An analysis of occupants response to thermal discomfort in green and conventional buildings in New Zealand

Energy and Buildings 104 (2015) 191–198

Contents lists available at ScienceDirect

Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild

An analysis of occupants response to thermal discomfort in green and conventional buildings in New Zealand Nurul Sakina Mokhtar Azizi ∗ , Suzanne Wilkinson, Elizabeth Fassman Department of Civil & Environmental Engineering, Faculty of Engineering University of Auckland, 20 Symonds St, Auckland City 1010, New Zealand

a r t i c l e

i n f o

Article history: Received 27 June 2014 Received in revised form 22 June 2015 Accepted 4 July 2015 Available online 7 July 2015 Keywords: Green buildings Energy saving behaviour Coping mechanism Adaptive comfort Thermal discomfort

a b s t r a c t Studies have found high discomfort issues in green buildings where occupants find it too cold during the winter and too hot during the summer. Green buildings are highly climate responsive since they are usually dependent upon natural ventilation and natural daylight. In conventional buildings, occupants are not so dependent on the building design to moderate temperature and lighting. This paper investigates occupants responses to discomfort in conventional and green buildings to better understand how they behave, and whether they behave differently. This study examines what people do when they are too hot or too cold. Three coping mechanism were tested (i) environmental adjustment, (ii) personal adjustment and (iii) psychological adjustment. Results in this paper showed that in response to being cold, occupants in green buildings engaged more in personal adjustments, less environmental adjustment, and more in psychological adjustment compared to conventional buildings. While in response to being hot, these coping mechanisms were less apparent. The paper examines what adjustments people make when they are too hot or too cold, and compares these behaviours in different building types. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Thermal comfort is defined by the International Standard American Society of Heating, Refrigerating and Air Conditioning Engineers (ASHRAE) as “that state of mind which expresses satisfaction with the thermal environment” [1]. Green buildings, which are mostly mechanically and naturally ventilated buildings adopt the ASHRAE 55 2010 [1] and ISO 7730 [2] guidelines which suggests a thermal comfort temperature between 20 ◦ C and 24 ◦ C. Studies showed that the thermal comfort at these temperatures is generally accepted by the occupants [3–5]. Post occupancy studies on comfort have shown that in general green buildings are more comfortable compared to conventional building [6,7]. Although the temperature of 20 ◦ C to 24 ◦ C is an accepted comfort range for most occupants, Nicol and Humphreys [8], argued that the temperature range advised by ASHRAE 55 2010 [1] standard and ISO 7730 [2] guideline is too narrow. This is supported by other studies that found high discomfort issues in green buildings where occupants find it too cold during the winter and too hot during the summer [9–12]. Occupants will perform environmental adjustment buildings when they are experiencing discomfort in thermal, daylight and natural ventilation in the building [13–17].

∗ Corresponding author. Tel.: +64 226809427. E-mail address: [email protected] (N.S.M. Azizi). http://dx.doi.org/10.1016/j.enbuild.2015.07.012 0378-7788/© 2015 Elsevier B.V. All rights reserved.

The adaptive thermal comfort model (i.e. International standard ASHRAE RP884 [18] and European Standard EN 15251 [19] proposes a wider temperature range up to 30 ◦ C and claims that buildings should be designed in a way that provides wider opportunity for occupants to adopt behaviour adaptations [8,20,21]. The adaptive thermal comfort model is not widely adopted in a controlled thermal environment such as an air-conditioned building [4]. This is because there is a huge challenge in behavioural change as it requires a lifestyle change that is too onerous [4,22]. Scholars have listed examples of behaviour adaptations in response to discomfort to being cold and hot [8,14]. There are three types of behaviour adaptations which are (i) personal adjustment—i.e. adjusting activity, adjusting posture, (ii) technological or environmental adjustment—i.e. turning on fans or heaters and (iii) psychological adjustments—i.e. just put up with it, or try to ignore the problem. However, limited studies have been conducted on the level of practice of these behaviours. By investigating the level of practice of behaviour adaptations to thermal comfort it is possible to gain a better understand how buildings can encourage more behaviour adaptations. Environmental adjustments have energy implications for the building. Among the post-occupancy issues in green buildings are lack of knowledge and skills on how to operate the environmental control systems efficiently [23–25], and limitation of accessibility to the control systems which caused occupants to make their own personal modifications (i.e. use personal fan, heater, and etc)

192

N.S.M. Azizi et al. / Energy and Buildings 104 (2015) 191–198

Table 1 Personal adjustment. Personal adjustment

References

clothing adjustment when felt cold and hot alter timing of their work pattern to avoid uncomfortable working conditions adjusting posture consuming hot or cold food and drinks moving to a different location taking a walk inside or outside contact building manager share the problem with co-workers to see if they are also experiencing discomfort

