Issue1_2009.indb The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 9 2 The Inherent Building Energy-Cost Relationship: An analysis of thirty Melbourne case studiesYu Lay Langston and Craig Langston (Mirvac School of Sustainable Development, Bond University, Australia) ABSTRACT This study investigates the energy and cost performance of thirty recent buildings in Melbourne, Australia. Commonly, building design decisions are based on issues pertaining to construction cost, and consideration of energy performance is made only within the context of the initial project budget. Even where energy is elevated to more importance, operating energy is seen as the focus and embodied energy is nearly always ignored. For the fi rst time, a large sample of buildings has been assembled and analysed to improve the understanding of both energy and cost performance over their full life cycle. The aim of this paper is to determine the relationship between energy and cost using regression analysis for a range of building functional types. The conclusion is that energy and cost are strongly correlated, independent of building area, and equations are presented for future modelling of energy using cost as the independent variable. Keywords: energy-cost relationship, correlation, optimisation, energy prediction, Melbourne. INTRODUCTION Energy has become a signifi cant issue worldwide. Greenhouse gas emissions (GGE) and the perceived threat of climate change (caused by phenomena such as global warming and ozone depletion) is identifi ed by Beggs (2002; p.10) as driving, “more than any other issue”, change in energy consumption attitudes. Since the energy crisis of the mid-1970s attention has been directed towards strategies that lower operating energy demand (Robertson, 1991), yet it has been only recently that the impact of energy embodied in building materials themselves has come under scrutiny. Despite this trend, the routine analysis of embodied energy remains absent (Treloar et al., 2002). Proper energy analysis during the design process can no longer be simply overlooked. ASEC (2001; p.100) indicate that “total energy use has doubled [in Australia] over the last 25 years […] at a faster rate than GDP”. The rationale behind this paper fi rmly lies with the perceived lack of integration of energy analysis into current practice. Capital cost still remains the primary criterion for building procurement decisions (Brown and Yanuck, 1985; Langston, 1991; Bull, 1992), while other criteria are given less signifi cance either due to a narrow myopic focus (Ashworth, 1988) or because a suitable multi-criteria technique has not been satisfactorily identifi ed (van Pelt et al., 1990). Energy analysis is costly, time-consuming and, when undertaken during the design phase, usually based on a large number of assumptions (Verbeek and Wibberley, 1996). Even so, it is likely to produce confl icting advice to that generated from capital cost estimates (Arnold, 1993). This occurs because energy analysis takes a long-term view, one that introduces multiple stakeholders and wider social concerns, rather than merely refl ecting immediacy and profi t-centred objectives. It has been argued over many years (e.g. Stone, 1960; Kirk and Dell’Isola, 1995; Flanagan and Norman, 1983; Langston and Lauge-Kristensen, 2002) that costs should also be accounted over a longer time span. Known as life cycle costs (LCCs) or whole life-costing (WLC), these comprise both initial (capital) and recurrent (operating) components that can be aggregated to give a more realistic picture of the total expenditure commitment (Fuller, 1982). The problem essentially is how two criteria, one measured in fi nancial terms and the other in pure energy terms, can ever be reconciled to provide clear building design guidance. In fact, any comparison of particular material choices will usually indicate that cost and energy ratios vary widely (Irurah and Holm, 1999). Yet at the level of an entire building this differential is expected to be less – a view that is supported to some extent by the manner in which embodied energy intensities are often determined (i.e. from national input-output fi nancial tables) and operating energy interpreted (i.e. incurred cost). Embodied energy is defi ned as the energy used in building construction, including all the upstream processes such as raw material mining, manufacture, packaging and transportation to site. Operating energy is defi ned as the energy used in maintaining a comfortable indoor environment, including heating, ventilation and air conditioning, as well as lighting, power for equipment and other recurrent energy requirements. Both are measured in primary energy terms. Capital cost is defi ned as the expenditure used in building construction, including labour, materials, plant and overheads, but often excluding contingencies, professional fees, land acquisition costs and goods and services tax, which are all treated as purchase costs. Operating cost is defi ned as the expenditure required to clean, repair, maintain, replace and