Lr

La

b

a

ARR2A

KLEPW

1

frstownmosiul

i

0d

Energy and Buildings 47 (2012) 159–168

Contents lists available at SciVerse ScienceDirect

Energy and Buildings

j ourna l ho me p age: www.elsev ier .com/ locate /enbui ld

ife cycle greenhouse gas emissions and energy analysis of prefabricatedeusable building modules

u Ayea,∗, T. Ngoa, R.H. Crawfordb, R. Gammampilaa, P. Mendisa

Department of Infrastructure Engineering, The University of Melbourne, Victoria 3010, AustraliaFaculty of Architecture, Building and Planning, The University of Melbourne, Victoria 3010, Australia

r t i c l e i n f o

rticle history:eceived 15 October 2011eceived in revised form5 November 2011ccepted 26 November 2011

eywords:ife cycle energymbodied energyrefabricationaste minimisation

a b s t r a c t

Prefabrication is one strategy considered to provide improved environmental performance for build-ing construction. However, there is an absence of detailed scientific research or case studies dealingwith the potential environmental benefits of prefabrication, particularly the embodied energy savingsresulting from waste reduction and the improved efficiency of material usage. This paper aims to quan-tify the embodied energy of modular prefabricated steel and timber multi-residential buildings in orderto determine whether this form of construction provides improved environmental performance overconventional concrete construction methods. Furthermore this paper assesses the potential benefitsof reusability of materials, reducing the space required for landfill and need for additional resourcerequirements.

An eight-storey, 3943 m2 multi-residential building was investigated. It was found that a steel-structured prefabricated system resulted in reduced material consumption of up to 78% by mass

compared to conventional concrete construction. However, the prefabricated steel building resulted ina significant increase (∼50%) in embodied energy compared to the concrete building. It was shown thatthere was significant potential for the reuse of materials in the prefabricated steel building, representingup to an 81% saving in embodied energy and 51% materials saving by mass. This form of construc-tion has the potential to contribute significantly towards improved environmental sustainability in theconstruction industry.

. Introduction

The construction and operation of buildings is responsibleor significant environmental impacts, predominately throughesource consumption, waste production and greenhouse gas emis-ions. Treloar et al. [1] have shown the importance of consideringhe life cycle impacts of buildings, as the environmental impactsf initial construction can be just as significant as those associatedith their operation. The construction of buildings generates sig-ificant quantities of waste, on average up to 10% of the volume ofaterials used in constructing the building [2]. A large proportion

f this waste goes to landfill where, in Australia, 42% of the totalolid waste stream is construction and demolition waste [3]. Thiss contributing to the rapid depletion and inefficient use of our nat-ral resources and energy and resulting in increased pressure on

andfill availability.Numerous strategies have been adopted in an attempt to

mprove the efficiencies of construction and reduce waste (design

∗ Corresponding author. Tel.: +61 3 8344 6879; fax: +61 3 8344 6215.E-mail address: lua@unimelb.edu.au (L. Aye).

378-7788/$ – see front matter. Crown Copyright © 2011 Published by Elsevier B.V. All rioi:10.1016/j.enbuild.2011.11.049

Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved.

for disassembly, lean construction and waste management). Byappreciating the principles of handling and using materials on site,attitudes to prevent waste are being developed and the construc-tion process is beginning to be managed more efficiently [4].

Another strategy for reducing construction waste involves theprefabrication of building components which has also been knownto reduce construction costs and time. This involves assemblingcomponents of the building in a factory or other manufacturingsite, and transporting complete assemblies or sub-assemblies to theconstruction site where the building is to be located [5,6]. This prac-tice is in contrast to the more conventional construction practiceof transporting the basic materials to the construction site whereall assembly is carried out. The prefabrication of buildings has beenshown to provide savings in construction waste of up to 52% [7],mainly through the minimisation of off-cuts [8], and can signifi-cantly improve the energy, cost and time efficiency of construction.

To reduce life cycle environmental impacts of buildings, theirservice life should be extended as much as possible [9]. The dura-

bility of the structure plays an important role. For example, thestructure of commercial buildings in Australia is typically designedto last 100 years; however the average service life of buildings inthe Melbourne CBD is closer to 25 years. This figure is based on

ghts reserved.

1 Build

ttybasicc

mdeo

2

ictwe(taaiswirbim

ewcetsotbfiae

2

ciaaogt

iabie

60 L. Aye et al. / Energy and

he observation that most major refurbishments or deconstruc-ions of office buildings in Melbourne happen within the first 20–30ears of the building’s life. The life cycle environmental impacts cane significantly reduced if the structural components of a buildingre designed to be durable and reusable. Innovative design of thetructural connections at the initial development stage is extremelymportant to ensure that the deconstruction/demolition processan take place efficiently to maximise the reusability of buildingomponents.

This study aims to quantify the potential life cycle environ-ental benefits of prefabricated modular buildings in order to

etermine whether this form of construction provides improvednvironmental performance over conventional construction meth-ds.

