PLEA 2020 A CORUÑA

P la n ni n g P ost C ar b o n C it ie s

The influence of building form on energy use, thermal

comfort and social interaction
A post-occupancy comparison of two high-rise residential buildings in
Singapore
JUAN CARLOS GAMERO-SALINAS 1, NIRMAL KISHNANI 2, AURORA MONGE-BARRIO 1,
BHAVYA GANDHI 2, MEGHA BILGI 2, ANA SÁNCHEZ-OSTIZ 1

1 School of Architecture. Department of Construction, Building Services and Structure. University of Navarra
(UNAV), Pamplona, Spain
2 School of Design and Environment. Department of Architecture. National University of Singapore (NUS),
Singapore

ABSTRACT: Two recently completed high-rise residential developments, located side-by-side in a neighbourhood
in Singapore, are compared in a post-occupancy study. Both have near identical demographics, are exposed to
the same microclimate, and constructed with a similar palette of materials. The primary difference is form. One
has a high degree of porosity with inner voids that act as conduits for natural air flow and offer a sheltered space
for social engagement. The other is more compact, less porous and has social spaces attached to the building’s
exterior. The study included surveys of residents, behavioural observations and environmental measurements.
On three counts – self-reported energy use, thermal comfort and social interaction – the former appears to be
more successful than the latter. Findings suggest that building form affects multiple outcomes at once. A form
strategy that lowers energy use, for instance, can also improve social engagement. The implication of this socio-
environmental approach to form-making is discussed in the context of high-density tropical typologies.
KEYWORDS: Building Form, Energy Use, Social Interaction, Thermal Comfort, Tropical Climate

1. INTRODUCTION Gaps in buildings that facilitate airflow, for instance,

Building form is an important factor that shapes are also the spaces for social gatherings.
environmental performance. Form variables such as
geometry, compactness and porosity play a key role 2. BACKGROUND
in passive outcomes, such as shade, access to daylight In 2008, the Housing & Development Board (HDB)
and natural ventilation [1]. Passive design has long of Singapore commissioned two high-rise public
been a consideration at the drawing board, as this housing developments (Figure 1) within the same
pertains to indoor comfort and energy demand [2], neighbourhood. Both buildings, completed in 2015,
and more recently to overheating risk reduction have a near identical demographic breakdown, are
[3,4,5]. In the tropical context, the emphasis on exposed to a similar microclimate, and constructed
passive design and measured performance, as drivers with a similar palette of materials.
of form-making, was first advocated by Malaysian
architect, Ken Yeang, who applied it high-rise
buildings in dense urban conditions [6]. His case for
the bioclimatic model – which proposed form-
features such as skyterraces and form-strategies like
placement of service cores to reduce solar gains –
was influential in the 80s and 90s in Southeast Asia,
at a time when energy security was a concern [2].
With the advent of the Green movement in the
2000s, however, the question of performance was
assigned to electro-mechanical solutions such as air
conditioning. At this time, design firms like WOHA
(Singapore) also began experimenting with new form
typologies that could push the limits of passive design
[7]. What is noteworthy about strategies by WOHA is Figure 1: Building A (left) and Building B (right)
that they merge the environmental and the social [8].
The primary difference is their approach to form.
Building A (Figure 2), by architects WOHA