otherwise manage proper usage of the building over its life, including the cost of energy provision. All costs need to be expressed in real terms (constant dollars). If there is an inherent relationship between energy and cost that can be exploited to enable better design solutions to be identifi ed, then it should be possible to quantify energy directly from a life-cost investigation. The outcomes will naturally be dependent on the chosen time horizon for the study but will simplify the process of embodied energy calculation (in particular) that to date has proved elusive to common practice. Nevertheless, it should be remembered that the biggest cost to an organisation is usually the salaries of its employees, and that this matter has no real connection with energy demand. Salaries can be shown to represent more than 80% of life-costs (Evans et al. 1998; Langston 2005). For the purposes of this study, staff salaries and related occupancy costs are ignored. The purpose of this paper is to discover the extent of any relationship between energy and cost, using regression analysis, for a range of building types in Melbourne, Australia. From this information, a better understanding of facilities performance can The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 10 be obtained, leading to further insight into the relationship between energy and cost. The structure of this paper is to review previous research fi ndings, to outline the method adopted in this research, to analyse the results, and to make observations and draw conclusions for practice. BACKGROUND Previously published research about the relationship between energy and cost is largely confi ned to a few studies (Costanza, 1980; Costanza, 1984; Lavine and Butler, 1982; Oka et al., 1993; Ding, 2004). These are explored in more detail below. Costanza (1980) adapt input-output analysis to calculate the total (direct plus indirect) energy required to produce goods and services in the U.S. economy. He states that usually the energy required to produce labour and government services and the solar energy input to the economy is overlooked. The omission can be traced to the assumption that traditional primary energy factors of economic production (land, labour and capital) are independent. He makes the strong case that these input factors are not independent and that energy is required for their production. Embodied energy intensities can be calculated in this case by using input-output data. The results of such an analysis show that that there is a strong correlation between embodied energy and dollar values for a 92-sector U.S. economy if the energy required to produce labour and government services is included. Costanza (1980; p.1224) concludes his research with the observation that “the most important implication is that the physical dimensions of economic activity are not separable from limitations of energy supply”. He continues by arguing that the universally-appealing notion of unlimited economic growth with reduced energy consumption must be fi rmly put to rest beside the equally appealing but impossible idea of perpetual motion. It is easy to get a “free lunch” by looking only at small parts of the system in isolation, but when the entire system is analysed it becomes clear that the cost of “your lunch” is just being transferred to another part of the system. Costanza (1984) reaffi rms his earlier work. Input-output models of the economy, modifi ed to include households and government as endogenous sectors, are used to calculate each sector’s direct plus indirect (embodied) energy consumption based on estimates of the distribution of direct energy inputs to the sectors. Dollar value of sector output is highly correlated with this energy indicator. Therefore an economy can be said to operate on an energy theory of value since it can be scientifi cally proven to be proportional to an appropriate energy indicator. Lavine and Butler (1982; p.2) determine from their research that the dollar value of economic output is accurately predicted by embodied energy fi gures. They quote other studies by Brown and Lugo, Kemp et al. and Hall et al. (not separately cited), as well as Costanza, undertaken in the 1980s. In all cases strong correlations between the embodied energy value of inputs to an economy and the monetary value of the economy’s output are shown. They conclude: “Since the global system operates at approximately steady state, the net work output of the global system must equal the sunlight energy input. […] This ‘work output’ interpretation of embodied energy suggests the hypothesis that embodied energy values can predict economic values. The reasoning is that economic value is achieved in direct proportion to economic work, and economic work is in direct proportion to the work output achieved with the use of resource inputs to the economy. Because embodied energy is a measure of such work output, embodied energy also may be a measure of economic value.” Lavine and Butler (1982) also argue that embodied energy values can provide a consistent