. Background

Waste minimisation strategies have been popular for some timen the construction industry. Many studies measure waste fromonstruction sites on the basis of either volume or mass, to gaugehe effect on disposal costs [10–12]. The savings from reducingaste can be best measured in terms of the environment by consid-

ring their embodied impacts [13]. For example, embodied energyi.e. the energy consumed during extraction, processing, manufac-uring, and transportation at all stages [14]) and natural resourcesre conserved when energy intensive materials are used efficientlynd waste is minimised [15]. Embodied energy is thus a usefulndicator of resource value and environmental impacts. Previoustudies have focused on the recycling potential of constructionaste and demolition materials, valuing waste in terms of embod-

ed energy [13]. However, there is an absence of detailed scientificesearch or case studies dealing with the overall environmentalenefits of prefabrication [7], particularly the embodied energy sav-

ngs resulting from waste reduction and the improved efficiency ofaterial usage.Guggemos and Horvath [16] have identified and quantified the

nergy required for the construction of two office buildings, oneith a structural steel frame and the other with a cast in place con-

rete frame. The study included the energy associated with materialxtraction, construction, use, maintenance, and end-of-life demoli-ion. The findings revealed that the total life cycle energy use of bothteel and concrete framed buildings were comparable. An analysisf the construction elements showed that the concrete slabs con-ributed the greatest to the overall embodied energy of the studieduildings. A further study of office buildings has supported thisnding, indicating that the reinforced concrete in slabs and beamslone can contribute from 59 to 67% of a building’s total embodiednergy [17].

.1. Embodied energy analysis methods

At best, studies that have compared the life cycle energy asso-iated with conventional and prefabricated construction methodsn the past, have used incomplete methods of embodied energynalysis, known to exclude up to 87% of the energy requirementsssociated with construction [18,19]. There are a number of meth-ds that can be used to estimate the energy consumed in providingoods and services. The accuracy and extent of analysis depends onhe method chosen [19,20].

Embodied energy analysis techniques can be classified broadlynto three separate groups, process analysis, input–output (I–O)

nalysis and hybrid analysis [21,22]. Process analysis uses a com-ination of process-, product- and location-specific data from

ndividual manufacturers or suppliers to calculate embodiednergy. This approach is generally seen to produce results that are

ings 47 (2012) 159–168

more accurate and relevant to the product being analysed, but onthe other hand, the collection of process data can be labour- andtime-intensive. Moreover, this approach suffers from a systemicincompleteness, which is due to the representation of the productsystem by a finite boundary, and the omission of contributions out-side this boundary. This truncation of the product system is oftendue to a lack of available data and whilst the magnitude of thesetruncation errors depends on the type of product or activity, it canbe in the order of 50–90% [see for example 19,23,24–27]. This issueof system boundary incompleteness in process analyses is not aproblem that brute force can solve, even with practically unlimitedresources. As a result, life cycle assessment (LCA) based on processanalysis does not usually cover the input system of the functionalunit to a sufficient degree. Lenzen [27] and Treloar [20] have shownthat within conventional process-type LCA, this error is not usuallyreducible to an acceptable level by extending the system boundarybecause of the complexity of the supply chain that would have tobe investigated.

Input–output analysis uses sector-based financial data in orderto trace energy requirements between industry sectors [1]. Com-bining this financial data with energy tariffs it is possible to quantifythe energy embodied in any product or service produced withinthe economy. Generalised input–output frameworks have beenapplied extensively to environmental analysis since the late 1960s(see for example [28,29]). An introduction to the input–outputmethod and its application to environmental problems can befound in papers by Leontief and Ford [29], Proops [30] and Dixon[31].

The advantage of input–output analysis is that it provides a sys-temically complete analysis of the energy requirements associatedwith any product system as it covers completely the economic sys-tem defined by the national or regional statistics. These modelstreat the whole economy as a system and any number of inputsfrom other sectors are included, an almost limitless number ofpotential transactions upstream through the supply chain. How-ever, as it is based on national average data and commodities areaggregated into sectors, its use limits the applicability and relia-bility of results for any specific product. However, Lenzen [27] hasused Monte Carlo simulation to show that the uncertainties asso-ciated with input–output analysis are lower than the truncationerrors of a process analysis.

Due to the quite specific limitations with process analysis andinput–output analysis, some have sought to combine the best ofboth methods in an attempt to minimise the limitations and errorsof these traditional assessment methods. A number of researchershave suggested and demonstrated hybrid approaches [inter alia20,27,32–37]. These hybrid methods combine process data and I–Odata in a variety of formats [inter alia 20,27,30,33,38] with I–Odata used to fill the gaps that are traditionally excluded from theprocess-based system boundary [30].

In a tiered hybrid LCA [21,33], the direct and downstreamrequirements (for manufacture, use, and end-of-life), and someimportant lower-order upstream requirements of the functionalunit are examined in a detailed process analysis, whilst remaininghigher–order (‘upstream’) requirements (for materials extractionand manufacturing) are covered by input–output analysis. In thisway, advantages of both methods, completeness and specificity, arecombined. Moreover, the selection of a boundary for the productionsystem becomes obsolete. ‘Upstream’ truncation error is eliminatedfor the inputs for which process data has been collected. But usingthis tiered hybrid technique it is still virtually impossible to identifyall individual requirements of goods and services, relying on the

consultant to decide which processes are important, and requireanalysis. This can only resolve the upstream truncation error foritems that the user decides are relevant. As the supply chain is dis-aggregated to allow the integration of process data, the potential

Buildings 47 (2012) 159–168 161

ettcwt

iIcoptehi[pa

simtsiibot(ndotp

3

atteab

3

tvtcosbsaasflda

Table 1Densities and embodied energy intensities of basic construction materials.