[14,23–28] [23,25,27,28] [25] [14,23,25,26,28] [14,23–25,27–30] [14,26] [14] [14]

to achieve optimum comfort [16,23,26]. Studies have also found that occupants’ desire to achieve optimum comfort caused controls to be overridden, such as mechanical cooling and heating systems [16,17,24]. Against the need to change controls, Heerwagen and Diamond [14] suggest that an adverse impact of providing good automated building control systems is the “desk couch potato” where there may be a lack of muscle movement, and an increase of social isolation. Heerwagen and Diamond [14] examined the three types of behaviour adjustments (personal, environmental, psychological) in green buildings. The findings showed that the green buildings encouraged more personal adjustments than environmental adjustments. Personal adjustments were made more than environmental adjustments in spaces which occupants have limited access to the control systems such as the open plan space. While in private offices within the building, the occupants made more environmental adjustments than personal adjustments. Advocates for personal adjustments believe it not only helps reduce energy consumptions in buildings, but it is also believed to create healthier personal actions for the occupants since there is more muscle movement [4,14,27]. In order to further the debate about thermal comfort in buildings, the study in this paper examines what people do when they are too hot or too cold, and whether there are significant differences in the behaviour of occupants between green and conventional buildings. 2. Coping mechanisms in response to discomfort Reviews of the international literature showed that there are three basic types of coping mechanism in response to discomfort that occupants normally take in buildings [14,21]. These coping mechanisms are (i) personal adjustment—i.e. adjusting activity, adjusting posture, (ii) technological or environmental adjustment—i.e. turning on fans or heaters and (iii) psychological adjustments—i.e. just put up with it, or try to ignore the problem. Table 1 provides a detailed list of adjustments considered to be personal adjustments in response to thermal discomfort. As shown in Table 1, clothing adjustment is a common personal adjustment made in response to discomfort. This behaviour has been promoted in office buildings. For example, a campaign on no neck ties in Japan in 2005 [28] and employees were encouraged to adopt casual dress code in United Nation Headquarters, New York [29]. The rationalisation for this campaign was that flexibility in dress code in office buildings provides occupants more adaptive strategies to cope with thermal discomfort. O’Connor et al. [30] categorised these behaviour changes as “suffer discomfort”. Although discomfort is not relieved entirely by personal adjustment, these behaviours have important functions such as making people move around more and engage in social interactions [4,14]. The mental and social benefits generated from personal adjustments are worthwhile and create a healthier environment for

the occupants. There are limited understanding on whether the design of a green building encourages occupants to engage in personal adaptation [27,31,32]. For example, Healey and WebsterMannison [27] reported that occupants engaged in more personal adjustments (i.e. dress in layers, cons umed hot/cold beverages, disposition) due to the influences of the socio-cultural aspects within the building, but did not relate the adaptive behaviour responses to the physical environment in the building. Moezzi and Goins [33] reported that occupants in commercial buildings engaged in less personal adjustments (i.e. drink hot/cold beverages; dress in layers, walk around more) than environmental adjustments and speculated that it is due to the lack of physical environment such as a place to buy coffee, and a place to retreat. However, findings by Gauthier and Shipworth [34] showed a different result to Moezzi and Goins [33] where physical environment does not necessarily encourage occupants to engage in personal adjustments. Several earlier studies indicated that building design features such as spacious common room and access view to the natural environment reduces occupants stress level and increases work productivity [35–38]. Environmental adjustments are how occupants interact with the building control systems (i.e. windows, blinds, switches, and other controls). Occupants who engage in this thermal discomfort coping mechanism can impact energy usage if the building control systems are not operated efficiently. Inefficient operation of the building control systems are described in the following studies. For example, Gabe [24] and Sawyer et al. [17] discovered that occupants increased the load of the cooling and heating systems to accommodate comfort. Reiss [16] discovered that occupants routinely override switches for natural ventilation or mechanical cooling because they don’t know what conditions each option is intended for. Reiss [16] also discovered that occupants did not open the window when they were supposed to which caused the heating system to consume energy five times more than predicted. Heerwagen and Wise [39] showed that occupants kept doors open for fresh air causing mechanical systems to consume more energy. Bordass et al. [23] and Brown [40] reported that occupants used personal heaters or fans to relieve discomfort. These studies led to the assumption that when occupants are provided with high access to the environmental control systems, they will be more likely to make adjustments that will impact energy usage in buildings [14,26]. This prediction was further supported by findings from Ricciardi and Burrati [41] and Moezzi and Goins [33] where the occupants engaged in less environmental adjustments when they had limited access to the building environmental control systems. O’Brien and Gunay [42] raised concern that contextual factors such as occupants’ awareness and perception of working in a green building can influence their choice of adaptive behaviour. For instance, occupants may adopt poor energy saving behaviours in energy efficient design buildings due to the ‘rebound effect’ [43,44]. Heerwagen and Diamond [14] defined psychological coping mechanism as an attempt to adjust to a situation by managing emotions or thoughts about the situation. Occupants responded to either feeling hot/cold by just putting up with the discomfort, believing there was nothing they could or trying to ignore the discomfort. Heerwagen and Diamond [14] found that almost one fifth of the occupants who experienced thermal discomfort either feeling too hot or cold chose to not do anything. Occupants engaged more in this coping mechanism when environmental adjustments are limited, and when other coping mechanisms’ are not effective to relieve discomfort. Previous studies describe adjustments made by occupants to relieve discomfort. These studies did not quantify the frequency of the behaviours. Quantification of the frequency of behaviours can aid building designers to make better prediction of energy usage. Current energy modelling software assume occupants schedules

