Energy and Buildings 141 (2017) 175–185

Contents lists available at ScienceDirect

Energy and Buildings

journal homepage: www.elsevier.com/locate/enbuild

Daylight performance and users’ visual appraisal for green building

offices in Malaysia
Gene-Harn Lim a,∗ , Michael Barry Hirning b , Nila Keumala a , Norafida Ab. Ghafar a
a
Faculty of Built Environment, Universiti Malaya, Kuala Lumpur, Malaysia
b
IEN Consultants, Kuala Lumpur, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: Lighting energy savings, as well as visual and non-visual user benefits have been widely attributed to
Received 13 August 2016 daylighting. This paper explores daylight design strategy, visual appraisal, Daylight Factor (DF), lighting
Received in revised form 23 January 2017 energy usage and discomfort glare using two green building offices in Malaysia, which have incorporated
Accepted 11 February 2017
daylighting into both façade and interior design. Visual appraisal surveys were collected from 39 and
Available online 20 February 2017
145 subjects in the open plan working space of the Energy Commission Building (ECB) and Public Works
Department Block G (PWD), respectively. The survey focused on task brightness, colour appearance,
Keyword:
uniformity and lighting preference. Discomfort glare assessed via occupant point-of-view luminance
Daylight factor
Daylighting
maps was juxtaposed here from a glare study involving the same buildings. Illuminance loggers were
Discomfort glare used to monitor artificial lighting usage as well as the DF on a selected floor of each building. There were
Green building no significant differences in occupant responses to the visual appraisal survey for both office spaces. Using
Visual appraisal MS1525:2014 and Green Building Index (GBI NRNC) tool as baselines, the DF performance of both offices
differs significantly: PWD had a 45.5% daylit area, with ECB a 14.8% daylit area for DF >1%. However,
lighting energy usage results show substantial savings of 53% and 41% occurred from daylighting. These
findings of visual appraisal, DF, lighting energy savings and discomfort glare show a discrepancy in using
only the DF to justify the daylight performance of an office space in a tropical climate such as Malaysia. The
findings suggest that equivalent consideration should be given to interior design to facilitate daylighting,
which is often beyond the control of designer, but in the hands of office end users.
© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction which similarly evaluate the sustainability of a building, taking

In the wake of energy efficiency practices to mitigate the effects
of climate change, various sustainable building assessment tools
have been established since the late 20th century [1]. The build-
ing sector has the highest potential to reduce its carbon output
given the same mitigation costs across various other sectors such as
transportation and agriculture [2]. In response, the Malaysian gov-
ernment intends to reduce its greenhouse gas emissions intensity
by 45% (scaled by GDP) by the year 2030 from its 2005 intensity by
introducing various environmental related policies and programs
[3–5]. One such program was the establishment of Malaysia’s Green
Building Index (GBI) in 2009, which signified the start of the green
building movement in Malaysia [1].
To date, there are multiple green building certifications in
use in Malaysia, such as LEED, Green Mark, GBI and MyCREST,
∗ Corresponding author.
E-mail address: limgeneharn@gmail.com (G.-H. Lim).
into account various aspects of design and construction; such as
energy efficiency, water efficiency, indoor environmental quality
(IEQ), sustainable site management and materials & resources [1,3].
Daylighting is an important aspect of green building design. The
benefits of good daylighting to both energy efficiency and visual
comfort are well known, as are the consequences [6–9]. Presently,
the justification for what constitutes “good” versus “bad” daylight
design in green buildings is measured via the Daylight Factor (DF).
This paper evaluates the success of the Daylight Factor in producing
a well day-lit space in the tropics using two GBI platinum rated
government office buildings in Malaysia.
1.1. Daylighting and visual appraisal
The human preference for daylight over artificial light for office
spaces is well established [7,10–12]. Daylight both stimulates and
regulates our circadian system, which subsequently affects our
alertness and mood [13,14]. It also provides variation of lumi-
nance and colour that influence the attractiveness and desirability

