Life-cycle energy modelling for urban precinct systems

Journal of Cleaner Production 142 (2017) 3254e3268

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage:

Life-cycle energy modelling for urban precinct systems Bin Huang a, *, Ke Xing a, Stephen Pullen b a b

School of Engineering, University of South Australia, Adelaide, South Australia 5095, Australia School of Natural and Built Environments, University of South Australia, Adelaide, South Australia 5095, Australia

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 May 2016 Received in revised form 24 October 2016 Accepted 26 October 2016 Available online 26 October 2016

Buildings are major contributors to global energy use and greenhouse gas emissions. There has been increasing effort and attention from industry and academia towards improving the energy efficiency of buildings and lowering the carbon footprint of this sector in the context of urban development. However, due to highly integrated and complex interactions between buildings, occupant behaviours and the surrounding environment, most current studies are largely limited to the energy modelling and assessment of individual buildings, rather than on the whole of a precinct system. This research proposes a precinct-scale life-cycle model to support the energy evaluation at the precinct level, as well as the optimal planning and redevelopment of urban precincts. The model encompasses three main types of energy components, i.e. operational, embodied and travelling energy, which are related to the physical and functional characteristics of an urban precinct. The model construct is underpinned by establishing a precinct baseline energy profile for precinct objects. It uses energy intensity measures and incorporates the impacts of environmental, morphological, and socio-economic factors of the precinct based on a systems perspective. The model application is demonstrated by a case study which analyses an outer suburban precinct in Adelaide, South Australia, which represents a typical example of urban expansion and transit-oriented development in Australian cities. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Life-cycle energy Energy intensities Building Precinct Environmental factors

1. Introduction Buildings are essential components of the urban built environment. The typologies of different buildings (residential, commercial, and industrial) and their mixes have a major impact on urban morphological configurations and their life-cycle energy performance. According to the analysis published by the International Energy Agency (2008), around 40.0% of global energy consumption can be attributed, both directly and indirectly, to the construction and operation of buildings. In Australia, the building sector contributes by approximately 20.0% to the overall energy use, nearly half of which is derived from commercial buildings (Australia Government Report, 2012). In recent years there has been a growing interest amongst urban planners and academia to study the life-cycle energy behaviour of buildings and urban forms with the aims of improving their performance, informing design and management decisions, and addressing associated environmental issues arising from modern

* Corresponding author. E-mail addresses: [email protected] (B. Huang), [email protected] (K. Xing), [email protected] (S. Pullen). 0959-6526/© 2016 Elsevier Ltd. All rights reserved.

lifestyles. According to the principle and framework defined by ISO14040 (ISO, 2006), the life-cycle energy of buildings can be divided into embodied energy and operational energy. Embodied energy is the energy required for the construction, maintenance and demolition of a building, while operational energy is the energy consumed during the operational phase. Embodied energy may be further split into initial embodied energy and recurring embodied energy based on different contribution types to the energy balance. Recent studies indicate that during the building lifecycle, embodied energy counts for approximately 10.0e20.0% of the total life-cycle energy demand in conventional buildings and can increase significantly to more than 60.0% in low-energy buildings (Dixit et al., 2010; Karimpour et al., 2014). Meanwhile, the operational energy offers great potential to reduce its share by 30e50% with the application of energy-efficient planning and design approaches. In addition, environmental impacts of buildings' space heating, cooling and natural lighting, which consequently affects the life-cycle energy consumption of buildings, are also widely observed (Newman and Kenworth, 1989; Steemers, 2003; Wong et al., 2011). Despite their efficacy in measuring energy performance of a building form, the extant studies and approaches are generally

B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

limited and less suited for the energy modelling and assessment at a precinct scale. There are a number of reasons for this. Firstly, buildings are inevitably dependent on and influenced by the surrounding natural, built and socio-economic environments in which they are established, occupied and operated. Therefore, assessing the energy and emission performance of a precinct by just summing the life-cycle energy values of individual buildings in the precinct might result in significant truncation errors (Huang et al., 2015). A further reason is that the large-scale analysis has been constrained by a large number of input parameters, as well as the complexity and uncertainty of urban flows (Ratti et al., 2005). The amount of input parameters employed for detailed physical and geometrical modelling of buildings will be significantly increased with the extension of a modelling boundary. The interactions between buildings and the environment also become more complicated and uncertain due to the large number of buildings included in precinct-scale modelling. Additionally, a heavy dependence on expert knowledge as well as detailed design information and material data is another concern in applying extant building models at an early stage of precinct planning when detailed information is normally scarce (Hviid et al., 2008). Moreover, the morphology and functions of an urban precinct are shaped by the cluster of various types of buildings and infrastructure established within the geographical boundary of the precinct (Jaccard et al., 1997; Hyman, 2013). Taking these factors into account, modelling and assessment of energy profiles at the precinct level needs to also take into consideration the life-cycle energy performance of relevant infrastructure systems (such as energy, transport, water and waste processing) and their operations, in addition to that of buildings. By examining current approaches for building and precinct energy measures, it is observed that the challenges in practice and research of precinct energy evaluation in existing studies can be summarised in three aspects as follows. Firstly, the system boundary is a critical component that directly affects the completeness and accuracy of precinct energy evaluation. However, the unclear and inconsistent definitions of system boundary employed in current models and exercises for precinct-level energy evaluation present as a major concern. Such divergences in current studies make comparative analysis which aims to gain a comprehensive understanding of precinct energy behaviours unattainable (Ding, 2005; Karimpour et al., 2014; Dixit et al., 2010). It is therefore commonly accepted that striving for a clear, subjective and comprehensive system boundary definition is essential. However, the consequent model complexity and computational burden also cannot be overlooked. Thus, a reasonable balance between accuracy and efficiency is critical in energy evaluation. Secondly, having an integrated model is another great challenge in precinct energy evaluation. Buildings and infrastructure (hereafter: precinct objects) are not isolated objects, but are evolving subsystems of a precinct which are closely linked with and affected by the occupant behaviours as well as the dynamic natural, built and socioeconomic environments where they are situated and operated (Steemers, 2003; Norman et al., 2006; Wong et al., 2011). Therefore, it is also necessary to incorporate the interplays of precinct objects, precinct occupants, and precinct environments for a holistic understanding and measure of the life-cycle energy of a precinct system in relation to the occupant behaviour and its geographical and environmental features (e.g. location, landscape, climate zone, etc). However, heavy computational burdens will be placed on integrated modelling with certain requirements such as large scale and variety of data input, highly uncertain variables, as well as the non-linear and complex interactions among buildings, occupants and environments. Finally, the identification of maintenance schemes and the lifespan of precinct objects is another concern which can affect the accuracy of evaluation or result in