Authors ORCID: Vol.1 | 61

1st : 0000-0002-0219-9848
3rd: 0000-0003-4928-8773 35th PLEA Conference. Planning Post Carbon Cities.
6th: 0000-0002-4440-4643 DOI: https://doi.org/10.17979/spudc.9788497497947
(Singapore), has a significant degree of porosity – estimation of building form variables and
gaps in the façade that let natural airflow pass environmental measurements.
through the towers. There are 12 vertically
distributed skyterraces (four per tower) that, with the Table 1: Population and sample size for each building.
sky roof, act as social spaces. An inner void that runs Building Nº of apartments Nº of surveys %
Building A 960 49 5.10
vertically through each tower accelerates air flow,
Building B 758 46 6.07
and acts as a semi-outdoor buffer space that
mediates between outdoor conditions and apartment
3.1 Survey
interiors. Each apartment opens onto this inner void,
Surveys were carried out in the common areas of
with which it interacts socially and environmentally.
each building. Surveyees were randomly selected
Building B (Figure 3), by comparison, is compact
On energy use, surveyees were asked: ‘what is
and less porous. It has six skybridges for its residents,
your approximate monthly electricity bill?’ (Q1). The
and one skygarden above a multi-storey carpark.
options they were given were based on Singapore
None of these, however, are fully sheltered, nor are
Power National Average Household Consumption for
they directly connected to the apartments.
2019 [11]. Surveyees were also asked ‘do you have
WOHA has stated that the design goals for
air-conditioning (AC) installed in your home?’ (Q2),
Building A are occupant comfort, lower energy
and ‘at what time of the day is it usually turned on?
demand and social engagement [8]. The designer of
(Q3)’.
Building B has spoken of creating a community
On use of social spaces, surveyees were asked ‘do
building in the Modernist vocabulary [9,10].
you visit the skyterraces in your estate?’ (Q4). If they
answered ‘yes’, they were then asked: ‘how often?’
(Q5), ‘how much time do you spend in them?’ (Q6),
and ‘why you visit them?’ (Q7). Additionally, residents
were asked (on a Likert scale) if ‘skyterraces
(including roof) are used by their neighbours’ (Q8).
To gauge social interaction between residents,
surveyees were asked ‘how many neighbours in your
estate are you friends with?’ (Q9). For this question,
the answer options were ‘0-4’, ‘5-9’, ‘10-14’, ’15-19’,
’20-25’ and ’25-30’.
On the question of comfort, surveyees were asked
‘in terms of thermal comfort, how do you generally
feel in the skyterraces and roof of your estate?’ (Q10).
They could answer from a 5-point thermal comfort
Figure 2: Building A showing section, floor plan with scale, commonly used in comfort studies [12].
skyterraces and tower axonometric with inner void
Each surveyee was asked his/her age, household
size, floor level where s/he lives and how long s/he
had been living in the development.

3.2 Building Form

The ratio of social space to total built-up area was
calculated. The proportion of social areas in shade vs
without shade was estimated. These calculations
included both skyterraces and roofs above carpark.
Another consideration was the percentage of building
Figure 3: Building B axonometric showing skyterraces façade exposed to outdoor conditions. In Building A,
the façade facing the inner void was deemed ‘not
This study set out to assess performance, as exposed’. For Building B, the façade adjacent to
stipulated by the architects, and to gauge the extent circulation corridors, voids and staircases was
to which performance can be linked to building form. likewise categorised ‘not exposed’.

3. METHODOLOGY 3.3 Behavioural Observations

The survey study began in August 2019 with Skyterraces in both buildings were visited every
residents of each building. Table 1 summarises the two hours from 14:00 until 18:00 (i.e. a total three
number of survey respondents in relation to the total times per afternoon) from August 6 to 9 and again,
number of apartments per building. This was between August 12 to 14. During each visit, the
augmented with behavioural observations, number of visitors was counted. Counting was carried