http://dx.doi.org/10.1016/j.enbuild.2017.02.028
0378-7788/© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.
0/).
176 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185
of spaces [15]. Provision of windows satisfies the occupants desire
for a connection to the outside, consequently improving their mood
[12,16]. Carter & Marwaee [17] found that people were more sat-
isfied with the lighting when they had windows, even if windows
were not appreciably affecting the lighting at their task location.
Galasiu & Veitch [12] also found that glare may be better toler-
ated from daylight than from artificial light if there was a good
view available. A survey of readers’ satisfaction at Raja Tun Uda
Public Library in Malaysia found 74% of the respondents agreed
that their seating preference is affected by daylight [18]. There
are also claims that daylight improves work productivity, however
as pointed out by Sullivan, Donn, & Baird [19], the findings from
laboratory research are not very reliable as people usually take
days to adjust to a different luminous environment. This intangible
connection between productivity and daylight has constrained the
economic feasibility aspect of daylighting to only consider energy
savings, which does not take occupants’ well-being into account
[20].
1.2. Daylighting and energy efficiency
Daylighting can reduce the reliance on artificial lighting, which
has been shown to help reduce the cooling load and building
energy demand [21]. This is possible as diffuse daylight has a higher
luminous efficacy, 110–130 lm/watt, than most artificial lighting,
70–100 lm/watt [22]. Yu & Su [8] reviewed 20 papers and found
daylight harvesting can lead to energy savings in lighting of 20–87%.
However, these calculations were assessed from non-field mea-
surement methods such as simulation and algorithm calculations.
In the context of tropical skies, Kamaruzzaman et al. [23] eval-
uated the Klang District Office in Malaysia and found an average
lighting consumption saving of 37% due to daylighting. However, it
is important to note that occupancy and usage trends will impact
energy savings. A study of 4 offices in Korea revealed that despite
a 43% reduction in lighting energy use due to automatic dimming,
the change in occupancy patterns lead to an increase in lighting
energy use by up to 50% [24].
1.3. Current recommendations for daylighting office spaces in
Malaysia
The local Malaysia standards required for new green office
buildings are the Green Building Index Non-Residential New Con-
struction (GBI NRNC) tool and MS1525:2014 Code of Practice for
Energy Efficiency and Use of Renewable Energy for Non-Residential
Buildings. Under the daylighting credit EQ8 (for GBI NRNC) points
are awarded based on the percentage of coverage of the Net Let-
table Area (NLA) that achieves a Daylight Factor (DF) of 1.0–3.5%
measured at the work plane [25]. A separate credit on the Post Occu-
pancy Comfort Survey, EQ15, requires that if 20% of occupants are
dissatisfied with the overall comfort, including lighting level and
glare problems, then corrective action must be taken [25]. There is
no specific credit awarded for reducing lighting energy from day-
light harvesting; however, there is a collective credit that looks
into the reduction of total building energy consumption (Credit EE5
Advanced EE Performance). This means that in design it is possible
to score credits in daylighting through DF performance alone. This
requires a simulation model without any interior fittings during the
Design Assessment stage, which is not representative of the actual
interior office condition [26].
MS1525:2014 briefly mentions a recommended DF of 1.0–3.5%,
the same as GBI NRNC EQ8. It also recommends an average illumi-
nance of 200 lx for infrequent reading and 300–400 lx for general
office spaces [27]. However, these illumination ranges are for artifi-
cial lighting only and not optimized for skylight or daylight design
in Malaysia [7,28,29].
1.4. Daylighting in Malaysia
A survey of 41 rooms in 5 office buildings across Malaysia
showed that none of these office spaces achieved more than 0.5%
DF due to the deployment of internal shading devices [28]. Despite
the offices having deep overhangs to block direct sun, glare from
the high luminance ratio of the window to internal surfaces caused
occupants to engage their internal shading. Lim et al. [30] identified
18 government offices in Malaysia with identical façade design, and
placement of individual occupant rooms at the perimeter. A simu-
lation study showed that a light shelf could have reduced excessive
daylight illuminance and improved uniformity in these offices.
A recent (2016) survey of discomfort glare in six office buildings
in Malaysia, including three GBI certified green office buildings,
found the most common source of glare in green buildings came
from windows [31]. With glare from windows in green buildings
experienced by 35% of occupants compared to just 7% in non-green
office buildings [31]. In tropical climates, a view of the bright sky is
a major glare concern. These findings highlight that façade design
principles adopted from temperate climates may not adequately
utilize daylight in the tropics [32].
2. Methodology
2.1. Buildings
This daylight performance study took place in two government
offices, the Energy Commission Building (ECB) and Public Works
Department Block G (PWD) located in Putrajaya and Kuala Lumpur
respectively (Figs. 1 and 2). The ECB, completed in 2010, is a multi-
award winning green building which obtained GBI Platinum and
Green Mark Platinum awards in the Non-Residential category. It
was recognised as the ASEAN Energy Award Winner (2012) and
ASHRAE Technology Award Runner-Up (2013) [33]. Completed in
2013, PWD was the first high rise building to be certified GBI Plat-
inum [34].
Nikpour et al. [35] explicitly studied the daylight quality of ECB.
Despite having an office depth of 18 m, the daylighting strategy of
ECB focuses on using a light shelf and atrium to reflect only dif-
fuse daylight into the space. In addition to the removal of ceiling
panels and using white finishes for internal surfaces, the cubicle
was designed to facilitate daylight across the office space by using
translucent partitions (Fig. 1). Roller blinds were provided for occu-
pant’s visual comfort at the vision glazing below the light shelf,
while louvers between the light shelf and ceiling prevented low
angle direct sun from entering the space from above the light shelf.
Also of note is the usage of an automatic blind system at the atrium
opening which allows only diffuse light from the sky. This was
designed to block direct afternoon insolation, which would bring
excessive heat gain into the building.
PWD is a 37 storey office tower that allows daylight penetra-
tion by taking advantage of perforated horizontal louvers along the
vision window (Fig. 2). The closing angle of the horizontal louver is
limited to ensure there is a minimum opening for diffuse daylight
to enter. Fundamentally, both buildings have daylighting strate-
gies that allow occupants to manually adjust blinds in response to
brightness at the window whilst not jeopardizing the abundance of
diffuse daylight. Both buildings emphasize daylighting strategies,
though different in approach, in an attempt to achieve benefits in
energy efficiency and visual comfort [26]. Both office spaces use
efficient luminaires (Philips Essential 2 × 28W T5 Fluorescent) con-
trolled by on/off daylight sensors with a set point of 250 lx. The
 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185 177

Fig. 1. Pictures of Energy Commission Building office in Putrajaya.

Fig. 2. Pictures of PWD Block G Level 31 office in Kuala Lumpur.