inconsistencies and difficulties in comparative studies. Tracking and assessing long-lifespan objects, such as buildings and infrastructure, usually require considerable effort in data collection and interpretation (Dixit et al., 2012). Meanwhile, the states of precinct objects and the surrounding built environment can be dynamic in nature. Therefore, their maintenance, renovation and redevelopment are affected by various factors such as urban (re)planning, occupant choices, government policies, construction methods and climate conditions, which consequently makes defining actual lifespans and maintenance cycles for precinct objects a long-term and complex issue. Bearing in mind these current knowledge gaps and challenges, this paper aims to develop an integrated model to evaluate precinct energy performance with the considerations of environmental influence and occupant behaviors embracing three components: embodied, operational and travelling energy. Since the energy consumption can be converted into carbon emissions with a given energy/carbon dioxide equivalent conversion factor (kg CO2-eq/MJ) (which is determined by the local energy production profile) only the precinct energy is examined in this research. The proposed model is expected to provide a more comprehensive picture of precinct energy profile, which will consequently contribute to the optimal planning and (re)development of energy efficient urban environments. In this paper, Section 2 discusses system attributes and modelling challenges with the setting up of a system boundary. A system framework for constructing the precinct life-cycle energy model is discussed in Section 3 with some basic assumptions stated, followed by the formulation of mathematical models for precinct energy evaluation in Section 4. A case study is presented in Section 5 to demonstrate how the model works in analysing the life-cycle energy profile of an urban suburb in South Australia as a precinct system. Finally, Section 6 draws concluding remarks and suggestions for future work. 2. System attributes for modelling and evaluation Over the years, a growing body of research has been conducted to study the impacts on the actual energy consumption of buildings resulting from both physical factors (e.g. manufacturing, construction, maintenance, recycling, etc.) and occupant behaviours on building operations (e.g. comfort setting and operating schedules). In recent years BIM (building information modelling) models have also been developed to assist in the management of design features, construction material selection, building system information and project location, etc. (Volk et al., 2014). These studies and subsequent tools strongly support the optimal design and operation of buildings. However, although a building is a self-contained system with certain physical features, it is inevitably dependent on and influenced by the surrounding natural, built and socio-economic environments where it is established, occupied and operated. Thus, environmental factors such as precinct density, local climate and solar access should be integrated for a more comprehensive understanding and assessment of energy expenditures. In addition to these, and in order to support the optimisation of urban planning and resource distributions, it is essential to assess and understand the energy performance comprehensively at a macro level. It has to be kept in mind that a greater emphasis should be placed on the assessment of energy within a broader and more inclusive system boundary by taking a precinct perspective. 2.1. Nature of precinct scale modelling As the precinct is ‘a system of many interconnected systems’, a realistic, holistic and accurate evaluation of its energy performance can be achieved by integrating the life-cycle energy balance of


B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

precinct objects with the energy consumption contributed by occupant activities (e.g. daily commuting, recreational travelling, etc.). Therefore, at the precinct level, the energy consumption can be assessed with three distinct but inter-linked components: 1) total energy consumption embodied in the construction and maintenance of precinct objects, 2) total energy required for the operation of precinct objects, and 3) total energy related to occupant travel, including commuting for work, school, and leisure. Based on these inputs, the modelling of precinct energy consumption can be implemented in two phases. The downstream evaluation is focused on the baseline life-cycle energy of precinct objects, and the baseline energy (which is the average embodied, operational and travelling energy consumed within a precinct under ideal conditions) required by occupant travels with the consideration of environmental factors and occupant behaviours in the precinct context. Conversely, the upstream work aims to improve the evaluation accuracy through applying impacts resulting from urban precinct features. This is different from lifecycle energy evaluation of buildings where a great deal of detailed building design, building materials and construction features are emphasized. Hence, a precinct level evaluation would devote more effort to the influence from urban morphology, occupant activities and associated behaviours, as well as socioeconomic environment, since the evaluations are conducted with a systems perspective. 2.2. System boundary selection In the context of stand-alone buildings, processes such as raw material extraction, building components manufacturing, delivery, construction activities, operations, maintenance and demolition should be included in the system boundary setting with a consideration of the influence from the surrounding environment. As discussed in the previous section, the precinct-level or urban-level assessment can be formulated and evaluated with an extended consideration of embodied, operational and travelling energy. Therefore, the system boundary selected for precinct energy modelling and assessment is described as Fig. 1.

From the precinct perspective, it is necessary to consider the effects of time and spatial morphology on the environmental footprint assessment for a comprehensive understanding. Hence, the scope of a precinct-level or urban-level energy evaluation needs to be broadened and described in terms of both time and spatial dimensions. For the time dimension, operational energy changes over the whole lifespan of precinct objects, whereas the maintenance cycle, which consequently affects the overall embodied energy consumption of precinct objects, is determined by the servicelife of construction components and occupant choices. With respect to the spatial dimension, morphology, location and urban density contribute significantly to the operational and travelling energy components. Therefore, the system boundary for precinct level evaluations is expanded with the inclusion of energy consumption contributed by infrastructure and occupant travel, as well as more consideration of the influence from natural, built and socio-economic environments (as shown in Fig. 1). 2.3. Impacts of built environment As mentioned in previous sections, precinct objects are evolving subsystems closely linked to the dynamic surrounding environment. Built environment related factors such as urban form, interbuilding effects and residential density contribute significantly to the energy balance of precincts. Therefore, failing to integrate these factors in precinct energy assessment risks reaching biased or oversimplified solutions that have very limited effects on energy efficiency improvement. Over the years, many studies have been conducted in qualitative terms to support policy applications. However, a comprehensive understanding and specific identification of the built environmental contributions on precinct energy consumption remains sparse (Norman et al., 2006). Previous studies (Norman et al., 2006; Dujardin et al., 2012; Marique et al., 2013) indicate that built environmental factors can greatly and directly affect operational and travelling energy consumption, as well as the associated emissions. Despite the fact that built environmental factors might affect the selection of construction materials and design of precinct objects to some extent (which

Fig. 1. System boundary for global energy evaluation at the precinct level (Huang et al., 2015).

B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268


consequently result in variations of embodied energy expenditure) their direct influence on embodied energy evaluations is often overlooked due to their relative insignificant contributions (Morel et al., 2001; Lupísek et al., 2015; Cole, 1998). It is necessary to extend the current studies of building life-cycle energy assessment to the precinct level by integrating specific moderating coefficients. These adjust the baseline energy with the consideration of relevant effects imposed on objects and their energy behaviours by the built- environment features of a precinct. 2.3.1. Impacts on natural lighting The influence on daylight availability imposed by urban form and local climate, which consequently affects the energy consumption for artificial lighting, has been widely examined. A review of the relevant literature indicates that effective use of daylight can save 20.0e40.0% of lighting consumption (Brekke and Hansen, 1995; Embrechts and Van Bellegem, 1997). Although solar access might be affected by a great number of factors such as glazing ratio, U-values, building orientation, weather conditions and obstruction angle, at a precinct level only parameters such as obstruction angle and building orientation are commonly taken into account with the application of standardised construction components and building code in the local built environment. In this paper, a factor ml, reflecting impacts of obstruction angle and building orientation on the energy expenditure of artificial lighting, is employed to adjust the baseline lighting energy demand. As shown in Table 1, the orientation factors based on the studies conducted by Hunt (1979), Bodart and De Herde (2002), as well as Tzempelikos and Athienitis (2007) are used to recognise the influence on different building orientations. The models developed for internal luminance prediction which consider orientation factors may underestimate energy use by up to 40.0%. Nevertheless, the factors themselves are adequate to reflect the inherent correlation between building orientation and daylight access (Littlefair, 1998). Thus, despite falling short of providing accurate measures in an absolute sense, using these factors to adjust baseline energy consumption of artificial lighting is effective for examining and comparing relative differences in the impacts of various precinct configurations. Steemers (2003) explored the impact of obstruction angle on lighting energy requirement, and found that there is an approximate linear relationship between obstruction angle and lighting energy consumption (as shown in Fig. 2). 2.3.2. Impacts on heating, ventilation and air conditioning Apart from the factors such as occupant comfort setting and operating schedule, energy required for space heating, mechanical ventilation and air conditioning (HVAC) is also greatly affected by the surrounding environment. It is commonly accepted that for large buildings, the “Perimeter vs. Core” method is useful in zoningbased HVAC analysis, where the environmental influence on the HVAC energy consumption of core zones is noted as minimal. The surrounding environment plays an essential role in the HVAC energy evaluations of normal residential buildings and the perimeter area of large buildings. Studies conducted by Steemers (2003) and Pacheco et al. (2012) have examined the impacts on HVAC energy performance resulting in environmental factors such as urban density, obstruction angle and orientation. A common finding of these studies is that density might cause an approximate difference

Table 1 Orientation factors (based on Hunt, 1979). Orientation










Fig. 2. Effect of obstruction angle on energy use (adapted from Steemers, 2003).

of 4.0% on HVAC energy demand, orientation has a potential of 15.0% energy saving on space heating, while obstruction angle might contribute to the energy saving in space heating by 30.0% and that of space cooling by 20.0%. Fig. 2 adapted from a study conducted by Steemers (2003) also indicated the effect of obstruction angle on heating and cooling energy consumption. Since the influence of urban density can be affected by obstruction angle, a factor mHVAC, reflecting the impact of obstruction angle and building orientation on HVAC energy consumption, is applied for the modification of baseline HVAC energy consumption.

