Lee Siew Eang

Department of Building

School of Design and Environment

National University of Singapore

Energy in the form of electricity is used in building to operate equipment for the safety, efficiency and comfort of its occupants and users. Such equipment includes emergency systems, airconditioning, lighting, transportation, office systems and other appliances.

In the absence of natural resources fundamental to the generation of electricity, energy will remain a critical factor for the Singapore economy for the immediate and long-term future. This would have direct impacts on building designers, managers and owners. It is important to monitor and manage the cost pressure in the operation and maintenance of buildings with respect to energy. It is important to ensure global competitiveness in the area of environmental performance and efficiency.

Hence, the trend towards energy efficient building is not a passing trend like the energy crisis of the 1970s. Energy efficiency of building has emerged as a permanent performance factor to be considered in the cost and environmental equations. Knowledge base and understanding in energy efficiency of buildings will formalize and become a core competence of building designers, managers and owners.

An office building is defined as a building where spaces in a building or part of the building are used or intended to be used for rendering services such as agency, commission, banking, administrative, legal, architectural, engineering and other professional services. According to the Urban Redevelopment Authority’s statistics of 1998, there are a total of 511 office and office cum retail buildings in Singapore. Among them 21% are within public sector ownership.

Figure 1 below shows the electricity consumption distribution in Singapore for the year 1999. The office buildings’ consumption has been included in the non-manufacturing sector. It is additional shown to illustrate its contribution to the whole sector. It can be seen that electricity consumption in buildings, excluding the industry sector, is 57%. The domestic sector alone, that is the residential portion of the electricity consumption, is 20%. The consumption of office buildings, which is a part of the non-manufacturing sector, is responsible for 12% of the overall non-manufacturing sector’s consumption. In view of the large residential build-up space as compared to the total office space, office may indeed be considered as major energy user among the various building types.

The data of Figure 1 also clearly illustrate the importance of the building sector in contributing to the overall energy efficiency of Singapore.

Figure 1: Electricity consumption among different economic sectors in Singapore.

(Source: Singapore Power’s Annual Report 1999)

This paper presents findings of a recent study conducted. It examines the overall trends of energy consumption among office buildings in Singapore. An attempt is made to benchmark the energy performance of office buildings. A total of 104 office and office cum retail buildings have been studied.

Surveys were conducted on the energy consumption of buildings. The survey aimed to obtain the total as well as a break down of energy consumption, between the landlord and the tenants, of a building. Data were collected through a questionnaire survey including information about the function, and operational and occupancy characteristics of the building. 10% of the data collected, including both the landlords’ and tenants’ electrical consumptions were verified on site.

The data set collected includes the age of the building, age since the latest refurbishment, occupancy rate (OR), the total gross floor area (GFA), scheduled operating hours (OH), indoor environmental setting, systems’ performance data, and two years’ monthly electricity consumptions.

Statistical analyses on the characteristics and distribution of data, as well as relationship such as correlations between different parameters were investigated. The observed data values for the total energy consumption, the landlord and the tenants have all been tested and were found to satisfy the normal distribution function.

A total of 163 letters of invitation and questionnaires were sent out inviting participation for the study. An 86% response rate was achieved with 104 building managements responded to the survey. The building samples collected therefore represent a 3.7% sampling error using the stratified sampling random estimation technique. Among the 104 buildings that responded, there are 44 public sector buildings and 60 private sector buildings. 82 are office buildings in use and 22 are office cum retail buildings. In the latter group the retails function is a negligible portion of the total office space.

The buildings studied have average design efficiency, gross lettable area (GLA) to gross floor area ratio, of 74% with a 95% confidence interval of +/- 6.3%. They also have an average ratio of airconditioned area to gross floor area of 80% with a 95% confidence interval of +/- 2.5%. These show that the cohort of buildings studied have rather similar design profiles and standards. The low spread of building design efficiency, and the large sample population with a sampling error of about 3.7% means that the data obtained provide a reliable profile of office buildings’ energy performance in Singapore.

The respondents may be classified into three categories as follows:

1. Owner Occupiers

There are 30 such buildings. Typically, these buildings are owned, occupied and operated by one user group. There is therefore no tenant. The owner pays one bill. The data available are global energy consumption of the entire building including both common and office space.

2. Landlord Operated

There are 64 such buildings. These buildings are owned and managed by landlords or his agents on a rental basis. There are two groups of data obtained, the landlords’ consumption for the common area and facilities, and the tenants’ consumption. The landlords’ consumption typically covers the central airconditioning system, artificial lighting (both indoor and outdoor) for the common area not within any tenant’s premises, vertical transportation system, mechanical ventilation systems and cleaning and other functions in the common area. The tenants’ consumption refer to that used within the tenants’ premises. These usually include consumptions by lighting, office equipment, computers, and other small appliances.