Table 1 satisfied”, “Somewhat Satisfied”, “Satisfied”, “Strongly Satisfied”,

Comparison of daylight design strategy for both offices.
and also had an additional option of “No Opinion” at the end (see
Design Feature ECB PWD Appendix A). A six point Likert scale was chosen as the neutral
Window to Wall Ratio 50% 90% option of a 5 or 7 point Likert scale may be misrepresented as no
Glazing Transmittance 53% 54% opinion, thus causing misleading results [39]. Responses from pri-
Blinds Transmittance 8% 30% vate rooms or any other non-open plan working area were excluded
Ceiling/Wall/Floor 90%/70%/30% 90%/70%/30% from the results, as were surveys with responses of “No Opinion”.
Reflectance
Subsequently, a total of 39 and 145 surveys from ECB and PWD
Orientation of Openings All Orientations with Core All Orientations with
occupying South Segment Core occupying South were analysed using the Mann-Whitney U Test via IBM SPSS Soft-
Segment ware. The Mann-Whitney U Test is used to compare the differences
Notable Daylight Light Shelf, Atrium Perforated horizontal between two independent groups with an ordinal dependent vari-
Harvesting Strategy Opening blinds, high
able.
transmittance interior
partitions
Occupancy Density 27.15 m2 /pax 11.89 m2 /pax
Depth of Office Space 18 m 13 m
2.3. Daylight factor
Workspace Typology Mixture of private rooms Open Plan concept
and cubicles with with partitions below The DF is defined as the ratio of the internal horizontal illumi-
partitions above sitting sitting eye level nance on the work plane to the external horizontal illuminance
level
measured under CIE overcast skies. It has been well documented
Cubicle Partition <30% >90%
Transmittance that the CIE overcast sky is an idealised sky condition available only
in simulation and is impossible to observe in reality [43,44]. Three
units of TENMARS TM-203 illuminance loggers were synchronized
horizontal illuminance was designed to comply with MS1525:2014 to record at 1-s intervals. Two units (Logger B & C) were placed
(300–400 lx) [27]. The design features of each building are listed in on the roof of the building to ensure there were no surrounding
Table 1. obstructions. A mirror was used to shade one logger against the
direct sun while the other was left unobstructed (Fig. 3). Using this
2.2. Visual appraisal survey setup, the readings from loggers were compared to determine if
there was any direct sunlight. If both loggers registered a global
The parameters of the office space that affect daylighting were horizontal illuminance value below 60,000 lx, and both readings
evaluated by occupants via an online visual appraisal survey. The were within 10% of each other, the sky condition was considered
survey (in Malay language) was sent out by the administration of as overcast and suitable for DF calculation. This method eliminated
both office spaces during the measuring period in 2015 (August for relying solely on visual inspection to assess sky conditions. The sky
ECB, October for PWD). It assessed occupant demographics, con- condition in Malaysia is predominantly intermediate (85.6%), fol-
trollability of the lighting environment and visual appraisal under lowed by overcast (14%) [41]. Weather data from the measurement
paper work and computer work. It is known that visual appraisal period (August and October) was extracted from IWEC (Interna-
and cognitive performance differ under different luminous envi- tional Weather for Energy Calculations) weather data, which is
ronment parameters, such as illuminance, colour temperature derived from up to 18 years of hourly weather data (Table 2) [42].
and uniformity [36–38]. A 6 points Likert scale was used which This data was used a guide to accept only horizontal illuminance
consisted of “Strongly Dissatisfied”, “Dissatisfied”, “Somewhat Dis- readings below 60,000 lx for DF calculations. Concurrently, dur-
178 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185
Table 2
Comparison of Weather Data of September and November [42].
Month Global Horizontal Radiation Direct Normal Radiation
(Avg Daily Total) (Avg Daily Total)
August 4336 Wh/sq m 1044 Wh/sq m
October 4219 Wh/sq m 853 Wh/sq m
Fig. 3. Illustration of the DF Measurement Settings.
ing acceptable overcast conditions, the third logger (Logger A) was
used to measure the indoor illuminance at 800 mm from the floor
throughout the entire space. The illuminance measurement points
covered all work desks and walkways, with a spacing of 1.5–2 m
between points. All measurements were completed in about thirty
minutes for each building. The positions of the blinds were not dis-
turbed, and the electric lights were switched off. This procedure
measured a total of 201 and 304 points for ECB and PWD respec-
tively.
The DF was obtained by the illuminance ratio of Logger A to
Logger B. The DF performance was evaluated using two different
approaches. The first approach calculated the number of measured
points achieving the recommended DF range (1–3.5%) by GBI NRNC
and MS1525:2014. This was then divided by the total number of
measuring points to express the percentage of points in the recom-
mended DF range. The second approach inserts the DF value of each
coordinate into the layout plan. Then the area of the recommended
DF range is calculated from the layout plan (Figs. 10 and 11). Simi-
larly, the area of recommended DF was divided by the total area to
express the percentage of area in the recommended DF range. This
made it easy to compare both approaches.
2.4. Lighting circuit usage
A typical floor layout of each building was selected to conduct
both the DF (see Section 2.3) and lighting usage study. Level 31 of
PWD and Level 2 of ECB were selected for monitoring as they were
occupied by administration departments, and expected to have
consistent occupancy patterns throughout the measuring period.
The usage trend of the recessed artificial light circuits was
tracked with HOBO-U12 loggers (in 5 min intervals) over a one
month period during working hours (8.30 a.m.–5.30 p.m.) for both
ECB and PWD respectively. The loggers were hung 10 cm below
each lighting circuit, and were considered activated upon exposure
to high illuminances >2000 lx (Fig. 4). Another two units of HOBO-
U12 loggers were placed 1 m from the perimeter of the glazing
to record the vertical illuminance of daylight from windows. This
allowed a check of the responsiveness of electric lighting controls
to the available daylight.
A total of 11 and 6 lighting circuits were monitored for ECB and
PWD respectively (Fig. 5). Energy savings in lighting were com-
puted by considering the duration that lighting circuits remained
Diffuse Radiation Cloud Cover Global Horizontal
(Avg Daily Total) (Occurrence) Illumination (Avg Hourly)
3521 Wh/sq m 83%–92% 40,730 lx
3533 Wh/sq m 85%–92%. 40,203 lx
Fig. 4. Method of identifying luminaire activation.
inactive compared to if all light circuits were activated throughout
the working hours (8:30 a.m.–5:30 p.m.).
2.5. Discomfort glare
The visual comfort in both ECB and PWD were assessed inde-
pendently of the DF and lighting usage studies using data from
a published study by Hirning et al., assessing discomfort glare
in Malaysian buildings [31]. Both ECB and PWD were buildings
included in that study, which updated the Unified Glare Probabil-
ity (UGP); a discomfort glare metric derived from in-situ survey
responses and occupant point of view luminance maps. The UGP
is a modification of the UGR and was developed for assessing glare
specifically in open plan spaces in tropical and sub-tropical climates
(Eq. (1)) [31,45].
1
UGP =  10 (1)
  L2 ωs,i
 −15
2 1 n
1+ 7 Lb i=1
s.i
Pi
Where Ls is the luminance of the glare source, Lb is the background
glare luminance, ωs is the solid angle subtended by the glare source,
P is Guth’s position index and n is the number of glare sources.
Clear, intermediate and overcast sky conditions were observed
during the experiment. The visual comfort assessments for ECB
were conducted over two days under clear sky conditions during
August 2015, and over 4 days under overcast and intermediate sky
conditions for PWD in November 2015. A total of 68 and 93 occu-
pant luminance maps were obtained ECB and PWD respectively.
Though the exact dates of the survey and surveyed occupants do
not match, they do overlap. Therefore to characterise the discom-
fort glare of occupants within ECB and PWD, the UGP calculated for
occupant luminance maps was averaged in two categories: Occu-
pants situated in areas that fell within the recommended DF range
of 1–3.5% (shown in Figs. 10 and 11) and those occupants situated
outside this range. The UGP estimates the likelihood that an occu-
pant will experience discomfort glare by returning a value between
0 and 1, which can also be expressed as a percentage. This gives a
snapshot of the glare experienced by occupants in these buildings
during the survey period. It should be noted that during the data
collection period for ECB the automated atrium blind was not fully
in operation.
 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185 179