2.3.3. Impacts on travelling energy consumption In relation to the impacts on travelling energy consumption, factors like employment status, private car ownership, and availability and convenience of public transport are widely accepted as direct contributors. Furthermore, environmental effects such as residential density and urban morphology are also found to be of great importance in travelling energy evaluation by affecting the local traffic efficiency. Indeed, this effect becomes increasingly significant with the deterioration of local traffic which might result from population growth, increase in vehicle use, as well as decrease of liveable residential land. Early research completed by Newman and Kenworth (1989) indicates that the influence on travelling energy by urban density is likely to be very significant. This finding is supported by a similar study conducted by Steemers (2003). By observing these two studies, it is postulated that the correlation between travelling energy consumption and urban density can be approximated as shown in Fig. 3. In this research a factor mt reflecting the influence of urban density is introduced for the moderation of travelling baseline energy consumption.


B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

Fig. 3. Effect of population density on travelling energy consumption (based on Newman and Kenworth, 1989).

3. A framework and assumptions for precinct energy modelling In this section, a framework for the precinct energy modelling is developed in accordance with the system boundary defined in Section 2, with some basic assumptions specified to support model formulation.

baseline energy consumption. In this stage, parameters relating to the precinct environment, local climate, and occupant behaviours (e.g. total floor area of each building type, operation schedule of appliances, travelling frequency and distance, etc.) are identified to support the precinct baseline energy assessment. At Phase 3, in order to improve the accuracy of precinct energy evaluation, factors such as ml, mHVAC and mt corresponding to urban planning and context, as well as environmental factors are applied to adjust the baseline energy consumption. At this stage, the actual precinct morphology is transformed into a notional grid, which is a virtual precinct structure retaining the key spatial attributes such as average height of building blocks, number of blocks, as well as distances between blocks in north-south and east-west axes. Then obstruction angles are calculated with the parameters obtained from the notional grid. Finally, factors ml and mHVAC are identified based on the orientation factors and obstruction angles using the models presented in existing studies. However, at the current stage of this research, the factor mt is identified with the population density of precincts and the model reflecting effect of population density on travelling energy change.

3.2. Basic assumptions Since the number of variables, complexity of interactions and constraints significantly increase with the expansion of system boundaries, it is necessary to satisfy the following assumptions for the simplification of modelling and evaluation. Assumptions applied at the current stage can be classified into three categories according to the three types of energy expenditure which occur within urban precincts.

3.1. Precinct energy modelling framework As shown in Fig. 4, the modelling of precinct energy in this research represents precinct baseline energy moderated by the influence of the ambient environment. Initially, the baseline energy demand of a precinct is calculated based on energy intensities and estimated total floor areas of different precinct object types within the studied precinct taking into account the characteristics of local climate and precinct occupants. On this basis, several coefficients reflecting environmental influence are employed to adjust the baseline life-cycle energy value for contextualising the assessment and improving the accuracy. Corresponding to the features presented, the modelling framework consists of three phases: At Phase 1, energy intensities for the embodied and operational energy of each precinct object type, as well as the travelling energy of each transport mode, are identified. In this study, energy intensities of the main building materials are estimated following a process-based life-cycle analysis approach. The embodied energy intensity (MJ/m2) of each precinct object type is then identified based on the life-cycle energy intensities of the main building materials and the amount of each material required for construction and maintenance, with a consideration for energy consumed by construction (e.g. equipment use, onsite assembly, etc.) and transportation activities. It is necessary to keep in mind that the energy embodied in the manufacturing of vehicles for travelling as well as construction and maintenance related transportation is also included as a component of the overall embodied energy of a precinct. Operational energy intensities (MJ/ m2/day) are calculated based on a model presented in the following section with the baseline HVAC energy intensity obtained from statistical analysis of software simulations and references. The energy intensity of each travelling mode is collected or derived from published references (e.g. Davis and Diegel, 2007; Saunders et al., 2008). Phase 2 is designed for the evaluation of precinct

1) As stated in the previous section, the overall embodied energy is calculated based on energy intensities and floor areas or units. Therefore the following assumptions can be adopted:  properties and vehicles of a similar standard type have the same embodied energy intensity (MJ/m2 or MJ/unit);  the same capita floor area (m2/person) can be used for the default calculation of total floor area of some precinct objects when the relevant data is uncertain;  precinct objects and vehicles of the similar type have the same lifespan;  the maintenance cycle of each object type can be determined by the lifespan of some critical building components, or a yearly average energy consumption on a given object type can be applied to measure its recurring embodied energy consumption. The embodied energy of regular services for personal vehicles is not included in this model, as such services normally have a very limited scope and very low material intensity relative to the composition of a vehicle. 2) In this research, the identification of precinct operational energy is based on the following assumptions:  the operating schedule of non-residential properties is determined by considering working hours, surrounding environment and local working schedules, whereas that of residential buildings is determined by employment profile, age structure, family size and income status; and  normalized per unit energy consumption (MJ/m2/hour) can be employed for space heating and cooling in both residential and non-residential objects with the consideration of local climate; and  operational energy consumption of appliances such as hot water and white goods (e.g. fridge, dishwasher, washing machine and dryer) can be calculated according to the average per capita daily consumption or daily operating cycles; and

B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268


Fig. 4. System framework for precinct energy evaluation.

 the operational energy consumption is constant over the whole lifespan of precinct objects and appliances. 3) For the modelling of precinct travelling energy, travelling mode choice is determined by the availability of public transportation and private car ownership. Other pertinent variables, such as travelling distance and frequency, are predicted based on employment profile, commuting requirements and statistical data of travel behaviours provided by local governments (ABS, 2011). 4. Mathematical models for precinct energy evaluation As discussed in previous sections, the precinct-scale model proposed in this paper measures energy consumption contributed by both precinct objects and occupant travels. This section presents the approaches for calculating intensities of the main precinct energy components, as well as the overall precinct life-cycle energy profile. The model is structured such that alterative energy intensities can be adopted as more refined data becomes available.

infrastructure, roads, supply stations (e.g. gas, electricity, water stations, etc.), cable networks, sewer and stormwater systems are taken into account.