Material Density(kg/m3)

Unit Embodied energyintensity (GJ/unit)

Concrete (30 MPa) 2400 m3 5.48Concrete (50 MPa) 2400 m3 8.55Structural steel 7850 t 85.46Glass (4 mm) 2600 m2 1.72Cellulose insulation (R2.5, 100 mm) 43 m2 2.17Plasterboard (10 mm) 950 m2 2.07Plywood 540 m3 10.92Aluminium 2700 t 252.60Timber (softwood) 700 m3 10.92MDF 500 m3 30.35Mortar 1900 t 2.00

L. Aye et al. / Energy and

xists for ‘sideways’ and ‘downstream’ truncation error. Sidewaysruncation includes inputs that are generally quantified in financialerms, such as advertising, as well as other items thought insignifi-ant. Downstream truncation can include the processes associatedith turning basic materials into more complex products, some-

imes difficult to quantify using physical data.The hybrid model developed by Treloar [20] (known as

nput–output-based hybrid analysis) starts with a disaggregated–O model to which available process data is integrated. The impli-ation of this is that the truncation errors associated with the usef process analysis are avoided and the analysis is more com-lete. The allocation of the various goods and services is accordingo their appropriate industrial sector in the input–output tablesnsuring that similar modelling principles are used. This approachas already been well established and demonstrated for calculat-

ng the energy embodied in buildings and other goods and services1,39–42]. A more in-depth analysis of the implications of using arocess analysis vs a hybrid analysis approach for embodied energyssessment is provided by Suh et al. [22].

Never before has a model utilising a systemically completeystem boundary been used to assess and compare the embod-ed energy associated with prefabricated construction. Due to

ainly the known deficiencies in the methods of analysis used,he knowledge gained from previous studies provides littleupport to industry in their need for environmental compar-sons between different construction approaches in order tonform design decision-making. Using the approach developedy Treloar [20,43], the current study extends on similar previ-us studies by providing a more comprehensive assessment ofhe energy embodied in prefabricated construction approachesresolving substantially the issue of system boundary incomplete-ess). The information provided by this study will facilitate theesign decision-making process and the environmental benefitsf prefabrication will be able to be better evaluated in ordero create buildings that are optimised for their environmentalerformance.

. Methodology

A multi-residential building has been used as a case study tossess the life cycle energy performance of prefabricated steel andimber constructions. This section outlines the case study buildinghat was analysed and the methods used to assess the life cyclenergy requirements associated with both conventional concretend prefabricated steel and timber construction approaches for thisuilding.

.1. Case study building

This study involved an assessment of the embodied and opera-ional energy associated with a multi-residential building, for threearying construction approaches, prefabricated modular steel andimber structures and a conventional concrete structure, used foromparative purposes. The building modelled has a gross floor areaf 3943 m2 with a total of 63 apartments consisting of 58 single-torey and five double-storey apartments. The first six floors of theuilding each consist of 9 single-storey apartments (Fig. 1) and theeventh floor consists of four single-storey and five double-storeypartments. The floor area of the single-storey and double-storeypartments is 63 and 118 m2, respectively. The ground floor con-

ists of seven tenancies together with other utilities. The groundoor and the sub-structure were not considered in this study. Theetails of the external/internal walls and the floor/ceiling panelsre for each scenario is by element (see Fig. 2).

Ceramic tiles 1700 m2 2.93

Source: Treloar and Crawford [48].

3.2. Embodied energy analysis

Embodied energy accounts for the energy consumed during themanufacture of products and materials, including those resultingfrom the manufacture of goods and services used during this pro-cess. For example, the energy embodied in steel products, typicallycomprise energy for iron ore extraction, transporting and process-ing the iron ore, manufacturing the steel products and deliveryto site. Energy is also embodied in goods and services, includingcapital, utilised during these processes, and so forth. Many factors(including technology, fuel supply structures, region, product spec-ification and analysis method) can result in considerable variabilityin embodied energy data.

The embodied energy assessment for the case study build-ing was performed using an input–output–based hybrid analysis[20,43]. This method is applied using an I–O model of Australianenergy use, developed by Professor Lenzen, Department of Physics,The University of Sydney [44]. The base I–O data was taken fromthe Australian National Accounts [45] and combined with energyintensity factors by fuel type. The combination of these two sourcescomprises the I–O model. The model includes the value of capi-tal purchased in previous-years, and capital imported from othercountries, amortised over the capital item’s life (as described andanalysed in Lenzen and Treloar [46]). Capital refers to the equip-ment and machinery used to make or transport products. TheI–O model was used as the basis for the embodied energy anal-ysis of the case study building. The best available process datawas incorporated for specific material manufacturers as per theinput–output-based hybrid method [20]. Process specific data forthe energy from the manufacture of specific materials was obtainedfrom the latest available SimaPro Australian database [47].

The calculation of the energy embodied in the two structuralsystems for the case study building was based on the embodiedenergy intensities from Table 1, which includes the energy fromfossil fuel consumption. These intensities were calculated using theinput–output-based hybrid method, combining available processdata for the specific materials, with I–O data [48].

The quantities of the materials used for each construction sys-tem for the case study building were determined and multipliedby their respective embodied energy intensities. The sum of theseresults gave the total embodied energy for each structural system.The proportion of materials available for reuse for both construc-tion approaches was determined and the energy embodied in thesematerials was also calculated using the above approach. The energyassociated with the end-of-life demolition, disposal and reuse pro-

cesses (e.g. making good) of materials has not been included in thisstudy. Crowther [49] has shown that the energy associated with thisstage of a building’s life represents less than 1% of the building’s lifecycle energy requirement.

162 L. Aye et al. / Energy and Buildings 47 (2012) 159–168

Fig. 1. Standard floor plan for single-storey apartments. FKA [54].

Fig. 2. Details of the main material used in the building for prefabricated steel, concrete and prefabricated timber scenarios, by element.