Fig. 5. Floorplan of ECB and PWD with démarcations of lighting circuits and open plan floor areas.

Fig. 6. Visual Appraisal for Paper Work and Computer Work for Both Office Spaces.

3. Results Table 3
Tabulation of Lighting Performance for Both Offices.

3.1. Visual appraisal survey Office Space ECB Office PWD Block G Office

Total measuring points 201 304

A complete visual appraisal survey was returned from 39 and recorded
145 users in ECB and PWD respectively. There were no significant Percentage of measuring 14.4% 42.4%
differences in both office users’ visual satisfaction between paper- points for DF >1%
Percentage of measured 14.8% 45.5%
based work and computer work (P > 0.05) (Fig. 6). This was similar
area for DF >1%
across all surveyed parameters i.e. brightness, colour appearance Average reduction in 41% 53%
and uniformity. Therefore, the survey results did not establish any artificial lighting energy
superiority in visual appraisal between the two office spaces (Fig. 7). due to Daylighting
Average UGP for DF >1% 0.45 (n = 35) 0.32 (n = 52)
More than 80% and 65% of users in ECB and PWD preferred hav-
Average UGP for DF <1% 0.33 (n = 33) 0.42 (n = 41)
ing some daylight in their office working space (Fig. 8). However,
there were 28% and 25% of occupants dissatisfied with task bright-
ness for paper work in ECB and PWD respectively. Similarly for area of daylit space compared to ECB using the recommended DF
computer work, 21% and 24%; of occupants were dissatisfied with range.
task brightness.
3.3. Lighting circuit usage
3.2. Daylight factor
The artificial lighting circuits’ usage was investigated to evaluate
The DF results for both calculation methodologies as outlined in the effectiveness of daylighting and lighting controls. Fig. 9 show
Section 2.3 are shown in Table 3. Figs. 10 and 11 graphically display selected results (from 1 month of usage data) for vertical illumi-
the DF results for ECB and PWD respectively. There was no signif- nance at the facade against the number of activated light circuits
icant difference between the two methods, with the percentage of for PWD. The results show that the lighting circuits in both build-
measuring points with DF >1% (14.4 and 42.4%) agreeing with the ings were responsive to daylight as expected. For example, Fig. 9
calculated area of DF >1% (14.8 and 45.5%) for both ECB and PWD shows a day in which half of the lighting circuits in PWD were
respectively. The results clearly show that PWD has a far greater switched off in the morning (8.30 a.m. to 12.00 p.m.) due to the
180 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185
Fig. 7. Satisfaction Percentage for Paper Work and Computer Work for Both Office Spaces.
Fig. 8. Cumulative Percentage of Users’ Lighting P.
Fig. 9. Extract of 2 days lighting circuit activation logging results for PWD.
sufficient daylight from east facing windows; with all monitored
lighting circuits switched off in the afternoon (12.00 p.m. to 4.00
p.m.).
The energy savings for lighting measured across the one month
period for ECB and PWD were 40.9% and 52.8% respectively
(Table 3). This shows a substantial interaction of the office space
with daylight. The calculation assumes a base case where there
is no daylight sensor at all, and the monitored lighting circuits
are activated throughout all working hours (8.30 a.m.–5:30 p.m.).
The activation frequency of the lighting circuits and the number
of luminaires in each circuit are used to obtain the lighting energy
usage.
3.4. Discomfort glare
Table 3 also shows the results of the UGP calculation from occu-
pant luminance maps. The visual comfort between the DF >1% and
DF <1% areas of the building are significantly different (p = ∼0.00) in
both ECB and PWD. ECB has more discomfort glare near the facades
(UGP 45%) compared to the interior (UGP 33%). However, the oppo-
site occurs in PWD, with more discomfort glare in the interior (UGP
42%) compared to near the façade (UGP 32%).
4. Discussion
4.1. Daylight performance and visual appraisal
Table 3 shows that ECB (14.8%) and PWD (45.5%) have vastly
different daylight performance in terms of measured area of DF
>1%. However, despite the different DF performance for both office
spaces, there is no statistically significant difference in occupants’
visual appraisal of both offices (Fig. 6). Though it should be noted
that the visual appraisal feedback is not confined to the effects of
daylight itself as both offices still use artificial lighting. However the
monitored lighting circuit usage shows a substantial dependency
on daylight (Fig. 9). Fig. 8 showed that more than 80% and 65% of
users in ECB and PWD, respectively, preferred having some daylight
in their office working space. This result agrees with Hirning et al.
[31] which found 62% of occupants in 6 office buildings in Malaysia
preferred working under daylight.
The dissatisfaction with task brightness for paper-based work
and computer work ranges from 21% to 28% of occupants. This find-
ing is consistent with a survey of more than 30,000 respondents
across 215 buildings in the United States and Canada, where it was
found that occupants in green buildings are on average more satis-
fied with indoor air quality and thermal comfort, but not in lighting,
office layout and acoustics [46]. Abbaszadeh et al. [46] found a cor-
relation between green building design with open plan offices that
have low or no partitions at all.
GBI NRNC credit EQ15 emphasises the need for corrective action
upon finding more than 20% of surveyed occupants are dissatisfied.
Even though the majority of occupants agreed that the brightness
level was sufficient, the dissatisfaction expressed by more than 20%
of occupant’s means there is concerning visual discomfort which
needs to be addressed in these spaces. The assessment using the
Unified Glare Probability (UGP) supports these findings, as the like-
lihood of glare on average in both buildings exceeds 20% (Table 3).
Similarly, Hirning et al. [31] found that 43% occupants in three green
buildings in Malaysia experienced discomfort glare from daylight.
It is difficult to compare the UGP results between the build-
ings as the sky conditions during assessment were very different.
However, the results show that visual comfort did not correlate
 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185 181