Vol.1 | 62
35th PLEA Conference. Planning Post Carbon Cities.
DOI: https://doi.org/10.17979/spudc.9788497497947
out in the afternoons since occupants were observed
to visit skyterraces mostly during the afternoon and Table 3: Summary of responses to social questions.
evening, as observed during surveys. Building A Building B
Q4. Do you visit the skyterraces of your own estate?
Yes (1) 46 (93.9%) 36 (78.3%)
3.4 Measurement of Environmental Conditions No (0) 3 (6.1%) 10 (21.7%)
The air temperature (Ta) of skyterraces and Q5. How often do you visit the skyterraces?
outdoor conditions was measured for 6 days (July 25 Every day or many times a week 18 (39.1%) 5 (13.9%)
– July 30, 2019). Air velocity (Va) was measured on Once a week 13 (28.3%) 12 (33.3%)
Once or twice a month 15 (32.6%) 19 (52.8%)
day 1 and 2 of that same period, but only for Building Q6. How much time do you spend on them?
A. < 30 minutes 18 (41.9%) 16 (44.4%)
Only skyterraces linked to towers of buildings A >= 30 minutes 25 (58.1%) 20 (55.6%)
and B were selected for temperature measurements Q7. Why do you visit the skyterraces and roof?
(i.e. excluding social spaces on roofs of carparks). a. Socialize with neighbours 13 (30.2%) 7 (19.4%)
b. Look at the views 27 (62.8%) 16 (44.4%)
Three skyterraces on 14th, 25th and 36th storeys of c. To take kids to play 10 (23.3%) 15 (41.7%)
Building A was selected, along with two on the 18th d. To exercise 5 (11.6%) 19 (52.8%)
and 33rd storeys of Building B, which were at a similar Q8. Are the skyterraces used by your neighbours?
height. 1 Yes, a lot 7 (14.9%) 5 (10.9%)
2 Yes, somewhat 26 (55.3%) 15 (32.6%)
Readings were taken at 4pm, when the day was 3 Yes, but not so much 12 (25.5%) 20 (43.5%)
typically warmest. Reference ‘outdoor’ readings were 4 Not at all 2 (4.3%) 6 (13.0%)
taken on the roof of carparks of both buildings; the Q9. How many neighbours in your estate are you friend with?
sensors here were sheltered from direct sun and 0-9 neighbours 16 (35.6%) 16 (44.4%)
10-19 neighbours 8 (17.8%) 3 (8.3%)
placed less than 50 meters away from nearest
20-29 neighbours 3 (6.7%) 1 (2.8%)
skyterrace. Q10. In terms of your thermal comfort, how do you generally feel
in the skyterraces of your estate?
3.5 Statistical Analysis +2 (very comfortable) 8 (16.7%) 3 (7.3%)
Non-parametric statistical tests were carried out +1 (Comfortable) 36 (75.0%) 21 (51.2%)
0 (Neutral) 4 (8.3%) 13 (31.7%)
on data from the surveys. For Q1, Q8, Q9 and Q10 -1 (Uncomfortable) 0 (0.0%) 4 (9.8%)
Mann-Whitney’s two sample test was performed; for -2 (Very uncomfortable) 0 (0.0%) 0 (0.0%)
Q2 and Q3 Fisher’s exact test; for Q4, Q5, Q6 and Q7
Pearson’s Chi-squared test. The analyses were done Table 4: Summary of building form variables
with Rstudio software. Rstudio functions chisq.test, Building A Building B
fisher.test, wilcox.test and shapiro.test were used, Skyterrace area (social space)
Shaded 13,404 m2 (63.2%) 582.6 m2 (11.8%)
respectively. Unshaded 7,800 m2 (36.8%) 4,352.2 m2 (88.2%)
Mann-Whitney’s two sample test was used to Total 21,204 m2 (100%) 4,934.8 m2 (100%)
compare measurements of buildings A and B. Gross floor area [13] 111,106 m2 87,000 m2
Skyterrace as % GFA 19.1% 5.7%
Percentage of façade directly exposed to outdoor conditions
4. RESULTS
Directly exposed 43.3% 53.6%
Tables 2 and 3 summarise responses to survey
questions. Table 4 summarises building form Table 5: Mean responses and statistical significances of the
variables. Table 5 summarises mean values for each difference between Building A and Building B.
question and statistical significance (p-value) of the Building A Building B p-value
difference between the two buildings. Energy-related aspects
Average self-reported $107.3 $130.4 0.038
Table 2: Summary of responses to energy questions. electricity bill (Q1) (n=31) (n=37) (*)
Average number of dwellings 98.0% 100% 0.477
Building A Building B
with AC (Q2) (n=49) (n=45)
Q1. What is your approximate monthly electricity bill?
Average number of dwellings 93.5% 97.7% 0.085
<$100 (SGD) 2 (6.5%) 1 (2.7%)
that turn on AC at night (Q3) (n=46) (n=44) (.)
$50-$99 (SGD) 13 (41.9%) 11 (29.7%)
Social aspects
$100-$149 (SGD) 12 (38.7%) 13 (35.1%)
$150-$199 (SGD) 2 (6.5%) 7 (18.9%) Average number of 10.2 7.5 0.0710
$200-$249 (SGD) 2 (6.5%) 5 (13.5%) neighbours that residents (n=27) (n=20) (.)
consider as friends (Q9)
Q2. Do you have installed AC in your home?
Average percentage of 93.9% 78.3% 0.002
Yes 48 (98.0%) 45 (100%)
residents visiting skyterraces (n=49) (n=46) (**)
No 1 (2%) 0 (0.0%)
(Q4)
Q3. At what time of the day are they usually turned on?
Average percentage of 39.1% 13.9% 0.000
Morning 0 (0.0%) 0 (0.0%)
residents visiting skyterraces (n=46) (n=36) (***)
Afternoon 1 (2.4%) 0 (0.0%)
every day or many times a
Evening 2 (4.8%) 0 (0.0%)
week (Q5)
Night 43 (93.5%) 43 (97.7%)
Average percentage of 58.1% 55.6% 0.886
residents that spend more (n=43) (n=36)