4.1.1. Intensities for embodied energy The development of a model for embodied energy intensity analysis is based on the principle of energy incurred per unit of floor area for the initial construction, renovation and demolition of properties, as well as the potential for recycling or reuse of critical building components. This research has developed a model for the evaluation of object life-cycle embodied energy intensity, which can be expressed as:

EEIi ¼ EEIIi þ EEIri þ EIci þ EItdi  EEIreci ¼

n1 X j¼1


þ EIci þ

4.1. Energy intensities for modelling According to usage, buildings are categorised into residential and non-residential buildings in this research. Residential buildings are further divided into detached houses, semi-detached villas, townhouses, and apartments due to different energy behaviours in construction, maintenance and operations. Non-residential buildings are those built for collective use purposes, such as offices, healthcare buildings, retail shops and schools. As for the

! n1   X   LPi Mi;j mi;j 1 þ ti;j þ Mi;j mj 1 þ ti;j 1 LCj


n1 X

 L mi;j 1 þ ti;j Pi etdi;j LCj j¼1

n1 X

 L mi;j 1 þ ti;j Pi erecj LCj j¼1

n1 X j¼1

L   mi;j 1 þ ti;j Mi;j þ etdi;j  erecj Pi þ EIci LCj (1)

In this equation, EEIi, EEIIi and EEIri are the intensities (MJ/m2)


B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

of overall, initial and recurring embodied energy respectively. EIci, EItdi and EIreci are the energy intensities (MJ/m2) required for onsite construction, transportation of components/materials, as well as the energy saving contributed by recycle or reuse of some building components/materials. Mi,j denotes the energy content of component/material (MJ/unit), which is estimated using the process-based LCA approach, and mi,j is the quantity of component/ material per floor area (unit/m2). The parameter ti,j represents the waste rate in construction and maintenance. The parameter etdi,j stands for an average energy demand for per unit of transportation and demolition (MJ/unit/m2) and erecj refers to an average per unit energy saving contributed by the recycle or reused of some construction components and materials (MJ/unit/m2), respectively. LPi indicates the lifespan of the precinct object type (i), whilst LCj is the lifespan of component/material. The mathematical operator is employed in the equation to find the least integer which is greater than or equal to a real number within. The embodied energy intensities of precinct objects are calculated based on:  the life cycle energy (including the amount of energy offset by materials recycle and components reuse) embodied in building materials/components,  quantities and replacement ratios of materials/components required for per m2 floor area, and  the energy expenditure on construction and related transportation activities.

4.1.2. Intensities for operational energy The operational energy is the energy required to operate infrastructure and buildings and maintain comfort conditions for occupants. It is broadly accepted that operational energy modelling should include HVAC (heating, ventilation and air conditioning), lighting, and other appliances. The types of appliances used are largely dependent on the types and functions of the building or infrastructure objects subject to analysis. In addition, operational energy demand varies depending on the local climate conditions, precinct morphology, comfort requirement and operating schedules. The developed precinct operational energy model can then be described as:

OEI ¼ OEIprop þ OEIinfr ¼ OEIl þ OEIhvac þ OEIap þ OEIinfr " n2 n3 X X ¼ OEIb;apk tyb  ðOEIlb  ð1  DARb Þ þ EUIb Þ þ b¼1


 tyb;k  nb;k þ

k¼1 n4 X

4.2. Travelling energy modelling Travelling for daily commuting and recreation is a major contributor to the increasing level of precinct energy consumption and carbon emissions worldwide. In a study on urban transport energy performance, Steemers (2003) stated that a conservative estimation of 2:1 ratio of building: transport energy use can be achieved for cities. It is partly because of the current availability and efficiency of urban public transport systems and the relatively low residential density and better traffic conditions that Australia has a high rate of private car ownership. The increase of private car travel has mirrored the relative decline in public transport use, which consequently causes higher travelling energy consumption since private vehicles typically consume more than twice the energy per passenger per kilometre than a train, and almost four times that of a bus (Steemers, 2003). Therefore, as a precondition, a better understanding of travelling energy is of great importance for the improvement of urban energy performance, especially for countries heavily dependent on low energy efficient traffic modes. The impact of travelling on carbon emissions is usually reported in term of tail-pipe emissions from vehicles. However, for a comprehensive coverage of travelling impact, energy consumption and associated carbon emissions caused by the manufacture of vehicles, as well as the construction and maintenance of transport infrastructure, should also be included. Based on such notions, a model adapted from the studies conducted by Zhou et al. (2013) is employed for the evaluation of precinct travelling energy consumption. With the input of more detailed information including local traffic situations, occupants' travelling behaviours and the operational energy consumption of different travelling options, this model can provide more accurate evaluation results. The formulation of the proposed model can be described as:



OEIInfrm  Rbm %  tym

operation schedules of representative buildings. Rbp% denotes the infrastructure floor area ratio expressed as % of object floor area. The parameters tyb, tym and tyb,k represent the annual operations hours of buildings, infrastructure and appliance (hours/year). OEIlb and OEIInfrm are the operational energy contents of lighting and infrastructure (MJ/m2/hour), whilstOEIb,apk and nb,k are the per unit energy consumption (MJ) and per floor area quantity (unit/m2) of appliance type (k) in building type (b). Lpc is the lifespan of the precinct.



¼ Lpc  popsize  @EEIt þ


(2) In this model, OEIprop and OEIinfr are the operational energy intensities of buildings and infrastructure (MJ/m2). OEIl, OEIhvac , OEIap and OEIinfr are the energy contents (MJ/m2) for artificial lighting, HVAC, appliance and infrastructure operations respectively. DARb is the daylight availability ratio, which is determined by factors such as window-to-wall ratio, orientation and obstruction. EUIb is the hourly per floor area HVAC energy content of building type (b) (GJ/ m2/hour), which is affected by local climate, precinct density, as well as the orientation, storey and zone allocation of buildings. In this context, the baseline operational energy consumption of HVAC can be obtained from simulations using a building operational energy evaluation tool, such as Energy Plus, with key input parameters such as local weather data, building footprints and envelope parameters of representative building types, HVAC system definition and zoning, temperature settings, as well as the

 EItmmin q

n5 X n6  X

 EItmmin þ EItmmax q q

q¼1 r¼1

  exp  mq  Vr  Utq %  dq þ EIta  Ua % 1

 da þ EItc  Uc %  dc A (3) Hereby, the per capita delivered transport embodied energy EEIt is employed to indicate the amount of energy expenditure on vehicle manufacturing. Previous research indicates that EEIt could be 8000MJ/person/year for outer and inner suburbs, and 11000MJ/ person/year for CBD area (Perkins et al., 2009). Meanwhile, the revised coefficient mq adopted from the model developed by Zhou et al. (2013) is utilised to moderate the travelling operational energy reflecting different traffic scenarios. EItmmin and EItmmax are q q the energy contents per unit distance at the free speed and the

B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

speed closing to zero respectively (MJ/km). Vr denotes the speed limit of road capacity (km/hour) and mq is the revision coefficient. Utq% represents the percentage of users choose in mode (q). The factor dq indicates the annual per person average travelling distance with model (q). EIta and EItc are the per unit distance energy contents of air and cruise travels per capita (MJ/km/person), respectively. Ua% and Uc% represent the percentage of occupants choosing air and cruise travel, with da and dc signifying the yearly average travelling distance by air and sea travel per capita. 4.3. Precinct energy modelling According to the modelling framework in Fig. 4, the precinctescale evaluations involve the energy embodied in infrastructure and buildings, as well as the energy consumption resulting in object operations and occupants' travelling. The overall lifecycle energy consumption of precinct (Epc) can then be calculated as the sum of the baseline energy components with the moderation in line with environmental factors.