Fig. 10. Daylight Factor Performance of Energy Commission Building Level 2.

with DF results. Particularly in PWD, where overcast and interme-
diate sky conditions were present during data collection. Occupants
were more likely to experience discomfort glare in the deep plan
areas (DF<1% and UGP 42%) compared to near the façade (DF>1%
and UGP 32%). This is likely due to the perforated blinds used for
internal shading (Fig. 12). Occupants experience discomfort from
contrast between viewing the brightly lit blinds at the façade from
the darker less bright areas of the interior. The façade is not self-
shading like ECB, and it has limited external shading. Therefore the
interior blinds become backlit from the daylight that reached the
façade.
However, in ECB, this trend is reversed in the deep plan areas
(DF<1% and UGP 33%) and at the façade (DF>1% and UGP 45%). Glare
results from bright sky views or high luminance views of the atrium.
Occupants in the deeper plan spaces are protected from these high
contrast views by their partitions and the louvers above the light
shelf (Fig. 12). As the DF in most areas of the buildings are 2% or less
except very close to the facade, the designer would be misled into
concluding that these spaces would be glare free if using only the
DF under the GBI NRNC guidelines to assess visual comfort.
The DF measurements in PWD (45.5% daylit area) outperform
ECB (14.8% daylit area) as shown in Table 3. This can be attributed
to the self-shading façade design in ECB which reduces the effective
182 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185

Fig. 11. Daylight Factor Performance of PWD Block G Office Level 31.