3. Mixed Landlord and Tenants Occupation

The monthly office energy consumption pattern in Singapore is different from that observed in the temperate zones. All the office buildings studied do not exhibit significant seasonal variation in energy consumption over the year. Figures 2 and 3 show some of the typical monthly energy consumption distribution over the year.

The monthly consumption variation over a year about the mean is 6.63% and 7.92% respectively for the total and landlord’s consumption. This may be attributed to the small variation in the external climatic conditions and the facts that the volume of commercial activities does not fluctuate significantly over the year.

Owing to the significant variation in building size among the sampled buildings, the energy performance of individual building is evaluated according to its energy efficiency. This is calculated using the formula,

Energy Consumption

GFA – GLA (1 - OR)

Where: GFA = gross floor area of the building in m² adjusted for car park area

GLA = gross lettable area in m2.

From eqn(1) above, it can be seen that the energy efficiency concerned has been normalized for the an operating hours of 55 hours and adjusted occupancy rate. This would provide better comparative study between buildings. Calculations have been made for both the total and the landlord’s consumption. In this way, the overall energy efficiency of a building, and individual landlord energy efficiency may be studied.

The distribution of the total energy efficiencies and the landlord’s energy efficiencies of the 104 buildings studied are given in Figures 4 and 5.

Figure 4: Frequency distribution of total energy efficiency of office buildings.

Figure 4 shows that the frequency distribution of the total energy efficiencies is largely a right skewed normal distribution with a skewness value of 0.89 indicating the closeness of the symmetry bell shape distribution. It follows that approximately 50% of the buildings studied have energy efficiencies below the mean value of 231.23kWh/m².

Figure 5 shows a similar frequency distribution for the landlord’s energy efficiency, with a very slight right skew to normal distribution. The skewness value is 0.73, and a mean value of 144.49 kWh/m².

The results therefore show that the energy efficiency and hence the various probability functions of the samples can be described using normal distribution function. Cumulative percentile curves may therefore be plotted for the two efficiency parameters.

Figure 5: Frequency distribution of landlord’s energy efficiency of office buildings.

The total energy consumption versus the gross floor area (GFA) of all the 104 buildings studied is plotted as shown in Figure 6. The results show that the energy consumptions of office buildings are directly proportionate to the gross floor areas. There is a correlation coefficient of 0.88 meaning that based on the GFA alone one can predict or estimate the energy consumption of a building to a good degree of accuracy with the exception of a small percentage of buildings. This can be a useful design benchmark for consultants or building managers. These results have been translated into a cumulative percentile distribution curve as shown in Figure 7. From the cumulative distribution curve, a building manager is able to determine the energy performance of his building in terms of percentile position based on the energy efficiency of his own building. This is a benchmark measure of his own building versus the energy performance of the cohort of 104 office buildings captured in this study.

Hence Figure 7 presents a useful data set for office buildings in Singapore where property manager can benchmark and target energy performance of his own building. Designers too can use this as a benchmark when they are designing new office buildings in Singapore. It is now possible for owners to identify and set realistic targets for the consultants to achieve, and in doing so ensure that his building stays within the top leagues of the industry. Building managers of existing buildings can determine where is the current performance level of his building and set realistic target to achieve for enhance energy efficiency. He can estimate the amount of saving possible and hence justify the amount of funding to achieve the target.

Figure 6: Distribution of total building energy consumption of office buildings

Figure 7: Cumulative percentage distribution curve of energy efficiency of office

Hence Figure 7 presents a useful data set for office buildings in Singapore where property manager can benchmark and target energy performance of his own building. Designers too can use this as a benchmark when they are designing new office buildings in Singapore. In another word, it is possible for a owner to identify and set a realistic target for the consultants to achieve, and in doing so ensure that his building stays within the top leagues of the industry. Building managers of existing buildings can determine where is the current performance level of his building and set realistic target to achieve for enhance energy efficiency. He can estimate the amount of saving possible and hence justify the amount of funding to achieve the target.

The data of the previous section is useful for measuring the total standing of an office building in Singapore. If a building manager found his building to be under performing, he may engage an energy services company to conduct an energy audit to determine where and how his building has under performed. In another word data with greater resolution or more details are needed. From a researcher’s point of view, further benchmark information is needed at lower levels such as landlord and tenant levels. It is also useful to determine the energy performance indicators and identify the key indicators. This is useful information for building designers and managers.