Vol.1 | 63
35th PLEA Conference. Planning Post Carbon Cities.
DOI: https://doi.org/10.17979/spudc.9788497497947
than 30 minutes on On perceived thermal comfort, 91.7% of
skyterraces (Q6) surveyees in Building A say they feel ‘very
Average likelihood of 2.19 2.59 0.008
neighbours visiting (n=47) (n=46) (**) comfortable’ or ‘comfortable’; in Building B, the
skyterraces, in Likert scale figure is 58.5%. The mean answer on the comfort
(Q8) scale is +1.07 for Building A (i.e. towards greater
Average thermal comfort in +1.07 +0.53 0.000 perceived comfort); the mean answer for Building B
skyterraces, in a 5-point (n=48) (n=41) (***)
thermal comfort scale (Q10)
respondents is 0.53 (i.e. towards neutrality).
Environmental conditions
ΔT (Ta outdoor – Ta 1.39ºC 0.78ºC 0.003 4.2 Building Form
skyterrace) (n=6) (n=6) (**) Building A has four times more surface area for
Air temperature at 29.65ºC 29.81ºC 0.189 skyterraces than B: 21,204 m2 and 4,935 m2,
skyterrace (n=6) (n=6)
Air temperature outdoors 31.04ºC 30.59ºC 0.115 respectively. As a proportion of total built up area,
(n=6) (n=6) skyterraces in Building A account for 19.1%; Building
Air velocity at skyterrace 1.39m/s - - B, 5.7%. In Building A, the surface area of skyterraces
Significance codes: >0.1 (.) < 0.05 > (*) < 0.01 > (**) < 0.001 > (***) that is shaded is 63.2%; in Building B, 11.8%. The
percentage of facade in A that is exposed to outdoors
4.1 Survey is 43%; in B it is 54%
Analysis of data suggests that the two groups,
Building A vs Building B, are not statistically different 4.3 Behavioural Observations
for any background variable. Normality is not found in Skyterraces in Building A account for 257 visitors
their answers either. during the period of measurement; Building B, 60
On energy use, the reported electricity bill of (Table 6). Normalised against number of households,
surveyees in Building A is lower than Building B by Building A has higher visitorship per household.
almost $23. Building A has a mean of $107.3; Building
B, $130.4. This finding has statistical significance. Table 6: Counting of people visiting skyterraces per building
Note: data is filtered to surveyees between 19 and 59 during period of observations (6-9, 12-14 August)
years of age, who are more likely to be aware of the Building Number of people Number normalised against
on skyterraces number of households
monthly bills.
Building A 257 0.27 per household
On use of AC, even though the two buildings are Building B 60 0.08 per household
not significantly different in ownership of AC,
approximately 93% of those from Building A say they 4.4 Environmental factors
use AC at night compared with almost 98% in Building The mean Ta of Building A is 0.16ºC lower than
B. This finding is marginally significant. Note: the that in B. This difference is not statistically significant.
analysis is filtered for surveyees who say they use AC Measured outdoor temperatures at both buildings
at night since this represents over 90% of responses in are not significantly different and are highly
both buildings. correlated (R = 0.92). However, the mean ΔT in A is
On use of skyterraces, 94% of surveyees from almost 1.4ºC (14thF: 0.99ºC, 25thF: 1.41ºC, 36thF:
Building A say they visit skyterraces, which is found to 1.79ºC), while in B ΔT mean is almost 0.8ºC (19thF:
be significantly higher than 78% of surveyees from 0.67ºC, 33thF: 0.89ºC). The difference between the
Building B. Additionally, skyterraces in A are visited two is statistically significant. A mean air velocity of
more frequently than in B, with approximately 39% of 1.85 m/s is measured in Building A (14thF: 2.26 m/s,
surveyees in former saying ‘every day or many times 25thF: 1.82 m/s, 36thF: 1.48 m/s).
a week’ compared with 14% in the latter. The
perception of neighbours visiting terraces in Building 5. DISCUSSION
A tends towards ‘yes, somewhat’; in Building B it is The architects for Building A, WOHA, have said
closer to ‘yes, but not so much’. they seek, through design, three outcomes: improved
On why skyterraces are visited, 30.2% in Building occupant comfort, lower energy use, and greater
A say ‘socialise with neighbours’ compared with social interaction [8,14]. The findings from this study
19.4% in Building B. The finding is not statistically suggest that Building A does better on all three
significant. counts than Building B. However, in what ways can
Regarding social interaction, surveyees in Building the success of Building A be linked, directly or
A say that they consider, on average, 10.2 neighbours indirectly, to its built form?
as friends in comparison to surveyees on Building B The distinguishing feature of Building A is its inner
who consider 7.5 neighbours as friends. The core – made up of voids and skyterraces – that acts as
difference is marginally significant. Note: analysis is a conduit for natural air-flow and holds spaces for
limited to those who say ‘yes’ to visiting skyterraces social interaction. As a result of this core, parts of the
and say they spend more than 30 minutes per visit, so building envelope are inward facing. By contrast,
as to eliminate those who are just passing through.