Build

3

iciTsydt

(il

3

sserc

birtbpbnttt

3

faatfigw

TV

L. Aye et al. / Energy and

.3. Operational energy analysis

The operational energy associated with the case study build-ng was estimated using TRNSYS simulation software. Based on theharacteristics of the building as well as assumed heating and cool-ng schedules. The simulation was performed using the MelbourneMY data developed and provided by Morrison and Litvak [50]. Theimulation was performed on an hourly basis for a period of oneear maintaining an indoor air temperature range of 21–24 ◦C. Theetailed occupational schedules and gains were not considered inhis study.

The seasonal average heat pump Coefficient of PerformanceCOP) values of heating = 3.0 and cooling = 2.2 were used in estimat-ng the electrical energy requirement from the heating and coolingoad outputs.

.4. Life cycle energy

The life cycle energy requirements associated with the casetudy building over a 50-year period were calculated for alltructural scenarios. This was achieved by combining the initialmbodied energy values with total estimated operational energyequirements over 50 years, assuming no heat pump system effi-iency losses or improvements over time.

Embodied energy associated with replacement of materials anduilding components over the life of the building was not included

n the analysis, although during the life of a building this energy canepresent up to 32% of its initial embodied energy [51]. The extent ofhis depends on a number of factors, including the useful life of theuilding and the anticipated life of the individual materials or com-onents. It was assumed that material replacement rates for bothuilding scenarios would be similar as they relate mainly to exter-al and internal finishes and not to the building structure. Despitehis, the study represents a much more comprehensive approacho the embodied energy assessment of a multi-residential buildinghan has been previously undertaken.

.5. Greenhouse gas emissions

Whilst calculating energy consumption is important in identi-ying areas where significant reductions in consumption may bechieved, energy consumption figures alone do not necessarily give

good indication of the environmental impacts associated withhis consumption. The same quantity of energy, but from different

uel sources (including coal, natural gas, wind and solar) will resultn a wide range of impacts on the environment. The greenhouseas (GHG) emissions produced from the combustion of fossil fuels,hich supply over 86% of global energy needs, is one of the main

able 2olume, mass and embodied energy breakdown of case study building for prefabricated s

Element Volume (m3) Mass (t)

Steel Concrete Timber Steel

Columns and beams 18.2 91.3 18.9 142.6

External walls 565.4 1114.9 653.2 227.9

Floor panels 305.9 941.5 381.3 132.4

External cladding 5.8 5.8 5.8 15.8

Ceiling 143.9 568.8 222.8 184.0

Internal walls 40.8 50.1 40.8 56.8

Roof 1.8 57.1 12.2 13.9

Doors and windows 30.1 33.1 30.1 29.5

Floor tiling 19.3 19.3 19.3 34.7

Staircase 0.5 3.6 0.5 2.0

First floora 12.7 – 12.7 31.0

Total 1144 2886 1398 871

Total per m2 0.29 0.73 0.35 0.22

a First floor slab is included in floor panel element for concrete building scenario.

ings 47 (2012) 159–168 163

contributors to the world’s key current environmental issue, globalwarming. The quantification of GHG emissions from consumedenergy is seen as a good indicator of the overall environmentalimpact resulting from energy consumption.

3.5.1. Embodied energy-related emissionsDue to the difficulties associated with determining the propor-

tion of embodied energy supplied by the various fuel types withinall of the processes involved in manufacturing and supplying thecomponents of the case study building, an average emissions fac-tor of 60 kg CO2−e per GJ of energy has been used to calculate thegreenhouse gas emissions related to the embodied energy of allconstruction types [51].

3.5.2. Operational energy-related emissionsEnergy required for heating and cooling was assumed to be

provided by brown coal-fired electricity, common for residentialbuildings in Victoria, Australia. Using the primary energy factor(3.5 for electricity in Victoria, Australia [52]), estimated opera-tional energy figures were converted to primary energy terms toaccount for the impacts associated with the energy production.Emissions factor of 1.35 kg CO2−e per kWh of electricity [53] wasused to estimate the greenhouse gas emissions from the electricityconsumption figures.

4. Results and discussion

This section presents the results and discussion of the life cycleenergy analysis of the case study building for prefabricated steeland timber, and concrete construction approaches.

4.1. Embodied energy analysis

This section presents the results of the embodied energy analysisof the case study building for both concrete and prefabricated steelconstruction approaches. Table 2 provides a breakdown of the totalbuilding material volume, mass and embodied energy for the majorconstruction elements, for all construction types.

Whilst the total mass of the concrete building is over fourtimes greater than that of the prefabricated steel building, the totalembodied energy in the steel building is about 50% higher than thatof the concrete building. This is predominately due to the muchmore energy intensive processes involved in steel manufacture as

compared to concrete production, for an equivalent functional unit(in this case a building’s structure). For the timber building withsteel columns and beams the total embodied energy is about 10%higher than that of concrete building.

teel, concrete and timber scenarios, by element.

