (1982), for 90% acceptability of thermal comfort, the
suggested value is Tn 2.5 C and for 80% acceptability the suggested value is Tn
3.5 C. In order of the levels heights, by adding 2.5C to these values for 90%
acceptability, the comfort zone is between 24C and 29C in the 3rd floor unit,
between 24.30C and 29.30C in the 6th floor unit and between 24.50C and 29.50C
in the 10th floor unit (Table 2).
Table 2: Outside Climatic parameters, Neutral Temperatures (Mean Values)
Outside Climatic
parameters (Mean
Values)

17-18
April

28.60

66.64

30.30

66.33

24

26.80

24.30

27

24.50

Tn+ 3.5

10th Floor

26.50

80%Accept.

Tn- 3.5

15-16
April

65.77

Tn + 2.5

6th Floor

29.10

90%Accept.

Tn- 2.5

19-20
April

Neutral Temp.
(C)

3th Floor

Humidity (%)

Date

Air Temp (C)

Units
Position

Neutral Temperature (C)

29

23

30

29.3

23.3

30.3

29.5

23.5

30.5

3.4 Discussion of the Subjective Measurements

This assessment involved 32 respondents real vote, from all units (monitored and
non-monitored units) who answered the surveys two times a day (at 6.30 pm and
10.00 pm during the 15th until the 20th of April). The analysis results of the
respondents votes from each of the three different levels (low, middle, high) units
according to the preference scale is shown in Table 3. This scale identifies the thermal
conditions people prefer to be in rather than the exact sensation they are experiencing.
Among the three levels, the low levels units are the ones with the highest number of
residents (62.5%) who expressed a neutral vote (no change). In this regard, the low
levels units are followed by the high and the middle levels units with 47.23% and
25% of residents expressing a neutral vote respectively.

Table 3: Relative frequency of ASHRAE Thermal votes: Analysis of Votes on ASHRAE scale in
different levels

Thermal
Preference

Low levels
(19th & 20th April)

Middle levels
(15th & 16th April)

High Levels
(17th & 18th April)

Frequency

Percent

Frequency

Percent

Frequency

Percent

Want Cooler

15

37.5

39

75

19

52.77

Want No
change

25

62.5

13

25

17

47.23

Total

40

100%

52

100%

36

100%

3.5 Discussion of Thermal Comfrot range and Actual Thermal Sensations

Comparison
Based on the subjective survey results, it was the low level unit that was shown to be
the most thermally comfortable, followed by the high and middle levels units
respectively.
However, this result is different from what was obtained through the field objective
measurements. Based on the PMV results it was found that the middle level unit was
the most thermally comfortable unit, followed by the high and low level units
respectively. This variation in the results of the study is attributed to the relative
humidity variations which have the lowest value (59%) in the lowest level unit (3rd
level) followed by the high (10th floor) and middle level (6th floor) units (61% and
62.5%) correspondingly. In other words, as the relative humidity increases, the range
of thermal comfort sensation gets narrower, which translates in an increased thermal
discomfort.
This is supported by the literature. Nicol (2004) states, it is generally assumed that in
hot conditions where loss of metabolic heat by convection and radiation decrease and
the bulk of heat losses are via evaporation, increased humidity will increase
discomfort. Furthermore, Nicol (2004) concludes in his study that in humid climates
or in conditions in which the relative humidity is high, temperatures that are about 1C
lower may be required by people to remain comfortable, but the main consequence of
higher humidity (or water vapor pressure) is that it has the effect of reducing the width
of the comfort zone. Besides, for the purpose of discovering whether PMV and actual
thermal sensation are compatible, the following analysis is presented.
The calculated PMV in the three different levels units (3rd, 6th and 10th floor) of the
naturally ventilated multi-storey residential building indicated that the three specified
units have a mean value of around +1.37 for PMV in most of the conditions.
According to the results which were obtained from the survey, in monitored and nonmonitored units, the mean value for the thermal comfort vote is 1.0397. Results
obtained from the survey show that in most conditions thermal vote centred around
1.0 (slightly warm) and by equating the central three categories of ASHRAE scale
with the notion of acceptability, some of the occupants are assumed to be satisfied
with thermal conditions in their units, while the calculated PMV indicated that the
indoor conditions of the units almost more than slightly warm in most conditions.
This means that by making a comparison between PMV and the actual thermal
8