N.S.M. Azizi et al. / Energy and Buildings 104 (2015) 191–198

are similar to building operation schedules (8am to 5pm) with no absence from their offices during workdays, lunch, meetings, etc. [45,46]. Frequent act of occupants using additional heaters/fans, and adjust temperature are not accounted for in the energy simulation tools [47–49]. Often designers make inaccurate assumptions that occupants would open windows to optimise usage of natural ventilation [48]. It is important to understand better occupants’ interaction with building control systems when faced with discomfort. When building control systems are operated efficiently, they may not relieve occupants discomfort entirely. Occupants have a high tolerance to discomfort. Moujalled et al. [50] discovered that occupants preferred more naturally ventilated buildings as compared to air-conditioned buildings even if these buildings were colder. To design buildings that encourage occupants to practice energy saving behaviour, designers must understand occupants’ level of interaction with the building control system. A study by Santin [51] found that energy-conscious households conserve more energy with systems that require active involvement, while less energyconscious households conserve more energy with systems that do not require active involvement. This paper provides a better understanding of occupants’ behaviour through comparison analysis between green and conventional buildings. The findings in this paper will not only help designers to design better green buildings that encourage green practice, but also assist designers to make better predictions of energy use. This paper extends the research by Heerwagen and Diamond [14] and Azar and Menassa [45] by comparing behaviours in conventional buildings to see if green buildings have any influence on how occupants behave in response to thermal discomfort.

193

system which is energy efficient compared to a typical airconditioning system. 3.2. Owen Glenn Building—Green building (OGGB) The Owen G. Glenn Building, a 7 storey building was completed in 2007. The area of the building is 74,000 m2 with an estimated population of 400 occupants. The main energy efficient features incorporated in the building were highly glazed windows to optimise natural daylight, with layered facades to provide solar shading. Occupancy sensors and automatic building control systems are connected to an energy management system. 3.3. Old Choral Hall—Conventional building (OCH) Old Choral Hall building is a 4 storey building completed in 1872. Total estimated population in the building is 100 occupants. This building was identified as an historic building in New Zealand. No energy efficient design was incorporated at the time it was built. As the building is protected under the Historic Places Act 1993, there is limited ability to incorporate significant changes to the building [55,56]. The energy in OCH building is monitored and controlled using an Energy Management System (EMS) in which it monitors sub-meter sub-meters of energy usage in lighting, cooling, and plugload. OCH is the only buildings that do not have an assigned building manager on site. This is because OCH is a relatively small building with less than 100 occupants as compared to the rest of the buildings. Hence, to have a building manager in OCH was not necessary. 3.4. Faculty of engineering—Conventional building (FoE)

3. Thermal discomfort adjustments in New Zealand buildings Four building were selected and used to compare thermal discomfort practices. Two buildings were green buildings and two were conventional buildings. New Zealand is known for its green image and awareness of green issues tends to be high amongst the general population [52]. The New Zealand Ministry for the Environment acknowledged that energy efficiency is an important aspect that needs to be addressed [53,54]. Two of each category of buildings was chosen to understand the influence of building design with occupant’s behaviour in saving energy. Thomas Building (TB) and Owen G Glenn Building (OGGB) were identified as green buildings with the design intent to be energy efficient. Faculty of Engineering (FoE) and Old Choral Hall (OCH) buildings was selected as conventional buildings without specific energy efficient design.

3.1. Thomas Building—Green building (TB) The extension Thomas Building, a 4 storey building was built in 2011 with the design intention for the building to be green building certified need by GreenStar New Zealand. The area of the building is 4,958 m2 with an estimated population of 160 occupants. The design intended a rating NZGreenStar between 4 star and 5 star. Therefore, this building has a higher level of greenness as compared to other green design intent buildings. Energy efficient features in the building incorporated double glazed tinted low E windows with double skin fac¸ade. The outer glazing with fritted dot pattern provides 30% shading. Natural ventilation is provided through inoperable window louvers. Most areas in the building have occupancy sensors. The building also adopts the variable air volume (VAV)

Faculty of engineering building is a 12 storey building with an estimation of 300 occupants. The building was reported to have no energy efficient features in the building. The building was built in year 1969. The Faculty of Engineering building is a 12 storey building with an estimated 300 occupants. The building was reported to have no energy efficient features in the building. The building was built in year 1969. In year 2003, the building was refurbished where an atrium was built with large space common room area including a cafeteria on the lower floor. The building provides 250-seat lecture theatre. At the end of the building, a long glass-enclosed colonnade was designed to create a transparent effect as well as gain natural daylight in the building. 3.5. Energy consumption of case study buildings For New Zealand, the energy index reference for university buildings in the industry to design green buildings is based on rough estimations of energy consumption in a building. The current energy index that is used in practice is 174 kW h/m2 /year. However, this energy index is not considered appropriate to be used for building performance analysis since the case study buildings in this study are unique as the function of the building is a combination of office and research laboratory space. The energy manager of the UoA argued that by just simplifying the total energy consumption to the gross floor area of the total numbers of buildings in a university without normalising other contextual variables such as the function of the buildings, local condition, and weather is not an accurate energy benchmark for energy efficiency analysis. The building managers reported that it would be beneficial if there were available energy benchmarks that can be used as targets. The current practice compares yearly historical data without any external energy targets. On average the green buildings (TB and OGGB) have increased energy consumption while the conventional buildings (FoE and OCH) both decreased their energy consumption.