Vol.1 | 64
35th PLEA Conference. Planning Post Carbon Cities.
DOI: https://doi.org/10.17979/spudc.9788497497947
Building B is compact, without voids or inner facades. regulatory limit for Residential Envelope Transfer
Its skyterraces are fewer and affixed to its façade, and Value (RETV) of 25 W/m2 [15].
therefore more exposed to outdoor conditions. From this study, it is evident that Buildings A and
The first form hypothesis relates to comfort in B are two distinct form typologies. Findings suggest
skyterraces. The sheltered skyterraces of Building A that form is a likely factor affecting performance in
should fare better in terms of thermal comfort than multiple ways. The significance of these findings
the exposed ones in Building B. The findings from this become clearer when they are extrapolated to the
study support this. There is a measurable difference urban scale. Singapore has 1 million HDB flats [16]. If
in the mean thermal comfort response of the two the entire stock of housing were to perform at the
groups: +1.07 for A versus +0.53 for B, i.e. the former same level as Building A, i.e. spending 17% less on
is more comfortable than the latter. This is energy, the impact at the city scale would be a saving
corroborated by temperature and air velocity of 729 GWh/year, based on a total of 4,287 GWh
readings. In Building A, ΔT between skyterraces and consumed in 2017 by public households [17,18]. This
outdoors is 1.39ºK, almost twice the ΔT in Building B. is equivalent to a reduction of 0.305 million metric
The mean Va in Building A voids is approximately 1.9 tons of equivalent CO2 emissions, based on
m/s. Lower temperatures and elevated air speeds Singapore’s 2018 Grid Emission Factor of 0.4188 kg
should lead to high perceived comfort. CO2/kWh [17]. If every HDB household were to also
It is likely that comfort is a variable affecting interact with 36% more neighbours, is likely that
visitor preferences. In Building A, where comfort social capital of the city would increase. Social capital
levels are higher, 94% of those surveyed say they visit is defined as the number of relationships between
skyterraces; in Building B, only 78%. Thirty-nine people in group that leads trust and cohesion [19].
percent in Building A also say they visit ‘every day or
many times a week’ compared with 14% in B. These 6. CONCLUSIONS
findings appear to align with observed visitor Thermal comfort, energy use and social
numbers: skyterraces in A have 257 visitors versus 60 interaction are complex outcomes, affected by many
in B. Normalised against the number of households, variables. In this study, building form is found to be a
skyterraces in Building A (0.27) appear to be more factor that contributes to each outcome in direct and
popular than in B (0.08), assuming all observed indirect ways. In Building A, the presence of sheltered
visitors are residents of the same building. It should and comfortable social spaces appears to lower
be noted there are other reasons for visitorship. The barriers to neighbourly interactions. The inner void
survey suggests ‘view’ is a factor. Skyterraces in and skyterraces act as a nexus of social interaction.
Building A are also substantially bigger than the ones The core also affects the energy performance of the
in Building B, with room for more people. apartments. What is seen in Building A, therefore, can
The second form hypothesis relates to social be described as the integration of social and
engagement: residents of Building A, who visit environmental objectives through form-based
skyterraces more frequently, ought to know more solutions.
neighbours. This is supported by the findings: Architects and researchers in the tropical regions
surveyees from Building A say they know an average have in the past argued for the importance of form
10.2 neighbours, compared with 7.5 in Building B, i.e. features and strategies, such as open-to-sky
36% more. The difference is found to be marginally courtyards and sunshades, however, primarily for
significant. social or place-making purposes [20, 21]. The notion
The third form hypothesis relates to energy use. that a form-based design approach can
Lower temperatures and higher air flows in the simultaneously affect multiple outcomes in high-rise
central void of Building A are likely to affect energy typologies is rare. Architect Ken Yeang made a case
use. Apartments that open onto this cooler core can for this in the 80s and 90s, applying the bioclimatic
divert air flow through their living spaces, thereby model to office buildings in Malaysia which were said
reducing the need for mechanical cooling. Inner to deliver better energy performance and occupant
facades, opening onto a cool void, are likely to comfort. Two noteworthy buildings of that era were
transmit lower solar heat gain into the apartments. evaluated in a study, with surveys and energy audits,
The findings show a difference in energy use. The that revealed them to be unsuccessful in both regards
monthly energy bill is 17% lower in Building A [2]. This was attributed to an inconsistent application
($107.3) than in Building B ($130.4), notwithstanding of bioclimatic principles and to the underestimation
identical ownership of air conditioners between the of comfort expectations and preferences.
groups. Several variables that might affect energy WOHA’s approach to form, represented by
consumption can be ruled out. Both buildings have Building A, differs from these earlier experiments in
near identical demographics; they rely on a similar two ways. It sees environmental performance and
palette of materials and comply with the same social engagement as interdependent outcomes. The