means for pricing environmental factors. Such prices may facilitate a more comprehensive consideration of the economic effects of many kinds of environmental policy decisions. The most relevant previous research, and most recent, is Oka et al. (1993). The total energy consumption and environmental pollution caused by construction are quantifi ed using input-output data (referred to as the Inter-Industry Relations Table). Six offi ce buildings in Japan, varying from 1,502 m2 to 216,000 m2, are evaluated. The major results are: • total energy consumption caused by construction of offi ce buildings is 8-12 GJ/m2 of fl oor area • CO2 production is 750-1140 kg, SO and SO2 production is 720-1430 g, NO and NO2 production is 700-1140 g, and dust is 70-130 g (per m2 of fl oor area) • the construction cost/m2 of fl oor space is proportional to the energy consumption and production of pollutants • structural work is shown to be high in energy consumption and CO2 exhausted per unit cost of construction, industrial waste exhausted in fi nishing work is higher than the other two categories of work (structure and equipment), while equipment work is shown to be relatively low in energy consumption, CO2, NO, NO2 and industrial waste produced They fi nd that the average values of energy consumption and pollutants released due to the construction of the six buildings is 32 MJ/1000 yen for energy consumption, 3.0 kg/1000 yen for CO2 production, 3.1 g/1000 yen for NO and NO2 production, and 1.0 g/1000 yen for industrial waste production. Although correlation coeffi cients are not shown, all cost, energy and pollutant relationships refl ect a tight linear pattern. Research completed by Ding (2004) concerns the use of multi- criteria analysis as a selection tool for sustainable development. She obtained detailed information relating to twenty educational (high school) projects in New South Wales, Australia. Among this data are fl oor area, embodied energy (construction), operating energy (including the energy embodied in forecast maintenance and repair work), capital (construction) cost and operating cost. All data are based on real performance where possible. The mean initial embodied energy across all projects is 8.05 GJ/m2, while the mean operating energy is 40.89 GJ/m2 or 0.68 GJ/m2/annum (this falls to 32.92 GJ/m2 or 0.55 GJ/m2/annum when recurrent embodied energy is removed). Floor area is measured as gross fl oor area (GFA1). The mean capital cost (A$2002) across all projects is $1,360/m2, while the mean operating cost is $1,946/m2 or $32.43/m2/annum. The cost of maintenance and repair cannot be extracted from the data. Floor area is again measured as GFA. METHOD The characteristics of the relationship between energy and cost require further study. In particular, it is important to determine whether the relationship is robust and independent of the infl uence 1GFA is defi ned as the fully enclosed fl oor area, measured to the internal line of external walls, plus the unenclosed covered area measured to the external face of external walls, at all fl oor levels of the building. The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 11 of building size. The selected research method adopted in this paper is sampling via case studies. Case study is an ideal methodology when a holistic in-depth investigation is needed (Feagin et al., 1991). It has been used in varied investigations, particularly in sociological studies but increasingly in construction. The procedures are robust, and when followed the approach is as well developed and tested as any in the scientifi c fi eld. Whether the study is experimental or quasi-experimental, the data collection and analysis methods are known to hide some details. Case studies, on the other hand, are designed to bring out the details from the viewpoint of the participants by using multiple sources of data (Tellis, 1997). The data for the research are drawn from actual case studies obtained courtesy of the Melbourne offi ce of Davis Langdon Australia, a large national quantity surveying practice. Case studies are located across Greater Melbourne and are specifi cally intended to refl ect a broad range of functional purpose. Capital cost data and fl oor areas are obtained directly from the elemental cost plans prepared by Davis Langdon Australia. Embodied energy intensities are estimated from the composite items of work listed in these documents using the input-output- based hybrid method (Treloar, 1998; Crawford, 2004). Operating cost is estimated from reasonable cycles for future maintenance and replacement work using LIFECOST™ software. Operating energy is based on data obtained from the Property Council of Australia for Melbourne offi ce buildings, adjusted to allow for extended opening hours for other functional uses. All costs are adjusted and expressed in fourth quarter 2006 dollars using published building price indices (BPI) also supplied by Davis Langdon Australia. None of these data have a direct relationship with GFA, but an indirect relationship (i.e. bigger buildings use