194

N.S.M. Azizi et al. / Energy and Buildings 104 (2015) 191–198

Table 2 Response rate. Type

Total population

Sample size

Total respondents received

Response rate (%)

Name of building

400

196

113

58

700

248

157

63

FoE OCH TB OGGB

1100

285

270

95

Conventional building Green buildings

Total

All four buildings are monitored and controlled using an Energy Management System (EMS) in which it monitors sub-meters of energy usage in lighting, cooling, and plugload. A building maintenance office equipped with facilities for monitoring energy consumption in the buildings is provided in all four buildings. The energy management system showing the track of energy usage is accessible online in all four case studies. The EMS tracks energy consumption of the building at an interval of 30 min daily. Thus, allowing building managers to capture energy waste by observing the current energy trend to the average baseline of what the building should be consuming. It also allows for the lighting, building services plant and temperature to be remotely controlled. In all four case studies, a detail energy scheduling is practiced. Airconditioning was scheduled to switch off an hour before office hours ends. Air-conditioning temperatures are set between 20 ◦ C and 24 ◦ C and is regularly reviewed according to occupancy density in the building. Lighting systems are also scheduled to switch off after hours. The building owners are aware of the importance of having an active facility-management team. In all case studies, there is an assigned building manager who reports back to an energy manager, a central coordinator on energy-related matters for the buildings Frequent meetings are held to understand and discuss how energy can further be reduced.

3.6. Research method Invitations to participate in this study were sent through an email and a follow up call to the building managers was made. The building managers in each of four case study buildings then distributed an online survey uploaded onto the building website to the occupants in the buildings. The researcher conducted a follow up email requesting the building manager to circulate the website link to the occupants in the building after two weeks. Hardcopies were also provided to the building manager for occupants who wished to fill in the questionnaire manually. In order to increase the response rate, the researcher was given access to the case study buildings to invite participants in the research face to face. Hardcopies as well as the website link were given to interested participants. The aim of the questionnaire was to understand how occupants use the building to adjust their thermal comfort. 10 questions about coping mechanism for each environmental condition were asked. Occupants were asked to rate their actions to achieve comfort using a Lickert scale to 5-I always do to 1-I never do. Option N/A was also provided for actions that are not relevant to them. Occupants were also asked to tick whether they had access to the building control systems such as being able to adjust temperature or open/close windows and doors. Analysis was undertaken using SPSS Statistic 22. The Mann–Whitney U test was used to identify which of the coping mechanism is significantly different between the two building types. The coping mechanisms identified as significantly different via Mann–Whitney U test were further analysed using frequency description and crosstab analysis to ascertain which building types practices the most coping mechanisms.

Total

Population

Sample size

Respondents received

Response rate (%)

300 100 300 400

169 80 169 196

80 33 68 89

47 41 40 45

1100

285

270

95

Occupants were also asked whether they agreed to this statement “Working in a green building means I have to sacrifice my comfort level and change my lifestyle to save energy” using a Lickert scale to 5-Strongly Agree to 1-Strongly Disagree. Occupants were also asked which environmental condition they prefer—feeling too hot or feeling too cold. Frequency analysis was used to identify the highest percentage of response for each factor.

4. Results The building managers estimated that there were 1100 people in the four buildings. A total of 270 responses were received. It is difficult to definitely know the response rate, as some of the occupants may never have received the information about the survey, or known of the study. However, if the actual numbers are taken as correct, then the calculated sample size of the population is 285 with confidence level at 95% and confidence interval of 5. This would mean a response rate of 95%. The sample size calculation is based on an established study by Krejcie and Morgan [57]. Table 2 shows the breakdown response rate in each building. Statistical analysis using Mann–Whitney U test showed that there are five actions significantly different between green (TB and OGGB) and conventional (FoE and OCH) buildings for occupants responses when they feel cold. These actions are two actions on environmental adjustment, two actions on behavioural adjustment and one on psychological adjustment. Under environmental adjustment, Fig. 1 shows that occupants in green (TB and OGGB) buildings are significantly less likely (p = 0.026) to adjust the temperature on the heating system in their working space. In addition, occupants in green (TB and OGGB) buildings are significantly less likely (p = 0.000) to use personal heater. As for personal adjustment, Fig. 1 shows that occupants in green (TB and OGGB) buildings complained significantly less (p = 0.000) to the building manager when they feel too cold. Occupants in green (TB and OGGB) buildings are significantly more likely (p = 0.000) to walk around to heat up their body. As for psychological adjustment, there are significantly more (p = 0.000) occupants in green (TB and OGGB) buildings who chose not to do anything about their discomfort Occupants in green (TB and OGGB) buildings are less likely than those in conventional buildings to change the environmental condition in the building (i.e. less likely to adjust the heating system, less likely to use personal heater) and chose more personal adjustment (i.e complained less to the building manager, walked more around in the building to heat up their body). Furthermore, occupants in green (TB and OGGB) buildings had significantly more occupants who do not do anything when they felt too cold. As for the thermal condition of the building being too hot, Fig. 2 shows that there are four actions which are statistically significantly different between the two building types. These actions are two actions on environmental adjustment, one action on behavioural adjustment, and one action on psychological adjustment. Under environmental adjustment, Fig. 2 shows that

N.S.M. Azizi et al. / Energy and Buildings 104 (2015) 191–198

195

Fig. 1. Occupants response when feeling cold in their working space. Note * energy saving behaviours that are significant different through Mann–Whitney U test (p values <0.05).

occupants in green (TB and OGGB) buildings are significantly less likely (p = 0.004) to adjust the temperature on the cooling system. In addition, they are also less likely (p = 0.031) to use a personal fan. Under personal adjustment, Fig. 2 shows that occupants in green (TB and OGGB) buildings complain significantly less (p = 0.002) to the building manager when feeling too hot.