Vol.1 | 65
35th PLEA Conference. Planning Post Carbon Cities.
DOI: https://doi.org/10.17979/spudc.9788497497947
building also creates an onsite microclimate with 5. Gamero-Salinas, J.C., Monge-Barrio, A., Sánchez-Ostiz,
sheltered inner voids. These spaces are more A., (2020). Overheating risk assessment of different
comfortable; they also reduce thermal load and dwellings during the hottest season of a warm tropical
climate. Building and Environment, 171, 106664. DOI:
enhance the potential for cross ventilation. This form
10.1016/j.buildenv.2020.106664
strategy is seen in other WOHA projects like the
6. Yeang, K., Balfour, A., Richards, I., (1994). Bioclimatic
School of the Arts and Oasia Hotel Downtown [8,14] Skyscrapers. Artemis, London
and in some of Yeang’s more recent projects such as 7. Kishnani, N., (2019). Ecopuncture: Transforming
the National Library in Singapore [22]. Architecture and Urbanism in Asia. BCI Media Group
Cities across the globe struggle to address 8. Bingham-Hall, P., WOHA, (2016). Garden City Mega
environmental goals and social goals, which are City: Rethinking Cities for the Age of Global Warming.
sometimes at odds. Singapore offers lessons on Pesaro Publishing
integration in the high-density tropical context [23]. 9. Powell, R., Chan, S., (2004). SCDA: The Architecture of
Soo Chan. Images Publishing
In this study, Building A demonstrates how this idea
10. SCDA. Skyterrace@Dawson. Available:
can be advanced further with form-based solutions at
http://www.scdaarchitects.com/ (accessed March 11,
the building scale. Lessons learnt here are particularly 2020).
relevant to developing countries where capital 11. Singapore Power Group (SGP), Billing, (2020). Avaiable:
investment and access to technology are limited. https://www.spgroup.com.sg/what-we-do/billing (accessed
The limits of the current study should be February 12, 2020)
countered in future research by increasing survey 12. Becker, R., Paciuk, M., (2009). Thermal comfort in
sample sizes and accessing actual energy bills. It residential buildings – Failure to predict by Standard Model,
should consider other factors that influence thermal Building and Environment, 44 (5): p. 948-960.
13. Council on Tall Buildings and Urban Habitat (CTBUH),
comfort such as radiant temperature, relative
The Skyscraper Center. https://www.skyscrapercenter.com
humidity, activity level, and clothing. In the Singapore
(accessed May 5, 2020).
context, it would be necessary to compare Buildings A 14. Wong, M.S., Hassell, R., Yeo, A., (2016). Garden City,
and B, both relatively new, with earlier generations of Megacity: Rethinking Cities for the Age of Global Warming.
public housing typologies, which had different sizes CTBUH Journal, 2016 Issue IV, p. 46-51
and arrangements of social space. 15. Building and Construction Authority (BCA), (2008). Code
on Envelope Thermal Performance for Buildings.
ACKNOWLEDGEMENTS 16. Housing and Development Board (HDB), (2020). Public