Figure 4 shows the distribution of energy consumption between the landlords and the overall consumption of tenants. The landlord’s consumption includes all the main and central airconditioning, lighting in common area, transportation systems and all other equipment such as pumps and ventilating fans and so on. The tenants’ consumptions are mainly in the areas of lighting and appliances including fans, printers, computers and so on. It can be seen from Figure 4 that the landlord’s consumption is 62% of the total consumption. The remaining 38% consumption is attributed to the tenants.

In this case, the results show a significant variation in the ratio of tenants to landlord energy consumption. This is due to the different arrangements for different buildings depending on the services and facilities provided by the landlord. However with a 95% confidence interval of +/- 2.80 % for the entire population of 104 buildings, it is reasonable to conclude that 62% of landlord consumption is a representative profile for the majority of the buildings studied.

Figure 8: Energy consumption distribution between landlord and tenants of office

The cumulative energy consumption curve of landlords is given in Figure 9 below. The total building consumption cumulative curve is also plotted for comparison. One can observe that the total building energy consumption is significantly spread as compared to that of the landlords’ consumption. All the 104 buildings studied are clustered in the range of 68 to 270 kWh/(sqm.yr). Energy performance ranking for landlord is therefore relatively more sensitive to minor increment. 75% of the buildings studied have energy efficiency indices of between 100 and 200 kWh/(sqm.yr). This is relatively competitive. In comparison, for total energy consumption, the bottom 20% of buildings had energy efficiency ranging between 300 and 460 kWh/(sqm.yr).

Figure 9: Cumulative energy consumption curve of landlords of office buildings.

The detailed consumption variations among the office buildings are presently being investigated under three grouping in terms of building energy consumption. The three different groups of buildings are those with energy efficiency (EE) falling within the top 25 percentile, those with EE falling between 26 and 75 percentile, and those with EE above the 75 percentile.

From the verified tenants’ energy consumption, a few main office user groups were investigated. They are the banking and finance offices, the trading and marketing firms, the professional consultants offices, and the representative offices. The cumulative energy consumption curves of these four types of offices are as shown in Figure 10.

Figure 10: The cumulative energy consumption curves of four tenancy groups.

Figure 10 results are plotted according to the EE of the various offices. The results show an interesting characteristic in that these users do not have linear energy consumption with respect to floor area. Hence for tenants in the same grouping, their energy consumption may be significantly different. Most notable is the group of representative firms, where energy consumption per square metre can vary from 25 kWh/(sqm.yr) to 370 kWh/(sqm.yr). This represents a 1400% increase. The professional offices exhibit the same trends. There is therefore important message for large tenants to examine their consumptions. A knowledgeable building manager can leverage on their core strength in energy efficiency and help his tenants to save energy and develop stronger owner and tenants business relationship.

There are two major groups of factors influencing the total energy consumption of buildings. They are:

1. Energy consumed by the physical building systems and elements in providing an internally desired environment. This group of factors includes the indoor environment settings, the various building services systems, and physical elements such as the envelope system.
2. Human factors such as the quality issues, habits and cultural background of the building users. This involves the thrift and attitude of building users, perception of owners, habits and educational standards of people and so on.

Researches of the 70s have shown that it is possible to bring about significant energy saving through improving building design and systems’ development. This saving is frequently negated by building users. For example, in the case of residential buildings, studies have shown that saving resulted from improved envelope design and construction is frequently absorbed by the corresponding increase in the standard of living resulting in zero real saving or even increased consumption. In another word, an improved envelope system does not automatically lead to a reduction of energy consumption at home. Instead it is more likely to have the energy consumption remained the same while the indoor temperature setting goes up and less clothing is worn.

In Singapore, although similar studies have not been attempted, it is likely that the same phenomenon will be observed. For instance, improved energy efficient design of home airconditioning system may lead to increased use of airconditioning and absorbed the real saving accrued.

This has the following implications:

1. Demand side management and the education of building users should not be neglected. It is equally important as compared to other areas of development in building energy efficiency.
2. For short-term improvement in energy performance of building, building managers and owners should focus on building systems, particularly those affecting the public or common areas where they are less influenced by users’ habit and responses.
3. It is important to develop integrated building energy efficient strategies which provide suitable level of users and environmental needs, and technology. Building energy efficiency practices must incorporate the human dimensions.

In enhancing the energy performance of buildings in Singapore, it is clear that detailed energy performance data and indicators are required as presented in this paper. To date, comprehensive data on the detailed energy performance of buildings in Singapore is still not available. This paper presented a small effort made over the last one and a half years on office buildings. Further significant efforts are needed, and such efforts must also be extended to other building types.

The information presented here allows building managers and designers to see what can be achieved for office buildings and able to set his targets for development. In addition to setting benchmark he can review consumption patterns within his building and compare them with the results reported here. This is useful. With this start the manager can develop his own strategies and database for enhanced energy performance. This can bring about significant energy saving.