As for psychological adjustment, there is significantly less (p = 0.018) occupants in green (TB and OGGB) buildings who do not do anything when they felt hot. Occupants in green (TB and OGGB) buildings appear to cope more with discomfort when they are hot than occupants in conventional (FoE and OCH) buildings. Hence, in overall occupants in green (TB and OGGB) buildings chose less

Fig. 2. Occupants response when feeling hot in their working space. Note * energy saving behaviour that are significant different through Mann–Whitney U test (p values <0.05).

196

N.S.M. Azizi et al. / Energy and Buildings 104 (2015) 191–198

90% 77% 77%

80% 70%

61% 56%

60% 50%

Convenonal

40% Green

30% 15%

20% 10%

7%

0% Adjust temperature on the cooling/heang system

Open/close the window

Open/close the door

Fig. 3. Accessibility to the building control system.

environmental adjustment (i.e. temperature on the cooling systems are less likely adjusted, less usage of personal fans) and less personal adjustment (i.e. less complain to building manager) when they feel too hot. As for psychological adjustment, there are more occupants in green (TB and OGGB) buildings than conventional (FoE and OCH) buildings who do not do anything in response to discomfort. Fig. 3 shows occupants accessibility to the building control systems. The results show that the occupants in green (TB and OGGB) buildings have more access to adjust temperature compared to occupants in conventional (FoE and OCH) buildings. As for access to the windows, occupants in conventional (FoE and OCH) building have more access than occupants in green (TB and OGGB) buildings. Occupants in green (TB and OGGB) and conventional (FoE and OCH) buildings have the same level of accessibility to the doors. Occupants in green (TB and OGGB) buildings expressed more willingness to suffer discomfort (such as being too hot or too cold) compared to conventional (FoE and OCH) building occupans. This can be seen from results in Fig. 4 where almost 40% of occupants in green (TB and OGGB) buildings claimed to be willing to suffer discomfort when feeling hot and cold, while almost 30% in conventional (FoE and OCH) buildings gave this response. 5. Why are occupants in green buildings more likely to accept discomfort? The overall results in Figs. 1 and 2 show that when feeling hot or cold, occupants in green (TB and OGGB) buildings are less likely to change their environment (i.e. less likely to adjust temperature and use personal heaters/fans) as compared to conventional (FoE and OCH) building occupants. In addition, Fig. 3 shows results that 15% of occupants in green (TB and OGGB) buildings have access to adjust the temperature against only 7% of occupants in conventional (FoE and OCH) buildings. So, even though the occupants in green buildings have higher access to the control system they

45% 40%

40% 37%

36%

35%

35% 28%

30%

23%

25%

Convenonal

20%

Green

15% 10% 5% 0% Feeling too hot in your working space

Feeling too cold in your working space

None

Fig. 4. Which condition are you willing to sacrifice?.

are still less likely to adjust the temperature. It appears that occupants in green buildings are behaving differently in response to their building design. One theory might be that as green (TB and OGGB) building occupants are more aware of the impact changing temperature and personal heaters have on the energy use and therefore do not change their building controls. The green buildings (TB and OGGB) have active on-site building managers where they frequently remind the building occupants to save energy which cause the occupants to be more aware on energy efficiency. As for the conventional building (FoE and OCH), only the FoE building had an on-site building manager but sending reminders on energy savings were not as frequent as the building managers in the green (TB and OGGB) buildings. Based on the discussion with the building managers, the reason for this is because the building managers in the green (TB and OGGB) buildings have the perception that since the green (TB and OGGB) buildings are designed to be energy efficient, the operation of the building should also support energy efficiency. This shows that despite the universities commitment to reduce energy usage, the perception of working in green buildings had encouraged the building managers more to persistently reduce energy use. The results in this study supports Moezzi and Goins [33] where building managers in green buildings had also discouraged the occupants in the buildings to use personal fans or heaters. Findings from previous studies have shown that occupants change energy usage in green building to achieve comfort [14,17,23,40]. The results in this paper shows that although occupants take actions that may impact energy usage of green (TB and OGGB) buildings, the level of practice is significantly less when compared to conventional (FoE and OCH) buildings. The perception of working in a green (TB and OGGB) buildings could influence occupants’ behaviour towards being green. The findings suggest that there is no evidence to support the hypothesis by earlier studies that higher accessibility to the building control systems creates the “desk couch potato” social phenomena [14,58] nor does it support the speculation where occupants adopt poor energy saving behaviours in energy efficient design buildings [43,44]. The only action under environmental adjustment that is not significantly different between green (TB and OGGB) and conventional (FoE and OCH) buildings is the act of opening/closing window/doors. Even though occupants in conventional (FoE and OCH) buildings have higher access to windows, they still open/close the windows at the same level of frequency as those occupants in green buildings (see Fig. 3). Results from Figs. 1 and 2 show that occupants in both building types are more likely to open windows/doors when they feel hot (at least 40% selected either always/often/sometimes), and close windows/doors when they feel cold (at least 40% selected either always/often/sometimes). The result in this paper extends the study by Heerwagen and Diamond [14] by showing a clearer demonstration on the variation of the behaviour in different building types. Heerwagen and Diamond [14] reported that 17% occupants are more likely to open windows/doors when they are hot, 5% close window when they are cold. While results in this paper showed a higher percentage of approximately 40% of the occupants in both building types are more likely to open the windows/doors when they are hot, and close the windows/doors when they are cold (see Figs. 1 and 2). The results also extends the studies by Azar and Menassa [45,46,59] which was limited to conventional office buildings for adjusting blinds, and turning off lights and office equipment. Results from Figs. 1 and 2 in this paper showed that adjusting temperature and additional usage of fans/heaters are made by 10% of occupants in both hot and cold conditions. Windows are opened by almost 40% of occupants during hot temperatures and closed by almost 40% during cold temperatures. This information on the frequency of behaviour is useful to assist better prediction of energy