Embodied energy (GJ)

Concrete Timber Steel Concrete Timber

228.6 148.4 13,402 2106 13,9481153.0 234.8 12,457 7792 74451116.2 111.7 9802 9226 2775

15.8 15.8 4378 4378 43781184.1 235.3 8179 9564 4103

44.8 56.8 2799 1016 2799139.5 95.8 1307 853 1470

23.7 29.5 3799 2733 379934.7 34.7 280 280 280

8.9 2.0 186 60 186– 31.0 189 – 189

3949 996 56,778 38,008 41,3731.00 0.25 14.40 9.64 10.49

164 L. Aye et al. / Energy and Buildings 47 (2012) 159–168

Fig. 3. Materials volumes for the three building types, by element.

ttu4caa

fs9t

nef

requirements among the three building construction types inves-tigated. The difference shown in operational energy is due to thedifference in the thermal mass and heat transfer characteristics of

Fig. 3 shows the comparison of material volume for all construc-ion approaches. This shows that the external walls, followed byhe floor panels, contribute the greatest to the overall material vol-me for all building construction approaches, representing 49%,7% and 39% of total material volume, for steel, timber and con-rete, respectively. These areas are therefore where greatest wastevoidance benefits can be achieved, through extending material lifend maximising eventual reuse, minimising the impact on landfill.

The breakdown of embodied energy for both concrete and pre-abricated steel construction systems is shown in Fig. 4. For the casetudy building, the total embodied energy equates to 14.4, 10.5,.6 GJ per m2 of floor area for the prefabricated steel, prefabricatedimber and concrete construction systems, respectively.

As the energy embodied in the prefabricated steel system is sig-ificantly higher than for the conventional concrete system, thenvironmental benefits of maximising material life and potential

or material reuse for the steel scenario are also greater.

Fig. 5. Typical TRNSYS output: indoor air tempera

Fig. 4. Embodied energy of the three building types, by element.

4.2. Operational energy analysis

This section details the annual operational energy requirementsassociated with the case study building for all construction typesinvestigated. The TRNSYS simulation performed to determine theoperational energy required for each zone to maintain an indoorair temperature between 21 and 24 ◦C (see Fig. 5).

The heating and cooling load patterns behave similarly for all theconstruction types investigated. The estimated heating and coolingloads were used to calculate operational energy consumption forall construction scenarios by using the heat pump seasonal averageCOP values described earlier.

The annual operational energy for the building clearly indi-cates that in Melbourne the heating energy requirements are muchgreater than cooling energy requirements for residential buildings(Fig. 6). There is slight difference in total heating and cooling energy

the construction materials selected (Table 3).

ture (21–24 ◦C) vs ambient air temperature.

L. Aye et al. / Energy and Buildings 47 (2012) 159–168 165

Table 3Annual operational energy requirements for steel and concrete structural scenarios by square metre of floor area (NLA = 3943 m2).

Building type Annual operational electricity (kWh/m2) Annual operational primary energy (GJ/m2)

Heating Cooling Total Heating Cooling Total

Steel 27.4 6.9 34.3 0.3451 0.0865 0.4316Concrete 26.9 5.3 32.2 0.3386 0.0666 0.4052Timber 27.0 6.5 33.5 0.3408 0.0825 0.4233

0

5

10

15

20

25

30

35

40

Hea ting Coolin g TotalHea

ting

and

cool

ing

ener

gy (k

Wh/

m2 y

r)

Steel Concr ete Timber

Fig. 6. Annual operational energy requirements of the three construction types, perm2

4

crpfirssrm

4

rcpV

F

characteristics and slight difference in thermal mass. It should benoted that the bulk insulation levels for all building types meet theBuilding Code of Australia minimum requirements.

of floor area.

.3. Life cycle energy

The embodied and annual operational energy requirements cal-ulated above were combined to determine the life cycle energyequirements of the case study building for both concrete andrefabricated steel construction types over a 50-year period. Thendings are presented in Table 4 and Fig. 7. The life cycle energyequirements were shown to be greater for the prefabricatedteel scenario at 36 GJ/m2, compared to 30 GJ/m2 for the concretecenario. For all scenarios the total heating and cooling energyepresents a larger component of the total life cycle energy require-ents than do the embodied energy requirements.

.4. Life cycle greenhouse gas emissions

The embodied and annual heating and cooling electricityequirements estimated above were used to determine the asso-iated greenhouse gas emissions for the case study building using

rimary energy and greenhouse emission factors for Melbourne,ictoria.

0

5

10

15

20

25

30

35

40

Total energy for 50 yrOperational energy for 50 yrEmbodied energy

Ene

rgy

(GJ/

m2 )

Stee l Concre te Timb er

ig. 7. Life cycle energy requirements of the three construction types over 50 years.

Fig. 8. Embodied greenhouse gas emissions of the three building types, by element.

4.4.1. Embodied energy-related emissionsThe greenhouse gas emissions associated with the energy

embodied in the building were 3407, 2482 and 2281 t CO2−e forthe prefabricated steel, prefabricated timber and concrete build-ing types, respectively. The elemental breakdown of embodiedgreenhouse gas emissions for all construction systems is shownin Fig. 8. It is evident that the steel framed building has about 50%more embodied greenhouse gas emissions compared to the con-crete framed alternative. The embodied greenhouse emissions persquare metre of floor area are 864, 630 and 578 kg CO2−e for thesteel, timber and concrete construction systems, respectively.

4.4.2. Operational energy-related emissionsThe annual heating and cooling energy-related greenhouse

emissions are shown in Table 5. This clearly indicates that thereis no big difference in the operational energy-related emissionsbetween the concrete and prefabricated steel and timber build-ings. The difference is attributed by the differences in heat transfer

Fig. 9. Life cycle greenhouse gas emissions of the three construction types over 50years.

166 L. Aye et al. / Energy and Buildings 47 (2012) 159–168

Table 4Total life cycle energy over 50 years (NLA = 3943 m2).