Epc ¼

n X

 Ai  EEIi þ ½OEIli OEIhvaci ½ml mhvac T þ OEIap


 þ OEIinfri þ TE  mt


In this context, Ai indicates the total floor area of precinct object type (i), As discussed in previous sections, ml, mhvac and mt are the factors reflecting the effects of environment and urban morphology on lighting, HVAC and travelling energy consumptions respectively. In order to support the evaluation of precinct energy performance at the planning stage, the required total floor area of each object type can be predicted with the following assumptions: 1) The required floor area of each residential building type (Rfli) is determined by parameters such as age group, income and family size of occupants, as well as the average per capita occupant floor area of each building type (Cpi), while 2) the required floor area of nonresidential buildings and infrastructure is predicted based on their functions, occupants' employment profile, local land use regulations and precinct density. 5. Case study and discussion of results The selection of the case study for applying the precinct model is guided by the fact that the majority of the dwellings in Australia are separate houses (around 70.0%) despite there being a substantial increase in the construction of apartments and other higher density housing over the last few decades (ABS, 2011). As an outer suburb in South Australia, Andrews Farm appears to be a suitable option for the case study to demonstrate the application of the proposed model for precinct life-cycle energy assessment due to its geographical, morphological and demographical features. Andrews Farm was founded in 1991 as a subdivision in the Munno Para Council. Being a relatively new suburb with planned development, the land area is rectangular in shape, making it very convenient for intensity and land size based energy calculation as a demonstration case. Also, this suburb consists of residential dwellings, which are predominantly detached singlestorey houses, and has the population profiles highly representative of typical Australian outer-metropolitan suburbs, according to the data from the Australian census in 2011. Furthermore, this particular residential area situates within the Adelaide Northern Rail Corridor and has local bus services to a nearby major rail station (about 3.5 km). Whilst the majority of the households still travel by private cars, the suburb has the potential for an improved access to use public transport facilities for low carbon commuting,


which forms an essential part of transit-oriented development (TOD) for the future urban planning. 5.1. Profiles of the selected precinct The studied precinct, Andrews Farm, lies northwest of Adelaide the capital city of South Australia, about 30 km away from the CBD area (as shown in Fig. 4). It is bound by the Northern Expressway, Curtis Road, Stebonheath Road and Petherton Road with an overall land size of 273 ha. The 2011 Census by the Australian Bureau of Statistics counted 6244 persons in this precinct with 2311 private dwellings (2034 of them were occupied) and a labour force of 3062 (2794 of whom were employed). Although the land use policy for future expansion of this precinct is in favour of TOD, the occupants of this precinct are still highly dependent on private cars, with 91.4% of them owning at least one vehicle and over 70.0% driving their cars for daily commuting. Such a situation is largely due to the current infrequency of public transport with the consideration of low population density and limited local employment opportunities under the prevailing economic circumstances. More detailed profiles of this precinct with respect to object types and amounts, age groups, household structure, car ownership, employment status and commute mode choice are shown in Tables 2 and 3 and Figs. 5e9. 5.2. Results and discussions The precinct-scale life-cycle energy is derived by integrating the three main components: embodied, operational and travelling energy with the consideration of environmental factors and occupant behaviours. The approach developed in this research consists of using energy intensities in combination with factors reflecting the influence from environmental and occupant choices to obtain the embodied and operational energy consumption at the precinct level. While the energy use associated with occupant travel is measured based on vehicle embodied energy, efficiency of local traffic network (e.g. speed limits, maximum and minimum speeds), precinct population density and traffic mode choice determined by occupants (as included in Eq. (3)). Meanwhile, based on the density and orientation data obtained, the actual precinct morphology is converted and analysed as a notional grid with obstruction angles of 3.28 (in north-south axis) and 4.36 (in east-west axis) respectively. As discussed in Section 2, the values of lighting factor, cooling and heating are then estimated as 1.09, 1.08 and 0.99, respectively. With the consideration of annual heating and cooling days in the local climate zone, the factor mHVAC is finally set as 1.06. The travelling factor mt is determined as 1.35 based on the precinct population density and its effect on travelling energy consumption, as discussed in previous sections. Figs. 10 and 11 show the embodied, operational and travelling life-cycle energy balances, as well as their relative proportions in Andrews Farm amortised over a 60-year lifespan. The results indicate that at the precinct level, occupants travelling energy is by far the largest component, which accounts for almost 55.0% (425.25 TJ) of the precinct annual energy consumption. This may be attributed to the case precinct's geographical location and current morphological characteristics as a typical low density outer suburb

Table 2 Orientation of buildings in precinct. Orientation Proportion Orientation Proportion

East 6.7% Northeastern 19.3%

West 7.0% Northwestern 13.3%

North 10.1% Southeastern 16.3%

South 8.1% Southwestern 19.2%


B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

Table 3 Dwelling and infrastructure types and amounts in precinct. Dwelling type


Average area

Separate house Villa and unit Townhouse Dwellings Hospital (estimated as requirement) Office (estimated as requirement) Commercial (estimated as requirement) Schools (estimated as required) Roads network Pipe network Electrical network Driveway Footpath

2006 20 8 2034 e e e e 138 e e e e

161.22 100.00 186.00 160.35 e e e e e e e e e

Fig. 5. The case study precinct.

Fig. 6. Precinct employment status and household structure.

m2 m2 m2 m2

Total area 323,411.00 m2 2000.00 m2 1488.00 m2 326,155.00 m2 12,550.00 m2 20,972.00 m2 94,125.00 m2 15,995.00 m2 370,237.20 m2 66,635.00 m 66,635.00 m 74,908.80 m2 95,755.00 m2

B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268


Fig. 7. Precinct age structure.

Fig. 8. Car ownership of households.

Fig. 9. Occupants' daily commuting methods.

of an Australian city. The limited access to and use of the public transport system make the occupants in this precinct highly private car-dependent for daily work and school commuting, which is more energy intensive and identified as the main contributor to the high travelling energy consumption. The annual energy balance values shown in Fig. 10, however, are subject to variations in embodied energy, operational energy and travelling energy consumption. Due to the statistical differences and different LCA approaches applied, embodied energy intensity data from extant studies can vary to a great extent, which consequently affects the embodied energy evaluation of urban precincts. For instance, a previous study by Stephan et al. (2013) shows that the initial embodied energy intensity of a house can be as high as 19.17 GJ/m2. Therefore, further studies on how to derive and use embodied

energy intensity data are of great importance. Meanwhile, as discussed previously, factors such as building design and construction material selection also contribute significant amount on the variation of embodied energy. Precinct operational energy is usually influenced by occupant behaviours, as well as the density and morphological features of the precinct, while precinct travelling energy is greatly affected by travelling mode choice, travelling distance and traffic conditions. Such variations could be as much as 54.5% in embodied energy consumption, 51.9% in operational energy expenditure and 50.6% in travelling energy demand. A scenario analysis was conducted to study the impact of travelling mode selection on precinct energy consumption. Whilst Scenario 1 represents the current travelling modes of the occupants of Andrews Farm; Scenario 2 is based on the assumption that 30.0% of the occupants switch to public transport, taking shuttle buses to the nearby train station, and then taking trains for commuting. Results shown in Fig. 12 indicate that in comparison with Scenario 1 the overall annual energy balance of the studied precinct is reduced by 51.88TJ (or 6.3%) with a decrease of 34.74TJ in travelling operational energy consumption. Moreover, the results also illustrate that the amount of energy embodied in vehicles, which is usually overlooked in most studies, indeed comprises a considerable amount, as high as 6.9%, of the precinct overall travelling energy. This is also reduced by 17.20TJ with the assumed change of precinct occupant travelling choice. Therefore, improvements in energy efficiency of travelling methods, as well as energy efficiency for vehicle manufacturing and usage are of great potential for precinct energy savings. Meanwhile, the share of operational energy, which accounts for 32.0% (247.42 TJ) of the precinct annual energy demand (as shown in Figs. 10 and 11), is lower when compared with results from other similar studies with the inclusion of travelling energy consumption. A breakdown of the precinct operational energy (in Fig. 11) shows that HVAC is the main end use with a proportion of 48.8% of the overall operational energy demand. With the consideration of nonresidential buildings and infrastructure, as well as environmental impacts, this value is higher than the household HVAC data (38.0%) provided by the Department of Water, Energy and Environment of South Australia (DWEE, 2015). The energy contributions by appliances and lighting are 24.0% and 13.1% respectively, which are consistent with the results, 18.0% and 11.0%, obtained by Sustainability Victoria (2014). In contrast to the results obtained by the Department of Water, Energy and Environment of South Australia (DWEE, 2015), the outcomes of the case study show that the amount of energy consumed in heating water (DHW) is very low except for that amount consumed by residential and healthcare buildings. Building type is a critical factor in affecting the distribution of energy consumption among end uses, as illustrated in Fig. 13. This highlights the necessity for more extensive studies on operational energy profiles and intensities of different building


B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

Fig. 10. Precinct energy balance on an annual basis.