N.S.M. Azizi et al. / Energy and Buildings 104 (2015) 191–198

usage for energy modellers. For example, when modelling energy, often the designers assume that occupants would open windows to optimise usage of natural ventilation [48], however the results in this paper showed only 40% of occupants would open the windows. Also, the current practice of predicting energy usage does not account for behaviours that use additional fans/heaters. The results in this paper showed almost 10% of occupants would use additional fans/heaters in response to discomfort. Therefore an adjustment of a more realistic percentage can be made for the designers when predicting energy usage. Personal adjustments can provide social and mental benefits to occupants such as talk to co-workers interact with neighbours, contact building manager, taking a walk inside or outside, getting hot/cold drink. This creates healthier personal actions for the occupants. The results in this paper showed that occupants in green (TB and OGGB) buildings practice more personal adjustments when they feel cold compared to when they feel hot. These results are different to Heerwagen and Diamond [14] where there was no difference in the response rate to discomfort when feeling hot and cold. Results in this paper showed that occupants in green (TB and OGGB) buildings display healthy adjustments such as walking more around in the building to heat up their body; social and mental benefits such as contact building manager by informing the building manager on the discomfort. Occupants in the green (TB and OGGB) buildings studied were less likely to adjust the temperature system, and chose more personal adjustments possibly showing that they demonstrate more green behaviour and supporting the notion green buildings do influence how occupants behave. The green (TB and OGGB) buildings are highly glazed buildings which provide wide access to views of the natural environment outside the buildings and have spacious common spaces for occupants to retreat. The different design attributes in the green buildings from the conventional buildings are speculated to encourage the occupants to exercise healthy adjustments in response to discomfort. The findings in this study also support the speculation by Moezzi and Goins [33] where physical environment can encourage occupants to make healthy adjustments such as by walking around more. However, findings in this study do not support Gauthier and Shipworth [34] indicating that occupants’ adaptive behaviour varies according to the different demand level of comfort. The findings extends earlier studies by Joye [35], Haynes [36], Miller et al. [37] and McCunn and Gifford [38], by showing that not only do these design features (i.e. spacious common room and access to view the natural environment) can make occupants feel less stressed, but it can potentially encourage occupants to engage in more personal adjustments. Similarly, the results from Figs. 1 and 2 showed that occupants in green (TB and OGGB) buildings cope more with discomfort when they are hot and cold compared to occupants in conventional (FoE and OCH) buildings. Either hot or cold, 19% of the occupants in green (TB and OGGB) buildings chose not to do anything, compared to 4% in conventional (FoE and OCH) buildings. The results in this paper and results in Heerwagen & Diamond [14] showed that this coping mechanism was adopted higher in green buildings. Heerwagen and Diamond [14] interpreted that psychological adjustment was chosen more because occupants had limited access to adjust the control systems. However, results in this paper showed that psychological adjustments are still chosen, even though green occupants had higher accessibility to adjust the control systems (adjust temperature, open/close window) as compared to occupants in conventional buildings (see Fig. 3). 6. Conclusion In conclusion, when faced with being too hot or too cold occupants in buildings make adjustments to make themselves more