sensation vote it is found that the PMV predicted warmer thermal conditions than the
real thermal comfort sensations of specified residents.
The statistical analysis of the mean comparison of paired samples t-test (SPSS
Program) is also performed, in order to find the compatibility of the PMV scale of
thermal comfort with the thermal sensation of the Iranian residents.
The results showed that the PMV scale of thermal comfort and the thermal sensation
of these Iranian residents do not have enough compatibility.
4. Conclusion
In the current study, the neutrality temperatures for the three units located in different
levels of the naturally ventilated multi-storey residential building have almost, in
terms of thermal condition, acceptable amounts that are between 26.50 C and 27 C,
falling within the neutrality temperatures range as predicted by Abdul Rahman and
Kannan (1997), Sabarinah, Ahmad (2006) and Daghigh, Sopian and Moshtagh
(2009).
This study has revealed that according to the field measurements, in terms of the
levels height, the middle floor unit (6th floor) was more comfortable, followed by the
highest (10th floor) and the lowest (3rd floor) levels units. One factor which has an
important effect on PMV and comfort range is air temperature. The middle level unit
which has the lowest mean value of PMV (+1.21) among the units had the lowest
mean temperature value (29.8C) which is almost near the comfort range mentioned
by Sapian et al. (2001).
In contrast to the field measurements, the subjective surveys results demonstrated
that it was the low level unit that was the most thermally comfortable unit, followed
by the high and middle level units respectively. This difference in results is attributed
to the relative humidity variations.
As a result, based on the studys outcomes, the PMV prediction of thermal comfort at
different levels of the naturally-ventilated multi-storey residential building was
warmer than the real thermal comfort desired by the occupants.
This result is consistent with the research conclusion which was obtained by Dahlan et
al. (2008) regarding two high-rise university hostels located in Universiti Malaya,
Petaling Jaya (HH1) and Universiti Putra Malaysia, Seradng (HH2) from May 12 to
June 19 in 2007. An explanation for these results could have roots in the Adaptive
Theory which suggests that people are not passively receptive of their thermal
environment (de Dear and Braer 2002) and (Zhang et al. 2007). It could be understood
that similar to literature results which mentioned the Malaysians acclimatization to
higher temperature, the residents of the current study, who were non-Malaysians, also
acclimatized to the higher temperatures but not as much as the locals.
This difference in results means that the PMV model of thermal comfort needs to be
modified in order to have an appropriate thermal index model that represents the real
thermal comfort conditions in real life situations.
To sum up, the PMV formula, when it is used to predict people's mean comfort votes
in their everyday situations, can be inaccurate to a high degree, especially in warm
environments. PMV has the capability of being greatly modified, especially in hot and
humid regions, so that the validity of its predictions improves (Humphreys and Nicol,
2002).

References:

Auliciems, A. "Towards a psycho-physiological model of thermal perception."

Int J Biometeorol, 25, 1981: 109122.
ASHRAE Standard 55-2004, (2004), Thermal Environmental Conditions for
Human Occupancy. (ANSI Approved).
Daghigh, Roonak, Kamaruzzaman Sopian, and Jalil Moshtagh (2009) , Thermal
Comfort in Naturally Ventilated Office Under Varied Opening Arrangements:
Objective and Subjective Approach. European Journal of Scientific Research, pp
260-276.
Dahlan, N.D, P.J Jones, D.K Alexander, E Salleh, and D Dixon (2008), Field
Measurement and Subjects' Votes Assessment on Thermal Comfort in High-rise
Hostels in Malaysia. Indoor and Built Environment, pp 334-345.
de Dear, RJ, and GS Braer (2002), Thermal Comfort in Naturally ventilated
buildings: Revisions to ASHRAE Standard 55. Energy Buildings, Vol.34, No.6,
pp 549561.
Fanger, P.O, and Jorn Toftum (2002), Extension of PMV model to non-airconditioned buildings in warm climates. Energy and Buildings, pp 533-536.
Feriadi, Henry, and Nyuk Hien Wong (2004), Thermal comfort for naturally
ventilated houses in Indonesia. Energy and Buildings, pp 614-626.
Fountain, M.E. (1991), Laboratory studies of the effect of air movement on
thermal comfort: a comparison and discussion of methods. ASHRAE Trasactions
Vol. 97, pp 863-863.
Humphreys, M.A., and J.F. Nicol (2002), The validity of ISO-PMV for
predicting comfort votes in every-day thermal environments. Energy and
Buildings, Vol. 34, No. 6, pp 667684.

Nicol, Fergus (2004), Adaptive thermal comfort standards in the hothumid

tropics. Energy and Buildings Vol. 36 , pp 628637.
Records of Meteorological Data (2009), Malaysian Metrological Department:
Petaling Jaya Station, Kuala Lumpur.
Rohles, F., and S. and Jones B. Konz (1982), Ceiling fans as extenders of the
summer comfort envelope. ASHRAE Transactions, Vol. 89, Part 1, pp 245-263.
Sabarinah, S.H, and Ahmad (2006), Thermal Comfort and Building Performance
of Naturally Ventilated Apartment Building in the Kelang valley: A Simulation
Study. Proceedings of the Energy in buildings(sustainable symbiosis) Seminar. pp
115-132.
Scheatzle, D., and H. and Yellot J Wu. (1989), Extending the summer comfort
envelope with ceiling fans in hot, arid climates. ASHRAE Transactions, Vol. 95,
Part 1.
10

Wong, N.H, H Feriadi, P.Y Lim, K.W Tham, C Sekhar, and CheongK.W (2002),
Thermal comfort evaluation of naturally ventilated public housing in Singapore.
Building and Environment,Vol. 37, pp 12671277.
Zhang, G, C Zheng, Yang W, Q Zhang, and DJ Moschandreas (2007), Thermal
comfort investigation of naturally ventilated classrooms in a subtropical region.
Indoor Built Environ ,Vol. 16, No.2, pp 148158.

11