Fig. 11. Breakdown of precinct energy balance.

Fig. 12. Scenario analysis with changing of travelling mode.

types. Meanwhile, given the result that heating and cooling represents the biggest contributor to energy use of all building types (except for schools), improving the energy efficiency of HVAC is still

considered as the most desirable and effective option to reduce precinct operation energy. Furthermore, Fig. 13 also gives the adjusted operational energy intensities, which are based on the

B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268


Fig. 13. Breakdown of precinct building operational energy and intensities.

baseline operational energy intensities combined with the application of factors reflecting built environmental effects. The demographical profiles of the precinct occupants, such as age structure, household size and employment status, are considered in the settings of occupant behaviour and building operation schedule, which are applied to obtain the baseline energy intensities. At a lifespan of 60 years with the current building standards, the embodied energy associated with the construction and maintenance of buildings and infrastructure is a main contributor, although this amount has been reducing with the improvements of optimal planning, design and construction techniques, as well as the application of energy efficient materials (Harris, 1999; Iddon and Firth, 2013). As shown in Figs. 10 and 11, with the inclusion of travelling energy, the share of embodied energy dropped to about 13.0% (100.51 TJ annually on average). However, as shown in Scenario 1 of Fig. 12, if the energy expenditure for the manufacture of vehicles (which is part of the embodied energy of occupant travel) is added into the picture, the proportion would be as high as 19.0%. In addition, with the exclusion of travelling energy, the precinct embodied energy accounts for about 38.9% of the total energy consumption, which is slightly lower than the results obtained from the studies conducted by Thormark (2007) and Chen et al. (2001), but falls in the range concluded by Dixit et al. (2013), which is 9.0e46.0% of the sum of embodied and operational energy. Fig. 14 illustrates that residential buildings and infrastructure are the most energy intensive typologies typically accounting for about 77.2% of the total embodied energy consumption. This is followed by retail buildings, offices, healthcare buildings and schools. As discussed earlier, the literature reflects the need for an integrated modelling for precinct energy evaluations. However, most current studies in this field are focused at the building level. The model developed in this research is based on the strategy for building energy evaluation, but expanded to a macro level i.e. an urban precinct perspective with further considerations such as

Fig. 14. Breakdown of precinct embodied energy.

environmental factors, inter-building effects and occupants' behaviours. The developed approach is expected to support urban planning, precinct (re)development and precinct objects design with a comprehensive understanding of precinct energy performance, as well as the interactions among energy consumption, environmental effects and occupants' behaviours. The case study involved in this paper, which indeed has a number of limitations at the current stage, is conducted to demonstrate how this model works for the energy evaluation of an urban precinct. Firstly, the model developed in this research is based on energy intensities, therefore accurate assessment of precinct energy would not possible without a reliable local data inventory, and currently data gaps still remain which are filled through non-local data sources (e.g. data referenced from Germany and USA studies is used for


B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

travelling operational energy evaluation in the demonstration case study). Future LCA studies on precincts would benefit greatly from greater data availability and the model can be refined to adopt developments in energy intensities. In addition, the effects of occupant behaviour are critical for precinct energy evaluation. In the demonstrated case study, this factor is taken into account but simplified with statistical data due to local data availability. Future research on this aspect would be greatly improved with local data collection and big data technologies. In addition, assuming the quantities of appliances based on floor area might suffer significant limitations, especially for non-residential buildings. However, this could be improved with the statistical analysis of large amount of local data. Despite these limitations, this study provides one of the most detailed life cycle energy analyses of precinct energy efficiency regulations to date. Accuracy of the demonstrated case study might suffer from local data availability. However, the analysis shows how this developed model works. With the universal system boundary, assumptions and strategy for energy assessment, the developed model is able to simulate different scenarios of urban precinct planning and can support an optimal planning for low-energy urban precinct development.

In addition, with the development of renewable energy harvest technologies, rooftop renewable energy collection equipment such as solar thermal water heater and photovoltaic (PV) systems have found a wide application within precincts. Although the application of these systems might not be able to promote energy saving, they can offset the overall carbon emission of precincts to some extent. Future studies will focus on the statistical analysis for the accurate improvement of energy intensity data, and the evaluation of carbon offset contributed by renewable energy appliances. Moreover, the proposed model is expected to compare precinct scenarios in following studies and reduce the overall greenhouse gas emissions through applying optimal redevelopment on urban precincts. Acknowledgements Funding that permitted this research was granted by the Australia Cooperative Research Centre (CRC) for Low Carbon Living through the project “Integrated Carbon Metrics (ICM)” (RP2007) and University of South Australia. Appendix 1. Per unit energy contents of construction activities (Based on Chen et al., 2001; Biswas, 2014).

6. Conclusions and future research This paper has presented an approach for the evaluation of energy performance of urban precincts and includes the interactions among precinct objects, natural environment and occupant behaviours. The research has provided a model supported by a case study for exploring the life cycle energy consumption of urban precincts. The use of the model is aimed at comparing different scenarios for urban form for the lowest energy and emissions outcomes. Whilst the case study in this paper is focused on one typical outer suburb of an Australian city, the following indicative results summarised from the study can inform and provide a basis for further precinct life-cycle energy studies in similar contexts:  the advantage of having predetermined energy intensity data is that it relieves the user from the burden of detailed expert knowledge. Furthermore, the energy intensity data of materials, construction activities and operations can also represent the physical features of precinct objects in a relative simple way. However, the limitation raised from the employment of energy intensity data in modelling cannot be overlooked. Due to the substantial differences existing in parameters such as material selection, local climate, building form, land use, occupant behaviours etc. the variation of energy intensity data is always significant. Therefore, statistical analysis on a large body of samples and impact factors setting are crucial to reliably improve energy intensities and evaluation results.  energy expenditure on space heating and cooling is of great importance to precinct operational energy balance.  as a predominant component, occupant travelling energy with the travelling embodied energy, (where the latter was commonly considered as an insignificant factor in previous studies), comprises a considerable amount of overall precinct energy consumption, should be included in precinct energy evaluations.  therefore, for low carbon living in the future, more effort should be devoted to the improvements of heating and cooling efficiency, as well as the energy saving from manufacturing and operating of vehicles, which can be achieved through planning and design optimisations.