We would like to deeply thank the Friends of the Housing – A Singapore Icon. Available:
https://www.hdb.gov.sg/ (accessed February 13, 2020).
University of Navarra. Also, we would like to thank
17. Energy Authority Market (EMA), (2019). Singapore
the University of Navarra, Obra Social ‘la Caixa’ and
Energy Statistics 2019. Available:
Caja Navarra Bank Foundation for the mobility grant https://www.ema.gov.sg/Singapore-Energy-Statistics-2019/
that allowed the research stay at National University (accessed March 11, 2020)
of Singapore (NUS). Special thanks to Mr. Wong Mun 18. Energy Authority Market (EMA), (2018). Singapore
Summ, founding director of WOHA Architects, for Energy Statistics 2018. Available:
supporting this research. We also acknowledge the https://www.ema.gov.sg/cmsmedia/Publications_and_Stati
kind permission from the Housing & Development stics/Publications/ses/2018/index.html (accessed March
Board (HDB) and Tanjong Pagar Town Council to 12, 2020)
19. Ramboll Foundation, (2016). Strengthening Blue-Green
access the public housing shown in this paper. Special
Infrastructure in our cities. Enhancing blue –green
thanks to Prof. Jesús Fidalgo of the University of
Infrastructure & Social performance in high density urban
Navarra for his guidance in the statistical analysis. environments. Available:
https://ramboll.com/megatrend/~/media/B350ABC1D9D04
REFERENCES 89AA54955B43D853EE9.ashx (accessed May 8, 2020)
1. Olgyay, V., (2015). Design with Climate: Bioclimatic 20. Bay, J.H., (2004). Sustainable community and
Approach to Architectural Regionalism - New and Expanded environment in tropical Singapore high-rise housing: the
Edition. Princeton University Press case of Bedok Court condominium. Cambridge University
2. Kishnani, N., (2002). Climate, Buildings and Occupant Press.
Expectations: a comfort-based model for the design and 21. Powell, R., Tay, K.S., Lim, A.K.S., (1997). Line, edge &
operation of office buildings in hot humid conditions (PhD shade: the search for a design language in tropical Asia: Tay
Thesis) Kheng Soon & Akitek Tenggara. Page One Publishing
3. Brotas, L., Nicol, F., (2016). Using Passive Strategies to 22. Hart, S., (2011). Ken Yeang’s National Library of
prevent overheating and promote resilient buildings. In: Singapore. Architecture Week.
PLEA 2016 Los Angeles – 32nd International Conference on http://www.architectureweek.com/2011/1026/environme
Passive and Low Energy Architecture. Cities, Buildings, nt_2-2.html
People: Towards Regenerative Environments 23. Centre for Liveable Cities and Urban Land Institute,
4. Hashemi, A., Khatami, N., (2017). Effects of Solar Shading (2013). Lessons from Singapore 10 Principles for Liveable
on Thermal Comfort in Low-income Tropical Housing. High-Density Cities. Available: http://www.clc.gov.sg/
Energy Procedia, 111, p.235-244 (accessed March 10, 2020).

Vol.1 | 66
35th PLEA Conference. Planning Post Carbon Cities.
DOI: https://doi.org/10.17979/spudc.9788497497947