sky component that affects daylight penetration [35]. Despite the
façade, self-shading design in ECB, Nikpour et al. [35] reinforces
that the light shelf and horizontal louvers have helped in pene-
trating diffuse daylight deeper into the office space. However, the
previous research only investigated two private office rooms in the
ECB with limited DF measurement points, only 4 and 9, respec-
tively. Furthermore, in the investigation of Nikpour et al. [35], the
roller blinds were seen to be rolled up fully, which explains why
the recorded DF near the window was as high as 4%.
Despite both offices having a façade design with an emphasis
on daylighting, the workspace design is visibly different (Fig. 12).
The workspace typology of ECB along the façade varies from private
rooms with plastered partitions and cubicles with partitions above
sitting eye level. Shelf space is provided for personal storage which
has subsequently cluttered the workspace. For ECB the high opacity
of partitions between workstations hampered daylight penetrating
beyond the cubicle closest to the vision window to the next deeper
cubicle. On the other hand, in PWD, the partitions for private rooms
and cubicles are highly translucent, and cubicle partitions are sit-
ting below eye level (Fig. 12). It was also intended design to place
the transition zone at the middle of the floor depth (Fig. 11). This
allowed the building to claim a lower DF (0.5%) qualification for
the transition zone, which allowed PWD to score more than 50% of
the office space area within the qualified DF range for GBI NRNC
credit EQ8 [26]. This confirms that the quantity and quality of day-
light penetrating into an office space depends on a large number of
design features and is not limited to façade design only [47].
 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185
Fig. 12. Perspectives of the Workspace Typology for Both ECB and PWD offices.
4.2. Ambiguity of DF metric application
While the usage of DF to justify the daylight effectiveness of an
office space via simulation is relatively well established, the usage
of DF to justify daylight performance is heavily criticized by many
[28,50–53]. The consensus is that DF is over simplified and is unable
to reflect complexity of daylight interaction with the office space.
It is common for many researchers and practitioners to be con-
fused over Daylight Factor (DF) and daylight level [51]. Even the
recently amended MS1525:2014 does not explicitly mention DF
to be measured under an overcast sky. The reference sky for DF,
which is the CIE overcast sky distinctively differs from the variable
and inconstant sky condition in Malaysia [28,44,52]. DF does not
take dynamic variations of daylight luminance into account as the
sun’s position changes under non-overcast skies.
Subsequently, various daylight performance metrics have been
proposed which take into account dynamic sky luminosity and
sun path, such as Daylight Coefficient (DC) [54], Daylight Auton-
omy (DA) [55], and Useful Daylight Illuminance [53]. However,
many researchers argue that illuminance metrics are unable to
adequately capture the light distribution of a space, especially in
regards to visual comfort [56–58]; and that luminance-based met-
rics, will either be used in conjunction with or supplant current
illuminance-based metrics in the future.
4.3. Justifying a daylight dependant office
Daylighting in an office space does not lead to energy savings
if there are no effective lighting control systems [7]. It was found
that there is substantial daylight interaction in both of the inves-
tigated offices which promoted lighting energy savings of 41% and
53% for ECB and PWD respectively. The energy savings in both
offices are significant and compare well with other similar stud-
ies [8,48,49]. From an energy savings stand point the daylighting
strategies employed in both these buildings are certainly effective.
However, the evident interaction with daylight also comes with
concern when around 20% of the occupants are not satisfied, as
stated in the post-occupancy evaluation visual appraisal survey.
The current green building rating tools, including GBI NRNC,
qualifies the daylight credit EQ8 solely by the performance of DF.
This paper has shown the discrepancy of DF performance in regards
to lighting energy savings, occupants’ visual appraisal and visual
discomfort. The lack of relation between these four aspects sug-
gests the need for a holistic sustainable approach of credits related
to daylighting.
As previous literature has suggested, DF is merely a metric to
justify the daylight performance. However, an office space with
183
daylight should show lighting energy savings [8], and have an effec-
tive lighting control system [12] which allows improvement in
visual appraisal [11,59,60]. It is suggested that green building rating
tools adopt holistic daylighting credits which comprise of lighting
energy saving with an office average illuminance value within the
recommended range of MS1525:2014 (200–400 lx) and satisfying
visual appraisal feedback. This will avoid green building rating tool
users from cherry picking credits within the convenience of cost
and implementation but implement it holistically.
5. Conclusion
This paper has conducted visual appraisal, Daylight Factor
field measurements, lighting energy savings and discomfort glare
assessment in two GBI platinum rated office buildings in Malaysia.
The comparison of design features between buildings has indicated
the importance of interior design to allow daylight penetration, in
addition to façade design to capture diffuse daylight. This is a com-
mon ‘design gap’ in the industry where there is a missing design
translation between the architect who designs the building ele-
ments while interior fit out is decided by the client. Despite the
difference in DF performance, both office spaces showed 41% and
53% energy savings from daylighting. While there are no significant
differences in visual appraisal by the occupants in both green office
buildings, it is important to note that more than 20% of subjects in
both buildings were dissatisfied with task brightness for paper and
computer-based work.
These findings support familiar arguments that DF is not an
accurate metric that reflects the daylight performance of a space.
This research shows a discrepancy in using the DF to justify an office
space as well day-lit, as there is no noticeable relation to lighting
energy savings, occupants’ visual appraisal or discomfort glare. The
paper recommends a holistic approach for green building rating
tools to award credits for daylighting based on lighting energy sav-
ings and satisfying visual appraisal which show the interaction of
occupants with daylight results in a glare free and workable envi-