Activities Transportation Crane lifting Crane installation Crane operations Tree chipper Levelling Fences Lighting Computers (0.2 kW) Telephone (10 W) Printer (0.35 kW) Air conditioner (1 kW)

Energy content Activities 2

Energy content

1.32 MJ/m 0.0072 MJ/kg 0.92 MJ/m2 184.48 MJ/m2 2.29 MJ/m2 4.78 MJ/m2 0.23 MJ/m2 47.92 MJ/m2 1.34 MJ/m2

Excavation operation Concrete pump Concrete pouring Concrete curing Concrete elements curing Precast concrete operation Precast material operation Heating of components Heating of sheds

9.56 MJ/m2 0.56 MJ/m2 34.25 MJ/m2 0.158 MJ/kg 0.9 MJ/kg 6.43 MJ/m2 5.53 MJ/m2 93.6 MJ/m2 50.4 MJ/m2

0.022 MJ/m2

Ready mix truck

13.92 MJ/m2

0.16 MJ/m2

Mortar operation

7.39 MJ/m2

4.48 MJ/m2

Waste removal

2.85 MJ/m2

*The m2 unit used in this table indicates the floor areas of precinct objects.

Appendix 2. Waste ratios of some construction materials (Based on Williams et al., 2012; Chau et al., 2012; Dixit et al., 2013).


Waste ratio (%)


Waste ratio (%)

Asphalt/bitumen Brick/block/ceramic Cement Cast iron/mineral wool Copper Durasteel Floor covering Glass Insulation Timber

5 6 10.5 5

Aluminium/concrete/metal Gypsum wallboard/coatings Plaster/fibre glass Plastic/rubber/PVC

5.5 5 8e8.5 3e10

2.5e10 3 5.8 3e7 5e10 2.5e11

Polystyrene/polythene Precast concrete elements Steel/steel products/finishes Stone/sand/soil Special aggregates Tiles and slates

5 2.5e5 5e7 5 10 2.5

B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

Appendix 3. Lifespan of some major elements (Based on llar-Franca and Azapagic, 2012; Suzuki and Oka, 1998; Cue Iddon and Firth, 2013).


Lifespan Element (years)

Lifespan (years)

Roof waterproofing Roof lights




Outer wall




Suspended ceilings


Concrete frame Floor finishes


Floor covering (carpet/tiles) Lighting system Paint Insulation Substation

Roof covering (PVC/asphalt/concrete) Stairs (concrete/aluminium/steel/plastic) Floor (reinforced concrete/laminated) Hot water equipment (storage/instantaneous) Pumps (air-conditioning/water supply/fire) Walling (tile/aluminium/plasterboard) Pipe/internal drainage system Windows (internal & external) Solar/rain screen cladding Doors (internal & external) Electric cable/systems Ventilation/air handling unit Cooling/heat rejection plant

15e20 5-18/20 15 5e10 50 20

Packaged 20 air-conditioner

74/21/50/15 71/20 8/7 20e30

37/43/30-39 20e25 20e30 20 20e42 15e20 15e25 15e20

Appendix 4. Operational energy contents of some key appliances (Based on Korolija et al., 2011; Mattinen et al., 2014; Papachristos, 2015).


Energy consumption


Energy consumption

Hot water heating Office Lighting Computer e desktop Computer e laptop TV

0.216 MJ/L


36.29 MJ/h/each


0.062 MJ/m /hour Kettle 0.73e0.98 Air conditioning MJ/h/each 0.09 MJ/h/each Lighting heat Gain 0.23 MJ/h/each Equipment heat Gain TV receiver box 0.13 MJ/h/each Occupant heat Gain DVD player 0.09 MJ/h/each Refrigerator Projector 1.68 MJ/h/each Lighting (annual) Fax machine 0.1 MJ/h/each Car Park Lighting (annual) Electric oven 9.85 MJ/h/each Hydraulic & fire pumps (annual) Coffee maker 5.18 MJ/h/each Heating/cooling (annual) Telephone 0.05 MJ/h/each Dish washer (annual) Photocopier 18.14 MJ/h/each Freezer (annual) Lift 233.28 Dryer (annual) MJ/h/each Printer 1.0e2.01 Washing machine MJ/h/each (annual) Microwave 5.72e7.78 oven MJ/h/each

12.44 MJ/h/each 213.58 MJ/h/each 0.06e0.1 MJ//m2/hour 0.01e0.13 MJ//m2/hour 0.29e0.41 MJ//m2/hour 0.73e2.07 MJ/h/each 25.2 MJ/m2 5.3 MJ/m2 8.0 MJ/m2 223.8e264.28 MJ/m2 792e962.21 MJ/each 603.14e757.1 MJ/each 763.2e1464.12 MJ/each 594.14e619.52 MJ/each

*The operational energy contents listed in this table were converted from the power capacity of each appliance type with a local delivered-primary energy index of 1.44 obtained from the Australian Energy Production database.


References ABS, 2011. Australian Bureau of Statistics. ABS Stat BETA. B31 Dwelling Structure. URL: Australia Government Report, 2012. Annual Report 2011- 2012, Commonwealth of Australia (Canberra, Australia). Biswas, W.K., 2014. Carbon footprint and embodied energy consumption assessment of building construction works in Western Australia. Int. J. Sustain. Built Environ. 3, 179e186. Bodart, M., De Herde, A., 2002. Global energy savings in offices buildings by the use of daylighting. Energy Build. 34 (5), 421e429. Brekke, B., Hansen, E.H., 1995. Energy saving in lighting installations by the utilization of daylight. Proc. Right Light 3, 875e886. Chen, T.Y., et al., 2001. Analysis of embodied energy use in the residential building of Hong Kong. Energy 26, 323e340. Chau, C.K., et al., 2012. Assessment of CO2 emissions reduction in high-rise concrete office buildings using different material use options. Resour. Conservation Recycl. 61, 22e34. Cole, R.J., 1998. Energy and greenhouse gas emissions associated with the construction of alternative structural systems. Build. Environ. 34 (3), 335e348. llar-Franca, R.M., Azapagic, A., 2012. Environmental impacts of the UK resiCue dential sector: life cycle assessment of houses. Build. Environ. 54, 86e99. Davis, S.C., Diegel, S.W., 2007. Transportation Energy Data Book: Edition 26; ORNL6978. Oak Ridge National Laboratory, Oak Ridge, TN. Ding, G.K.C., 2005. Developing a multi-criteria approach for the measurement of sustainable performance. Build. Res. Inf. 33 (1), 3e16. DWEE, 2015. Department of Water, Energy and Environment of South Australia. URL: Dujardin, S., et al., 2012. Home-to-work commuting, urban form and potential energy savings: a local scale approach to the regional statistics. Transp. Res. Part A 46, 1054e1065. Dixit, M.K., et al., 2010. Identification of parameters for embodied energy measurement: a literature review. Energy Build. 42 (8), 1238e1247. Dixit, M.K., et al., 2012. Need for an embodied energy measurement protocol for buildings: a review paper. Renew. Sustain. Energy Rev. 16, 3730e3743. Dixit, M.K., et al., 2013. System boundary for embodied energy in buildings: a conceptual model for definition. Renew. Sustain. Energy Rev. 21, 153e164. Embrechts, R., Bellegem, V.C., 1997. Increased energy savings by individual light control. In: Proceedings of Right Light 4, Copenhagen, pp. 179e182. Harris, D.J., 1999. A quantitative approach to the assessment of the environmental impact of building materials. Build. Environ. 34 (6), 751e758. Huang, B., et al., 2015. Energy and carbon performance evaluation for buildings and urban precincts: review and a new modelling concept. J. Clean. Prod. http:// Hunt, D.R.G., 1979. Improved daylight data for predicting energy savings from photoelectric controls. Light. Res. Technol. 41, 261e283. Hviid, C.A., et al., 2008. Simple tool to evaluate the impact of daylight on building energy consumption. Sol. Energy 82 (9), 787e798. Hyman, K., 2013. Urban infrastructure and natural resource flows: evidence from Cape Town. Sci. Total Environ. 461e462, 839e845. Iddon, C.R., Firth, S.K., 2013. Embodied and operational energy for new-build housing: a case study of construction methods in the UK. Energy Build. 67, 479e488. International Energy Agency, 2008. Energy Efficiency Requirements in Building Codes e Energy Efficiency Policies for New Buildings. ISO, 2006. ISO 14040: 2006. Environmental Managementdlife Cycle Assessmentdprinciples and Framework. International Organization for Standardization. Jaccard, M., et al., 1997. From equipment to infrastructure: community energy management and greenhouse gas emission reduction. Energy Policy 25 (13), 1065e1074. Karimpour, M., et al., 2014. Minimising the life cycle energy of buildings: review and analysis. Build. Environ. 73, 106e114. Korolija, I., et al., 2011. Influence of building parameters and HVAC systems coupling on building energy performance. Energy Build. 43 (6), 1247e1253. Littlefair, P.J., 1998. Predicting lighting energy use under daylight linked lighting controls. Build. Res. Inf. 26 (4), 208e222. Lupísek, A., et al., 2015. Design strategies for low embodied carbon and low embodied energy buildings: principles and examples. Energy Procedia 83, 147e156. Marique, A.F., et al., 2013. School commuting: the relationship between energy consumption and urban form. J. Transp. Geogr. 26, 1e11. Mattinen, M.K., et al., 2014. Modelling and visualization of residential sector energy consumption and greenhouse gas emissions. J. Clean. Prod. 81, 70e80. Morel, J., et al., 2001. Building houses with local materials: means to drastically reduce the environmental impact of construction. Build. Environ. 36 (10), 1119e1126. Newman, P.W.G., Kenworth, J.R., 1989. Gasoline consumption and cities: a comparison of US cities with a global survey. J. Am. Plan. Assoc. 55 (1), 24e37. Norman, J., et al., 2006. Comparing high and low residential density: life-cycle analysis of energy use and greenhouse gas emissions. J. Urban Plann. Dev. 132 (1), 10e21. Pacheco, et al., 2012. Energy efficient design of building: a review. Renew. Sustain. Energy Rev. 16, 3559e3573.