197

comfortable. The comparative study found that occupants in the green buildings engaged in less environmental adjustments, and adopted more personal and psychological coping mechanisms than those occupants in the conventional building. These results demonstrate that occupants in the green buildings prefer to accept the conditions the buildings provide, even though they had more access to building controls. It appears that green buildings are not only designed for energy efficiency but they encourage behaviour which saves on the energy consumption. The study provides better understanding on the occupants’ response to thermal discomfort by providing an estimation of how frequent these thermal coping mechanisms are practiced. This information is useful for building designers to make better predictions on the energy use in buildings and design buildings that encourage occupants to save energy as well as provide more opportunity for the occupants to make healthier adjustments. The results in this study add to the debate on linking green building use to socially responsible energy behaviour. Acknowledgments We thank the University of Auckland, Ministry of Education Malaysia, and University Science Malaysia for funding this research. We also express our gratitude to the building managers in the case study buildings for their cooperation, and assistance throughout this study. Lastly, we highly appreciate all the responses received from the building occupants in the case study buildings. References [1] ASHRAE, Thermal environmental conditions for human occupancy, in: ASHRAE Standard 55, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Atlanta, 2010. [2] ISO, Ergonomics of the thermal environment: Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria, in: ISO 7730, ISO, 2005. [3] F.W. Grimme, M. Laar, C. Moore, Man & climate—are we loosing our climate adaptation? in: Presented at the RIO3—World Clim. Energy Event, December 1–5, 2003, Rio de Janeiro, Brazil, 2003. [4] T. Law, Sch. Archit. Des., Springers, Switzerland, 2013. [5] F. Rohles, Temperature & temperament: a psychologist looks at comfort ASHRAE J. 49 (2007) 14–22. [6] A. Leaman, B. Bordass, Are users more tolerant of “green” buildings? Build. Res. Inf. 35 (2007) 662–673, http://dx.doi.org/10.1080/09613210701529518 [7] M.R. Esa, M.A. Marhani, R. Yaman, A.A. Hassan, N.H.N. Rashid, H. Adnan, Obtacles in implementing Green Building Projects in Malaysia, Aust. J. Basic Appl. Sci. 5 (2011) 1806–1812. [8] F. Nicol, M. Humphreys, S. Roaf, Adaptive Thermal Comfort: Principles and Practice, Routledge, London; New York, 2012, c2012. [9] S. Abbaszadeh, L. Zagreus, D. Lehrer, C. Huizenga, Occupant satisfaction with indoor environmental quality in green buildings, Proc. Heal. Build. 3 (2006) 365–370. [10] G. Baird, What the users think of sustainable buildings—a global overview, in: Sustain. Build. Conf., 26–27 May, 2010, Wellington, New Zealand, 2010. [11] A. Leaman, L. Thomas, M. Vanderberg, “Green” buildings: what Austrailian building users are saying, EcoLibrium (2007) 22–30. [12] W.L. Paul, P.A. Taylor, A comparison of occupant comfort and satisfaction between a green building and a conventional building, Build. Environ. 43 (2008) 1858–1870. [13] A. Leaman, F. Stevenson, B. Bordass, Building evaluation: Practice and principles, Build. Res. Inf. 38 (2010) 564–577. [14] J. Heerwagen, R. Diamond, Adaptations and coping: occupant response to discomfort in energy efficient buildings, in: Proc. ACEEE 1992 Summer Study Energy Effic. Build., Berkeley, CA, 1992, pp. 10.83–10.90. [15] J. Heerwagen, Office design meets (or bot) the energy challenges, in: Presented at the Behav. Energy Clim. Chang. Conf., November 14–17, 2010, Sacramento, CA, 2010. [16] R. Reiss, Improving the Energy Performance of Green Buildings. An Exclusive Report for E Source Members, E Sources Customer Direct, United States, 2005. [17] L. Sawyer, P. De Wilde, S. Turpin-Brooks, Energy performance and occupancy satisfaction: a comparison of two closely related buildings, Facilities 26 (2008) 542–551, http://dx.doi.org/10.1108/02632770810914299 [18] R. de Dear, G.S. Brager, R. de Dear, G.S. Brager, R. de Dear, G.S. Brager, Developing an Adaptive Model of Thermal Comfort and Preference, ASHRAE Trans. Inc., Macq, California, United States, 1998. [19] CEN, CEN Standard EN1521: Indoor Environmental Input Parameters for Design and Assessment of Energy Performance of Buildings Addressing Indoor Air Quality, Thermal Environment, Lighting and Acoustics, CEN, 2007.

198

N.S.M. Azizi et al. / Energy and Buildings 104 (2015) 191–198

[20] J.F. Nicol, M. Wilson, A critique of European Standard EN 15251: strengths, weaknesses and lessons for future standards, Build. Res. Inf. 39 (2011) 183–193, http://dx.doi.org/10.1080/09613218.2011.556824 [21] G.S. Brager, R.J. de Dear, Thermal adaptation in the built environment: a literature review, Energy Build. 27 (1998) 83–96, http://dx.doi.org/10.1016/ s0378-7788(97)00053-4 [22] E. Halawa, J. Van Hoof, The adaptive approach to thermal comfort: a critical overview, Energy Build. 51 (2012) 101–110, http://dx.doi.org/10.1016/j. enbuild.2012.04.011 [23] B. Bordass, R. Cohen, M. Standeven, A. Leaman, Assessing building performance in use 3: energy performance of the Probe buildings, Build. Res. Inf. 29 (2001) 114–128, http://dx.doi.org/10.1080/09613210010008036 [24] J. Gabe, Design versus performance: lessons from monitoring an energy-efficient commercial building in operation, in: Paper Presented at 3rd Int. Conf. Sustain. Eng. Sci., 9–12 December, 2008, Auckland, New Zealand, 2008. [25] I. Mosly, G. Zhang, Study on risk management for the implementation of energy efficient and renewable technologies in green office buildings, in: Paper Presented at 4th Int. Conf. Sustain. Eng. Sci. Transition to Sustain, November 30–December 3, 2010, Auckland, NZ, 2010. [26] R.J. Cole, Z. Brown, Reconciling human and automated intelligence in the provision of occupant comfort, Intell. Build. Int. 1 (2009) 39–55, http://dx.doi. org/10.3763/inbi.2009.0007 [27] K. Healey, M. Webster-Mannison, Exploring the influence of qualitative factors on the thermal comfort of office occupants, Archit. Sci. Rev. 55 (2012) 175–196. [28] Stephen Healy, Cool Biz: an STS analysis, in: Paper Presented at the Annual Meeting of the 4S Annual Meeting—Abstract and Session Submissions, Komaba I Campus, University of Tokyo, Aug 25, 2010, Tokyo, Japan, 2010. [29] C.K. Tan, Cool United Nations, United Nations University, China, 2008, http://ourworld.unu.edu/en/cool-united-nations/ (accessed September 13, 2011). [30] R.E. O’Connor, R.J. Bord, B. Yarnal, N. Wiefek, Who wants to reduce greenhouse gas emissions, Soc. Sci. Q. 83 (2002) 1–17, http://dx.doi.org/10. 1111/1540-6237.00067 [31] A. Keyvanfer, A. Shafaghat, M.Z.A. Majid, H. Lamit, K.N. Ali, Correlation study on user satisfaction from adaptive behavior and energy consumption in office buildings, J. Teknol. Sci. Eng. 70 (2014) 89–97. [32] Y. Zhang, P. Barret, Factors influencing occupant’s blind-control behaviour in a naturally ventilated office building, J. Build. Enviorn. 54 (2012) 137–147. [33] M. Moezzi, J. Goins, Text mining for occupant perspectives on the physical workplace, Build. Res. Inf. 39 (2011) 169–182, http://dx.doi.org/10.1080/ 09613218.2011.556008 [34] S. Gauthier, D. Shipworth, Behavioural responses to cold thermal discomfort, Build. Res. Inf. 43 (2015) 355–370, http://dx.doi.org/10.1080/09613218.2015. 1003277 [35] Y. Joye, Architectural lessons from environmental psychology: the case of biophilic architecture, Rev. Gen. Psychol. 11 (2007) 305–328. [36] B. Haynes, An evaluation of office productivity measurement, J. Corp. Real Estate 9 (2007) 144–155. [37] N. Miller, D. Pogue, Q. Gough, S. Davis, Green buildings and productivity, J. Sustainable Real Estate 1 (2009) 65–89, http://ares.metapress.com/content/6402637N11778213. [38] L.J. McCunn, R. Gifford, Do green offices affect employee engagement and environmental attitudes? Archit. Sci. Rev. 55 (2012) 128–134, http://dx.doi. org/10.1080/00038628.2012.667939