ronment. There is also an urgent need for green building assessment
tools to use dynamic daylight metrics such as useful daylight illumi-
nance and daylight autonomy, as well as luminance-based metrics
for visual comfort, such as the UGP used to assess discomfort glare
in this study; as the DF is proven to be obsolete in previous literature
and this research.
Acknowledgement
This work was supported by the Universiti Malaya High Impact
Research Grant [H-130001-00-H000001].
Appendix A. Survey Questionnaire
1 What is your Gender?
(Male/Female)
2 Which category below includes your age?
(<30, 31–40, 41–50, >50)
3 How long have you been occupying this office work space?
(Less than 2 months, More than 2 months)
4 What is the type of work space you are using currently?
184 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185
(Enclosed Private Office, Open Office with Just Desks, Cubicles with
Partitions above sitting eye level, Cubicles with Partitions below sitting
eye level)
5 Do you think that your window view to the outside of the building
is important?
(Important, Not Important, No Opinion, No Viewing Windows, Oth-
ers)
6 Regarding the balance between electrical and natural light, which
do you prefer?
(Total reliance on electrical light, Predominant reliance on electrical
light with supporting daylight, Equal reliance on daylight and electrical
light, Predominant reliance on daylight with supporting electrical light,
Total reliance on daylight)
7 Please rate the level of satisfaction on the following lighting con-
ditions for READING AND WRITING (PAPER WORK) in regards to
your work space – Lighting Brightness, Lighting Colour Appear-
ance, Uniform Lighting Distribution
(Strongly Dissatisfied, Dissatisfied, Somewhat Dissatisfied, Some-
what Satisfied, Satisfied, Strongly Satisfied, No Opinion)
8 Please rate the level of satisfaction on the following lighting con-
ditions for TYPING AND READING (COMPUTER WORK) in regards
to your work space – Lighting Brightness, Lighting Colour Appear-
ance, Uniform Lighting Distribution
(Strongly Dissatisfied, Dissatisfied, Somewhat Dissatisfied, Some-
what Satisfied, Satisfied, Strongly Satisfied, No Opinion)
References
[1] E. Papargyropoulou, R. Padfield, O. Harrison, C. Preece, The rise of
sustainability services for the built environment in Malaysia, Sustain. Cities
Soc. 5 (2012) 44–51.
[2] O.R. Metz, P.R. Davidson, R. Bosch, L.A. Dave, IPCC, Summary for policymakers,
in: Climate Change 2007: Mitigation. Contribution of Working Group III to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change,
IPCC, 2007, 2007 (Accessed 12 July 2015) https://www.ipcc.ch/publications
and data/ar4/wg3/en/spmsspm-c.html.
[3] Z.A. Hamid, A.F. Roslan, M.C. Ali, F.C. Hung, M.S.M. Noor, N.M. Kilau, Towards a
national green buidling rating system for Malaysia, Malaysian Constr. Res. J.
14 (2014) 0–16.
[4] S.C. Chua, T.H. Oh, Green progress and prospect in Malaysia, Renew. Sustain.
Energy Rev. 15 (2011) 2850–2861, http://dx.doi.org/10.1016/j.rser.2011.03.
008.
[5] UNFCCC, Intended Nationally Determined Contribution of the Government of
Malaysia, 2015, pp. 1–6 http://www4.unfccc.int/submissions/INDC/Published.
[6] P. Ihm, A. Nemri, M. Krarti, Estimation of lighting energy savings from
daylighting, Build. Environ. 44 (2009) 509–514, http://dx.doi.org/10.1016/j.
buildenv.2008.04.016.
[7] M.F.M. Alrubaih, N.L.N. Alghoul, M. a. Shameri, O. Elayeb, Research and
development on aspects of daylighting fundamentals, Renew. Sustain Energy
Rev. 21 (2013) 494–505, http://dx.doi.org/10.1016/j.rser.2012.12.057.
[8] X. Yu, Y. Su, Daylight availability assessment and its potential energy saving
estimation – a literature review, Renew. Sustain. Energy Rev. 52 (2015)
494–503, http://dx.doi.org/10.1016/j.rser.2015.07.142.
[9] P. Xue, C.M. Mak, H.D. Cheung, The effects of daylighting and human behavior
on luminous comfort in residential buildings: a questionnaire survey, Build.
Environ. 81 (2014) 51–59, http://dx.doi.org/10.1016/j.buildenv.2014.06.011.
[10] P.R. Boyce, Reviews of Technical Reports on Daylight and Productivity, 2004
http://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:Reviews+
of+Technical+Reports+on+Daylight+and+Productivity+Reviews+by+:#0.
[11] L. Edwards, P. Torcellini, A literature review of the effects of natural light on
building occupants, Natl. Renew. Energy Lab. (2002) 58 (Accessed 13
December 2016) http://www.nrel.gov/docs/fy02osti/30769.pdf.
[12] A.D. Galasiu, J.A. Veitch, Occupant preferences and satisfaction with the
luminous environment and control systems in daylit offices: a literature
review, Energy Build. 38 (2006) 728–742, http://dx.doi.org/10.1016/j.enbuild.
2006.03.001.
[13] B. Plitnick, M.G. Figueiro, B. Wood, M. Rea, The effects of red and blue light on
alertness and mood at night, Light. Res. Technol. 42 (2010) 449–458, abstract
http://lrt.sagepub.com/content/42/4/449.
[14] A.R. Webb, Considerations for lighting in the built environment: non-visual
effects of light, Energy Build. 38 (2006) 721–727, http://dx.doi.org/10.1016/j.
enbuild.2006.03.004.
[15] P.R. Boyce, Chapter 3: Lighting Quality, 2nd ed., 2003 http://www.lightinglab.
fi/IEAAnnex45/guidebook/3 lighting.quality.pdf.
[16] L. Heschong, F. Oaks, Windows and Classrooms: A Study of Student
Performance and the Indoor Environment, 2003, http://dx.doi.org/10.1175/
1520-0450(1998)037<0414:TDFBIM>2.0.CO;2.
[17] D.J. Carter, M. Al Marwaee, User attitudes toward tubular daylight guidance
systems, Light. Res. Technol. 41 (2009) 71–88, abstract http://lrt.sagepub.
com/content/41/1/71.
[18] A.R. Othman, M. Aiera, M. Mazli, Influences of daylighting satisfaction at raja
tun uda public library, shah alam, Soc. Behav. Sci. 68 (2012) 244–257, http://
dx.doi.org/10.1016/j.sbspro.2012.12.224.
[19] J. Sullivan, M. Donn, G. Baird, Measuring Productivity in the Office Workplace,
2013.
[20] M. Fontoynont, K. Ramananarivo, T. Soreze, G. Fernez, K.G. Skov, Economic
feasibility of maximising daylighting of a standard office building with
efficient electric lighting, Energy Build. 110 (2016) 435–442, http://dx.doi.
org/10.1016/j.enbuild.2015.09.045.
[21] D.H.W. Li, J.C. Lam, S.L. Wong, Daylighting and its effects on peak load
determination, Energy 30 (2005) 1817–1831, http://dx.doi.org/10.1016/j.
energy.2004.09.009.
[22] P.J. Littlefair, The luminous efficacy of daylight: a review, Light. Res. Technol.
17 (1985) 162–182, abstract http://lrt.sagepub.com/content/17/4/162.
[23] S.N. Kamaruzzaman, R. Edwards, E.M.A. Zawawi, A.I. Che-Ani, Achieving
energy and cost savings through simple daylighting control in tropical
historic buildings, Energy Build. 90 (2015) 85–93, http://dx.doi.org/10.1016/j.
enbuild.2014.12.045.
[24] G.Y. Yun, H. Kim, J.T. Kim, Effects of occupancy and lighting use patterns on
lighting energy consumption, Energy Build. 46 (2012) 152–158, http://dx.doi.