Building type Embodied energy (GJ) HVAC energy over 50 years (GJ) Life cycle energy over 50 years (GJ)

Heating Cooling Total

Steel 56,778 68,036 17,049 85,086 141,864Concrete 38,008 66,753 13,126 79,879 117,887Timber 41,373 67,180 16,265 83,445 124,818

Table 5Annual operational energy-related greenhouse gas emissions for concrete and prefabricated steel building types (NLA = 3943 m2).

Structure type Annual operational emissions (t CO2−e) Annual operational emissions (kg CO2−e/m2)

Heating Cooling Total Heating Cooling Total

Steel 145.8 36.5 182.3 37.0 9.3 46.2Concrete 143.0 28.1 171.2 36.3 7.1 43.4Timber 144.0 34.9 178.8 36.5 8.8 45.3

Table 6Total life cycle greenhouse emissions over 50 years (NLA = 3943 m2).

Building type Embodied emissions (t CO2−e) Operational emissions over 50 years (t CO2−e) Life cycle emissions over 50 years (t CO2−e)

Heating Cooling Total

yTcecwosn

aeo

4

stpsmics

pb

TT

Steel 3407 7290

Concrete 2280 7152

Timber 2482 7198

Table 6 shows the total life cycle greenhouse emissions for a 50-ear life span for each construction type for the case study building.his indicates that the concrete structure results in a 13% less lifeycle greenhouse emissions than prefabricated steel building. Thembodied emissions contribute between 21 and 27% of the total lifeycle emissions. Including the greenhouse gas emissions associatedith maintenance and replacement of materials and components

ver this period would further demonstrate the importance andignificance of this embodied greenhouse gas emissions compo-ent.

The following section quantifies the materials likely to be reusedt the end of the building’s useful life and the potential embodiednergy and material savings from this reuse, as opposed to the usef virgin materials.

.5. Material reuse benefits

Reuse of construction materials can lead to significant resourceavings together with other environmental benefits from a reduc-ion in waste disposed of in landfill and the energy required for theroduction of virgin materials. A major advantage of prefabricatedteel and timber construction is the ability for construction ele-ents to be disassembled at the end of their useful life and reused

n a new building. On the other hand, whilst concrete can be recy-led as aggregate in new concrete, it is typically not possible to reuse

tructural elements from one building in a subsequent building.

The potential material resource and embodied energy savingsossible from the reuse of materials for both concrete and steeluildings are shown in Table 7, based on assumptions of the likely

able 7otal volume, mass and embodied energy of concrete and prefabricated steel building scen

Volume (m3) Mass (t)

Steel Concrete Timber Steel Co

Initial total 1144 2886 1398 871 3949Quantity reused 60 20 35 441 87Saving (%) 5.3 0.7 2.5 50.7 2

1827 9117 12,5241406 8558 10,8381734 8941 11,423

materials and respective quantities available for reuse. Whilst theconcrete construction system accounts for a greater volume ofmaterial than the steel system, and thus a greater potential forreducing the quantity of waste sent to landfill, the potential forembodied energy savings from the reuse of materials is significantlygreater for the prefabricated steel construction system.

The potential future reuse of a material can never be guaranteed.For this reason it does not make sense to allocate any environmen-tal credit to its initial use. However, if a material can be reused afterits initial use, the building in which the material is reused shouldbe credited with the embodied energy saving resulting from theavoidance of the energy required for processing and manufacturingnew virgin materials [1]. Designers should always attempt to usematerials that have the potential to be reused rather than disposedof at the end of a building’s useful life. Table 7 shows the compar-ison between the proportions of total material volume, mass andembodied energy savings from the reuse of building componentsfor both the concrete and steel building scenarios.

The study revealed that the reuse of even a small proportion(by volume) of embodied energy intensive materials at the endof the building’s useful life can result in a substantial saving inembodied energy for both concrete and prefabricated steel andtimber systems. The proportion of embodied energy able to besaved by reusing existing materials in a new building is up to 81.3%or 46,157 GJ for the prefabricated steel building, up to 69.1% or

28,584 GJ for the prefabricated timber building and up to 32.3%or 12,259 GJ for a concrete building. It should be noted that thesefigures do not take into account the ability to recycle materials,such as concrete into aggregate, for use in new buildings which can

arios, with quantity and proportion of potential savings from the reuse of materials.

Embodied energy (GJ)

ncrete Timber Steel Concrete Timber

996 56,778 38,008 41,373 335 46,157 12,259 28,584.2 35.6 81.3 32.3 69.1

Build

ae

5

tvarcmTiupcrtarbtob

tAtttife

teHcpiafTcHcnw

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

L. Aye et al. / Energy and

lso save substantial quantities of virgin materials and embodiednergy.

. Conclusions

This study has assessed the life cycle energy requirements ofhree forms of construction for a multi-residential building, con-entional concrete construction, prefabricated steel construction,nd prefabricated timber construction to determine the envi-onmental benefits offered by modularised prefabrications. Thisomparison used an innovative hybrid embodied energy assess-ent approach that has never before been used in this manner.

he study has shown that the prefabricated steel system resultsn a significant reduction in the consumption of raw materials ofp to 50.7% by weight. Despite this, the energy embodied in therefabricated steel building is up to 50% greater than that for theoncrete building. However, the additional benefit of the prefab-icated system is the ability to reuse a significant proportion ofhe structure at the end of the building’s life. This may result in

significant reduction in waste being sent to landfill and reducedequirements for additional virgin materials. At the end of theuilding’s useful life, up to 81.3% of the embodied energy of the ini-ial steel building can be saved by reusing the main steel structuref the prefabricated modules and other components in another newuilding.