B. Huang et al. / Journal of Cleaner Production 142 (2017) 3254e3268

Papachristos, G., 2015. Household electricity consumption and CO2emissions in The Netherlands: a model-based analysis. Energy Build. 86, 403e414. Perkins, A., et al., 2009. Transport, housing and urban form: the life cycle energy consumption and emissions of city centre apartments compared with suburban dwellings. Urban Policy Res. 27 (4), 377e396. Ratti, C., et al., 2005. Energy consumption and urban texture. Energy Build. 37 (7), 762e776. Saunders, M.J., et al., 2008. Incorporating transport energy into urban planning. Transp. Res. Part A 42, 874e882. Suzuki, M., Oka, T., 1998. Estimation of life cycle energy consumption and CO2 emission of office buildings in Japan. Energy Build. 28, 33e41. Sustainability Victoria, 2014. Victorian Households Energy Report. Accessed on December 10, 2015. URL: Steemers, K., 2003. Energy and the city: density, buildings and transport. Energy Build. 35 (1), 3e14. Stephan, A., et al., 2013. A comprehensive assessment of the life cycle energy demand of passive houses. Appl. Energy 112, 23e34. Thormark, C., 2007. Energy and resources, material choice and recycling potential in low energy buildings. In: Proceedings of the CIB Conference, SB 07 Sustainable Construction Materials & Practices. Lisbon, Portugal. Tzempelikos, A., Athienitis, A.K., 2007. The impact of shading design and control on building cooling and lighting demand. Sol. Energy 81, 369e382. Volk, R., et al., 2014. Building Information Modelling (BIM) for existing buildings e literature review and future needs. Automation Constr. 38, 109e127. Williams, D., et al., 2012. Climate change influence on building lifecycle greenhouse gas emissions: case study of a UK mixed-use development. Energy Build. 48, 112e126. Wong, N.H., et al., 2011. Evaluation of the impact of the surrounding urban morphology on building energy consumption. Sol. Energy 85, 57e71. Zhou, J., et al., 2013. Exploring the relationship between urban transportation energy consumption and transition of settlement morphology: a case study on Xiamen Island, China. Habitat Int. 37, 70e79.


Acronyms; Energy Intensities Measured by MJ/m2 EItd: transportation energy intensity EEII: initial embodied energy intensity EEI: overall embodied energy intensity OEIhvac: operational Energy content of HVAC OEIl: operational Energy content of Lighting EEIr: recurring embodied energy intensity OEIap: operational Energy content of appliances OEIinfr: operational Energy content of infrastructure OEIprop: operational energy intensity of buildings OEIinfr: operational energy intensity of infrastructure OEI: operational energy intensity of precinct objects EIc: construction energy intensity (as listed in Appendix 1) EEIrec: embodied energy intensity offset contributed by components/materials recycle/reuse Acronyms; Ratios without units DARb: daylight availability ratio of building type (b) Ua%: percentage of occupants choosing air travels Uc: percentage of occupants choosing sea travels Utq%: percentage of occupants choose in mode (q) ml: factor reflecting the effects of environment and urban morphology on lighting energy consumption mhvac: factor reflecting the effects of environment and urban morphology on HVAC energy consumption mt: factor reflecting the effects of environment and urban morphology on travelling energy consumption mq: revision coefficient for the travelling energy model

Rbm%: infrastructure floor area ratio expressed as % of total precinct building floor area ti,j: waste rate of component/material type (j) in object type (i) for construction/ maintenance (%, as listed in Appendix 2) Acronyms; Per capita travelling distance measured by (km/person) dq: annual per capita travelling distance with model (q) da: annual per capita travelling distance by air dc: annual per capita travelling distance by sea Acronyms; Lifespan (years) Lpc: lifespan of precinct LP: lifespan of precinct object LC: lifespan of components/materials (as listed in Appendix 3) Acronyms; Other variables EEIt: per capita delivered transport embodied energy (MJ/person/year) EItmmin q : energy contents per unit distance at the free speed (MJ/km/person) EItmmax q : energy contents per unit distance at the speed closing to zero (MJ/km/ person) EIta: energy intensity of air travels (MJ/km/person) TE: travelling energy (MJ) Vr: the speed limit of road capacity (km/hour) TOE: transport operational energy (MJ) EItc: energy intensity of sea travels (MJ/km/person) Ai: the total floor area of precinct object type (i) TEE: transport embodied energy (MJ) popsize: population size of precinct Mi,j: energy content of component/material type (j) in object type (i) (MJ/unit) mi,j: per floor area quantity of component/material type (j) required in object type (i) (unit/m2) etd: average energy demand for per unit of transportation and demolition (MJ/unit/ m2) erec: average per unit energy saving contributed by the recycle or reused of some construction components and materials (MJ/unit/m2) EUIb: hourly per floor area HVAC energy content of building type (b) (MJ/m2/hour) tyb, tym, tyb,k: annual operations hours of buildings, infrastructure and appliance (hours/year) OEIl, OEIInfr: hourly operational energy contents of lighting and infrastructure (MJ/ m2/hour) OEIb,apk: per unit energy consumption (MJ) of appliance type (k) in building type (b) nb,k: per floor area quantity (unit/m2) of appliance type (k) in building type (b) Symbols; Find the least integer which is greater than or equal to a real number within n1: amount of components/materials types n3: amount of appliances types n5: amount of traffic types n2: amount of precinct building types n4: amount of infrastructure types n6: amount of road types n: amount of precinct object types, n¼n2þn4 Subscripts i: precinct object type (i) k: appliance type (k) q: traffic mode (q) b: precinct building type (b) m: precinct Infrastructure type (m) r: road type (r) j: building component or material type (j)