[39] J.H. Heerwagen, J.A. Wise, Green building benefits: differences in perceptions and experiences across manufacturing shifts, Heating Pip. Air Cond. 70 (2) (1998) 57–63, PBD Feb 1998. (1998) Medium: X; Size. [40] Z.B. Brown, Occupant Comfort and Engagement in Green Buildings: Examining the Effects of Knowledge, Feedback, and Workplace Culture, University of British Columbia, Vancouver, British Columbia, Canada, 2009 (Ph.D Thesis). [41] P. Ricciardi, C. Buratti, C. Burrati, Thermal comfort in open plan offices in nothern Italy: an adaptive approach, Build. Environ. 56 (2012) 314–320, http://dx.doi.org/10.1016/j.buildenv.2012.03.019 [42] W. O’Brien, H.B. Gunay, The contextual factors contributing to occupants’ adaptive comfort behaviors in offices—a review and proposed modeling framework, Build. Environ. 77 (2014) 77–87, http://dx.doi.org/10.1016/j. buildenv.2014.03.024 [43] S. Sorrell, J. Dimitropoulos, M. Sommerville, Empirical estimates of the direct rebound effect: a review, Energy Policy 37 (2009) 1356–1371, http://dx.doi. org/10.1016/j.enpol.2008.11.026 [44] J.S. Bourrelle, Zero energy buildings and the rebound effect: a solution to the paradox of energy efficiency? Energy Build. 84 (2014) 633–640. [45] E. Azar, C. Menassa, Agent-based modeling of occupants and their impact on energy use in commercial buildings, J. Comput. Civ. Eng. 26 (2012) 506–518, http://dx.doi.org/10.1061/(ASCE)CP.1943-5487.0000158 [46] C. Menassa, V. Kamat, S. Lee, E. Azar, C. Feng, K. Anderson, Conceptual framework to optimize building energy consumption by coupling distributed energy simulation and occupancy models, J. Comput. Civ. Eng. 28 (1) (2014) 50–62. [47] P. Hoes, J.L.M. Hensen, M.G.L.C. Loomans, B. de Vries, D. Bourgeois, User behavior in whole building simulation, Energy Build. 41 (2009) 295–302. [48] B. Li, D. Lim, Occupant Behavior and Building Performance, Des. Manag. Sustain. Built Environ., Springer, London, United Kingdom, 2013, pp. 279–304. [49] C. Turner, M. Frankel, Energy Performance of LEED for New Construction Buildings, New Buildings Institute, Vancouver, United states, 2008. [50] B. Moujalled, R. Cantin, G. Guarracino, Comparison of thermal comfort algorithms in naturally ventilated office buildings, Energy Build. 40 (2008) 2215–2223, http://dx.doi.org/10.1016/j.enbuild.2008.06.014 [51] O.G. Santin, O. Guerra Santin, Behavioural patterns and user profiles related to energy consumption for heating, Energy Build. 43 (2011) 2662–2672, http:// dx.doi.org/10.1016/j.enbuild.2011.06.024 [52] G. Brown, L. Stone, Cleaner production in New Zealand: taking stock, J. Clean. Prod. 15 (2007) 716–728. [53] D. Fullbrook, Q. Jackson, G. Finlay, Value Case for Sustainable Building in New Zealand, Ministry for Environment, Auckland, NZ, 2006. [54] EECA, Managing Energy—Getting Started (Energy Management Guideline), Energy Efficiency Conservation Authority, Auckland, NZ, 2012. [55] R. McClean, Planning for heritage sustainability in New Zealand—a safe heritage credit scheme, New Zealand historic places trust, in: Presented at the Refit and Retrofit Conf, September 26–27, 2012, Auckland, NZ, 2012. [56] MfE, Heritage Management Guidelines for Resource Management Practicioners, Historic Places Trust, Ministry for Environment, New Zealand, 2004. [57] R. Krejcie, D. Morgan, Determining sample size for research activities, Educ. Psychol. Meas. 30 (1970) 607–610. [58] Z. Brown, R. Cole, Influence of occupant knowledge on comfort expectations and behavior, Build. Res. Inf. 37 (2009) 227–245. [59] Ministry for the Environment, Heritage Management Guidelines for Resource Management Practitioners, New Zealand Historic Places Trust, Wellington, 2004.