There was also shown to be only a minor variance in the opera-ional energy requirements associated with the construction types.dditionally, the embodied energy component for all construc-

ion types investigated was shown to represent at least 32% ofhe total life cycle primary energy requirements. This reinforceshe importance of building embodied energy, particularly as rapidmprovements are made in buildings operational efficiency per-ormance, further increasing the relative significance of embodiednergy.

Form a life cycle energy perspective, over a 50-year period,he prefabricated steel scenario was shown to consume morenergy (16%, see Fig. 9) than for conventional concrete construction.owever, despite this the study has clearly indicated that prefabri-ated construction is capable of providing improved environmentalerformance over conventional construction methods if they are

nitially designed to be reused, either adaptively or through dis-ssembly. The reuse of materials may reduce the space requiredor landfill and the requirement for additional virgin raw materials.he choice of materials in the construction of buildings has a signifi-ant impact on the embodied energy requirements of construction.owever, embodied energy should be optimised in the broader lifeycle context, considering also the operational, recurrent, mainte-ance and end-of-life energy requirements and impacts associatedith buildings.

eferences

[1] G.J. Treloar, P.E.D. Love, G.D. Holt, Using national input–output data forembodied energy analysis of individual residential buildings, ConstructionManagement and Economics 19 (2001) 49–61.

[2] G.J. Treloar, H. Gupta, P.E.D. Love, B. Nguyen, An analysis of factors influenc-ing waste minimisation and use of recycled materials for the constructionof residential buildings, Management of Environmental Quality 14 (1) (2003)134–145.

[3] Productivity Commission, Waste Management, Report No. 38, Commonwealthof Australia, Canberra, 2006.

[4] E.R. Skoyles, J.R. Skoyles, Waste Prevention on Site, Mitchell’s ProfessionalLibrary, London, 1985.

[5] C.B. Tatum, J.A. Vanegas, J.M. Williams, Constructability Improvement UsingPrefabrication, Pre-assembly and Modularisation, Technical Report 297, Con-struction Industry Institute, Standford, USA, 1986.

[6] CIRIA, Standardisation and Pre-assembly Adding Value to ConstructionProjects, Report 176, CIRIA, London, 1999.

[

[

ings 47 (2012) 159–168 167

[7] L. Jaillon, C.S. Poon, Y.H. Chiang, Quantifying the waste reduction potential ofusing prefabrication in building construction in Hong Kong, Waste Manage-ment 29 (1) (2009) 309–320.

[8] M. Osmani, J. Glass, A. Price, Architect and contractor attitudes to waste min-imisation, Waste and Resource Management 2 (1) (2006) 65–72.

[9] L. Aye, M.A. Mirza, J.R.W. Robinson, Life cycle greenhouse gas emissions ofbuilding and construction: an indicator for sustainability, in: Proceedings ofthe MDCMS 1 – Vietnam First International Conference on Modern Design,Construction and Maintenance of Structures, Hanoi, 10–11 December, 2007,pp. S1–S6.

10] H. Johnston, W.R. Minks, Cost effective waste minimisation for constructionmanager, Cost Engineering 37 (1) (1995) 31–39.

11] P. Graham, G. Smithers, Construction waste minimisation for Australianresidential development, Asia Pacific Journal of Building and Construction Man-agement 2 (1) (1996) 14–19.

12] O.O. Faniran, G. Caban, Minimising waste on construction project sites,Engineering, Construction and Architectural Management 5 (2) (1998)182–188.

13] C. Thormark, Including recycling potential in energy use into the life cycle ofbuildings, Building Research and Information 28 (3) (2000) 176–183.

14] I. Boustead, G.F. Hancock, Handbook of Industrial Energy Analysis, Ellis Hor-wood Limited, Chichester, 1979.

15] W.R. Lawson, Building Materials, Energy and the Environment: Towards Eco-logically Sustainable Development, The Royal Australian Institute of Architects,Red Hill, 1996.

16] A.A. Guggemos, A. Horvath, Comparison of environmental effects of steel-and concrete-framed buildings, Journal of Infrastructure Systems (June) (2005)93–101.

17] A. Dimoudi, C. Tompa, Energy and environmental indicators related to con-struction of office buildings, Resources, Conservation and Recycling 53 (2008)86–95.

18] R.H. Crawford, Validation of the use of input–output data for embodiedenergy analysis of the Australian construction industry, Journal of ConstructionResearch 6 (1) (2005) 71–90.

19] R.H. Crawford, Validation of a hybrid life cycle inventory analysis method,Journal of Environmental Management 88 (3) (2008) 496–506.

20] G.J. Treloar, Extracting embodied energy paths from input–output tables:towards an input–output-based hybrid energy analysis method, Economic Sys-tems Research 9 (4) (1997) 375–391.

21] C.W. Bullard, P.S. Penner, D.A. Pilati, Net energy analysis: handbook for com-bining process and input–output analysis, Resources and Energy 1 (1978)267–313.

22] S. Suh, M. Lenzen, G.J. Treloar, H. Hondo, A. Horvath, G. Huppes, O. Jolliet, U.Klann, W. Krewitt, Y. Moriguchi, J. Munksgaard, G. Norris, System boundaryselection in life cycle inventories, Environmental Science and Technology 38(3) (2004) 657–664.

23] C.T. Hendrickson, A. Horvath, S. Joshi, L.B. Lave, Economic input–output modelsfor environmental life cycle assessment, Environmental Science and Technol-ogy 32 (7) (1998) 184A–191A.