org/10.1016/j.enbuild.2011.10.034.
[25] Green Building Index, GBI NRNC Non Residential New Construction Design
Tool, 2011, pp. 76.
[26] G. Reimann, Personal Communication with IEN Consultants, 2014.
[27] Departments of Standards Malaysia, MS1525–2014 Code of Practice on
Energy Efficiency and Use of Renewable Energy for Non Residential Buildings
(2nd Revision), 2014.
[28] K.M. Al-Obaidi, M.A. Ismail, A.M. Abdul Rahman, Assessing the allowable
daylight illuminance from skylights in single-storey buildings in Malaysia: a
review, Int. J. Sustain. Build. Technol. Urban Dev. 6 (2015) 236–248, http://dx.
doi.org/10.1080/2093761X.2015.1129369.
[29] M.Z. Kandar, M.S. Sulaiman, Y.R. Rashid, D.R. Ossen, M. Aminatuzuhariah,
Investigating daylight quality in malaysian government office buildings
through daylight factor and surface luminance, Int. J. Civil Archit. Struct.
Constr. Eng. 5 (2011) 52–57.
[30] Y.-W. Lim, M.H. Ahmad, Daylighting as a sustainable approach for high rise
office in the tropics, Int. J. Real Estate Stud. 8 (2013) 1–15.
[31] M.B. Hirning, G. Isoardi, V.G. Hansen, Prediction of discomfort glare from
windows under tropical skies, Build. Environ. (August) (2016), http://dx.doi.
org/10.1016/j.buildenv.2016.08.005.
[32] I.R. Edmonds, P.J. Greenup, Daylighting in the tropics, Sol. Energy 73 (2002)
111–121, http://dx.doi.org/10.1016/S0038-092X(02)00039-7.
[33] IEN Consultants, ST Diamond Building Platinum Award, (n.d.). http://www.
ien.com.my/projects/st.html (Accessed 11 August 2016).
[34] IEN Consultants, Menara Kerja Raya (KKR2) GBI Platinum CVA Certified, 2016
(Accessed 11 August 2016) http://www.ien.com.my/projects/kkr2.html.
[35] M. Nikpour, M.Z. Kandar, E. Mosavi, Investigating daylight quality using
self-shading strategy in energy commission building in Malaysia, Indoor Built
Environ. 22 (2013) 822–835, http://dx.doi.org/10.1177/1420326X12458512.
[36] F. Linhart, J.L. Scartezzini, Evening office lighting – visual comfort vs. energy
efficiency vs. performance, Build. Environ. 46 (2010) 981–989, http://dx.doi.
org/10.1016/j.buildenv.2010.10.002.
[37] B.M. Shamsul, C.C. Sia, Y. Ng, K. Karmegan, Effects of light’s colour
temperatures on visual comfort level, task performances, and alertness
among students, Am. J. Public Health Res. 1 (2013) 159–165, http://dx.doi.org/
10.12691/ajphr-1-7-3.
[38] G.R. Newsham, C. Arsenault, J.A. Veitch, A.M. Tosco, C. Duval, Task lighting
effects on office worker satisfaction and performance, and energy efficiency,
Leukos 1 (2005) 7–26, http://dx.doi.org/10.1582/LEUKOS.01.04.001.
[39] R.F. Guy, M. Norvell, The neutral point on a likert scale, J. Psychol. 95 (1977)
199–204, http://dx.doi.org/10.1080/00223980.1977.9915880.
[41] A. Zain-Ahmed, K. Sopian, Z. Zainol Abidin, M.Y.H. Othman, The availability of
daylight from tropical skies – a case study of Malaysia, Renew. Energy 25
(2002) 21–30, http://dx.doi.org/10.1016/S0960-1481(00)00209-3.
[42] EnergyPlus, 18 years hourly historical weather data, Int. Weather Energy Calc.
(2016) https://energyplus.net/weather-location/southwest pacific wmo
region 5/MYS//MYS Kuala.Lumpur.486470 IWEC.
[43] A. Nabil, J. Mardaljevic, Useful daylight illuminances: a replacement for
daylight factors, Energy Build. 38 (2006) 905–913, http://dx.doi.org/10.1016/j.
enbuild.2006.03.013.
 G.-H. Lim et al. / Energy and Buildings 141 (2017) 175–185
[44] D.H.W. Li, Daylight and energy implications for CIE standard skies, Energy
Convers. Manage. 48 (2007) 745–755, http://dx.doi.org/10.1016/j.enconman.
2006.09.009.
[45] M.B. Hirning, G.L. Isoardi, S. Coyne, I. Cowling, Discomfort glare in open plan
green buildings, Energy Build. 70 (2014) 427–440, http://dx.doi.org/10.1016/j.
enbuild.2013.11.053.
[46] S. Abbaszadeh, L. Zagreus, D. Lehrer, C. Huizenga, Occupant satisfaction with
indoor environmental quality in green buildings, in: Proc Heal. Build. 2006,
Lisbon, 2006, pp. 365–370.
[47] D.H.W. Li, J.C. Lam, Evaluation of lighting performance in office buildings with
daylighting controls, Energy Build. 33 (2001) 793–803, http://dx.doi.org/10.
1016/S0378-7788(01)00067-6.
[48] S. Onaygıl, Ö. Güler, Determination of the energy saving by daylight
responsive lighting control systems with an example from Istanbul, Build.
Environ. 38 (2003) 973–977, http://dx.doi.org/10.1016/S0360-
1323(03)00034-9.
[49] A.D. Galasiu, G.R. Newsham, C. Suvagau, D.M. Sander, Energy saving lighting
control systems for open-plan offices: a field study, J. Illum. Eng. Soc. 4 (2007)
7–29, http://dx.doi.org/10.1582/LEUKOS.2007.04.01.001.
[50] R. Kittler, M. Kocifaj, S. Darula, The 250th anniversary of daylight science:
looking back and looking forward, Light. Res. Technol. 42 (2010) 479–486,
http://dx.doi.org/10.1177/1477153510373569.
[51] J. Mardaljevic, L. Heschong, E. Lee, Daylight metrics and energy savings, Light.
Res. Technol. 41 (2009) 261–283, http://dx.doi.org/10.1177/
1477153509339703.
[52] K. Kensek, J. Suk, Daylight factor (overcast sky) versus daylight availability
(clear sky) in computer-based daylighting simulations, J. Creat. Sustain.
Archit. Built. 1 (2011) 3–14 http://www.sersc.org/journals/CSABE/vol1/2.pdf.
185
[53] A. Nabil, J. Mardaljevic, Useful daylight illuminance: a new paradigm for
assessing daylight in buildings, Light. Res. Technol. 37 (2005) 41–59, http://
dx.doi.org/10.1191/1365782805li128oa.
[54] P.R. Tregenza, I.M. Waters, Daylight coefficients, Light. Res. Technol. 15 (1983)
65–71, abstract http://lrt.sagepub.com/content/15/2/65.
[55] S. cheng Chien, K.J. Tseng, Assessment of climate-based daylight performance
in tropical office buildings: a case study, Int. J. Low-Carbon Technol. 9 (2014)
100–108, http://dx.doi.org/10.1093/ijlct/ctu014.
[56] M.B. Hiring, G.L. Isoardi, S. Coyne, V.R. Garcia-Hansen, I. Cowling, Post
occupancy evaluations relating to discomfort glare: a study of green buildings
in Brisbane, Build. Environ. 59 (2013) 349–357, http://dx.doi.org/10.1016/j.
buildenv.2012.08.032.
[57] J.A. Jakubiec, C.F. Reinhart, K. Van Den Wymelenberg, Towards an integrated
framework for predicting visual comfort conditions from luminance-based
metrics in perimeter daylit spaces, in: Proc. BS2015: 14th Conference of
International Building Performance Simulation Association, Hyderabad, India,
Dec. 7–9, 2015, pp. 1189–1196.
[58] J.A. Jakubiec, C.F. Reinhart, A concept for predicting occupants’ long-term
visual comfort within daylit spaces, Leukos (2015), http://dx.doi.org/10.1080/
15502724.2015.1090880.
[59] P.R. Boyce, C. Hunter, O. Howlett, The Benefits of Daylight Through Windows,
2003, pp. 12180–13352.
[60] L. Bellia, F. Bisegna, G. Spada, Lighting in indoor environments: visual and
non-visual effects of light sources with different spectral power distributions,
Build. Environ. 46 (2011) 1984–1992, http://dx.doi.org/10.1016/j.buildenv.
2011.04.007.