24] M. Lenzen, C.J. Dey, Truncation error in embodied energy analyses of basic ironand steel products, Energy 25 (2000) 577–585.

25] P. Born, Input–output analysis: input of energy, CO2 and work to produce goods,Journal of Policy Modeling 18 (1996) 217–221.

26] M. Lenzen, G.J. Treloar, Energy embodied in buildings: wood versus concrete,Energy Policy 30 (3) (2002) 249–255.

27] M. Lenzen, Errors in conventional and input–output-based life cycle invento-ries, Journal of Industrial Ecology 4 (4) (2000) 127–148.

28] W. Isard, K. Bassett, C. Choguill, J. Furtado, R. Izumita, J. Kissin, E. Romanoff, R.Seyfarth, R. Tatlock, On the linkage of socio-economic and ecologic systems,Papers in Regional Science 21 (1) (1968) 79–99.

29] W. Leontief, D. Ford, Environmental repercussions and the economic struc-ture: an input–output approach, Review of Economics and Statistics 52 (1970)262–271.

30] J.L.R. Proops, Input–output analysis and energy intensities: a comparisonof methodologies, Applied Mathematical Modelling 1 (March) (1977) 181–186.

31] R. Dixon, Inter-industry transactions and input–output analysis, The AustralianEconomic Review 3 (115) (1996) 327–336.

32] L.B. Lave, E. Cobas-Flores, C.T. Hendrickson, F. McMichael, Life cycle assessment:using input–output analysis to estimate economy-wide discharges, Environ-mental Science and Technology 29 (9) (1995) 420A–426A.

33] S. Suh, G. Huppes, Missing inventory estimation tool using input–outputanalysis, International Journal of Life Cycle Assessment 7 (3) (2002) 134–140.

34] R. Heijungs, S. Suh, The Computational Structure of Life Cycle Assessment,Kluwer Academic Publishers, Dordrecht, Netherlands, 2002.

35] S. Suh, Functions, commodities and environmental impacts in anecological–economic model, Ecological Economics 48 (2004) 451–467.

36] A.H. Strømman, G.P. Peters, E.G. Hertwich, Approaches to correct for doublecounting in tiered hybrid life cycle inventories, Journal of Cleaner Production17 (2) (2009) 248–254.

37] M. Lenzen, R.H. Crawford, The path exchange method for hybrid LCA, Environ-mental Science & Technology 43 (21) (2009) 8251–8256.

38] F. Duchin, Industrial input–output analysis: implications for industrial ecology,Proceedings of the National Academy of Sciences of the United States of America89 (1992) 851–855.

1 Build

[

[

[

[

[

[

[

[

[

[

[

[

[

[

68 L. Aye et al. / Energy and

39] G.J. Treloar, A. McCoubrie, P.E.D. Love, U. Iyer-Raniga, Embodied energy analysisof fixtures, fittings and furniture in office buildings, Facilities 17 (11) (1999)403–409.

40] R.H. Crawford, G.J. Treloar, An assessment of the energy and water embodied incommercial building construction, in: Proceedings of the 4th Australian Life-Cycle Assessment Conference, Sydney, February, 2005, p. 10.

41] R.H. Crawford, Life cycle energy and greenhouse emissions analysis of wind tur-bines and the effect of size on energy yield, Renewable and Sustainable EnergyReviews 13 (9) (2009) 2653–2660.

42] R.H. Crawford, G.J. Treloar, R.J. Fuller, M. Bazilian, Life cycle energy analysisof building integrated photovoltaic systems (BiPVs) with heat recovery unit,Renewable and Sustainable Energy Reviews 10 (6) (2006) 559–575.

43] G.J. Treloar, Environmental assessment using both financial and phys-ical quantities, in: Proceedings of the 41st Annual Conference of theArchitectural Science Association ANZAScA, Geelong, November, 2007, pp.247–255.

44] M. Lenzen, S. Lundie, Input–Output Model of the Australian Economy Basedon Published 1996–97 Australian Input–Output Data, University of Sydney,Sydney, 2002.

45] ABS, Australian National Accounts, Input–Output Tables, 1996–97, ABS Cat. No.5209.0, Australian Bureau of Statistics, Canberra, Australia, 2001.

[

[

ings 47 (2012) 159–168

46] M. Lenzen, G.J. Treloar, Endogenising capital – a comparison of two methods,Journal of Applied Input–Output Analysis 10 (December) (2004) 1–11.

47] T. Grant, Australian Material Inventory Database of Life Cycle Assessment Val-ues for Materials, RMIT, Melbourne, 2002.

48] G.J. Treloar, R.H. Crawford, Database of Embodied Energy and Water Values forMaterials, The University of Melbourne, Melbourne, 2010.

49] P. Crowther, Design for disassembly to recover embodied energy, in: Pro-ceedings of the 16th International Conference on Passive and Low EnergyArchitecture, Melbourne, September, 1999.

50] G.L. Morrison, A. Litvak, Condensed Solar Radiation Data Base forAustralia, Report 1988/FMT/1, University of New South Wales, Kensington,1994.

51] G.J. Treloar, Streamlined life cycle assessment of domestic structural wall mem-bers, Journal of Construction Research 1 (2000) 69–76.

52] G.J. Treloar, A comprehensive embodied energy analysis framework, PhD The-sis, Deakin University, Geelong, 1998.

53] Department of Climate Change, National Greenhouse and Energy Reporting(Measurement) Determination 2008, Commonwealth of Australia, Canberra,2008.

54] FKA, Architectural Plans of Russell Place Development, Fender Katsalidis Archi-tects, 2009.