more energy and cost more money) is implied. Thirty recent Melbourne projects are used as case studies. These projects represent diverse functions including provision of offi ce workspace, health facilities, residential accommodation, teaching and laboratory space, retail, hotel accommodation and a number of specialist uses. Projects comprise both new construction (73.3%) and redevelopment (26.7%). So-called residential projects, comprising apartment buildings and aged care facilities, account for 23.3% of the case studies, and the remainder are constructed for various other commercial uses. One-third of the case studies are hospitals. Projects range from 1997 to 2004, and vary in fl oor area from 249 m2 to 18,821 m2 GFA. The mean fl oor area is 3,749 m2 (coeffi cient of variation of 110.75%). They comprise a wide range of materials and standards, some are air-conditioned and some not, some have fi re sprinkler systems, some have loose furniture and special equipment, and some have substantial external works. This mix decreases the likelihood that projects exhibit similarities in energy and cost performance. Economies of scale also play a part in larger projects, which tend to have lower unit costs than identical designs of smaller size. The mix is therefore effectively random, enabling a range of statistical techniques to be applied to the sample. Table 1 lists the case studies used in this research by building type. Case studies are identifi ed by a numerical code, as the name and location of projects needs to be kept confi dential (this is a non- negotiable agreement made between the researchers and Davis Langdon Australia). Data supplied by Davis Langdon Australia comprise GFA and elemental capital costs based on hundreds of abbreviated measured quantities extracted from design cost plans. The full Building ID Year Building Type Gross Floor Area (m2) 1 2003 Residence (new) 1,409 2 2004 Residence (new) 450 3 2004 Residence (new) 1,791 4 2000 Office (new) 2,543 5 2003 Health Centre (redevelopment) 528 6 2003 Hospital (new) 6,761 7 2003 Residence (new) 328 8 2003 Information Centre (new) 1,223 9 2004 Hospital (redevelopment) 3,278 10 2003 Hospital (redevelopment) 3,760 11 2004 Library (new) 249 12 2004 Civic Hall (new) 625 13 2004 Primary School (new) 2,696 14 2001 Residence (new) 2,790 15 2001 Hospital (redevelopment) 5,677 16 2000 Hospital (new) 378 17 1999 Hotel (redevelopment) 652 18 2000 Car Parking Station (new) 5,412 19 1998 Hospital (new) 4,281 20 1998 Health Centre (new) 787 21 1998 Hospital (new) 1,159 22 1999 Hotel (new) 12,930 23 1999 Residence (redevelopment) 18,821 24 1999 University Building (new) 10,565 25 1999 Office (new) 4,704 26 1999 Hospital (redevelopment) 1,345 27 1999 Hospital (redevelopment) 5,940 28 1998 University Building (new) 2,502 29 1998 Residence (new) 5,223 30 1997 Hospital (new) 3,649 Source: Davis Langdon Australia (Melbourne Office) Table 1: Case study base information The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 12 project cost is presented, including Preliminaries, Site Works and External Services, and Special Provisions (such as allowances for loose furniture and equipment), but excluding contingencies, professional fees, land acquisition costs and goods and services tax. All other data are estimated using embodied energy models, promulgated operating energy targets (for Melbourne), expected maintenance and replacement cycles, and other operational assumptions. Embodied energy, including both initial and recurrent embodied energy, is determined using a sophisticated spreadsheet model. The model is an input-output-based hybrid method that embraces both process analysis data (where it is available) supplemented with the input-output data from published government statistics (1996-1997 fi nancial year), extracted and compiled at Deakin University by Dr Graham Treloar and Dr Robert Crawford. Looking at embodied energy, 41.65% of the overall calculation used process analysis data, varying between 25.56% and 62.60% across the thirty case studies (coeffi cient of variation equals 15.89%). Operating energy is estimated using a simple model based on occupancy hours per year and ‘good practice guidelines’ for new buildings in Melbourne (PCA, 2001)2. The latter translates to 0.56 GJ/m2 net lettable area per year (or 155.5 kWh/m2/annum), 2Note that the good practice guidelines were used in preference to the new building design target (PCA, 2001) in this study. The latter is a 28.5% reduction from the former, yet in the short-term this is unlikely to be achieved for the general run of projects except those that are specifi cally designed as energy effi cient. comprising 70% electricity and 30% gas (where gas supply is present), or 100% electricity (where no gas supply is present). Delivered energy is converted to primary energy using a factor of 2.72 for electricity (based on 80% brown coal at effi ciency=3.4 and 20% green power at effi ciency=1) and 1.4 for gas. Offi ce buildings assume nominal occupancy of 2,500 hours/annum (equivalent to 8am to 6pm Monday to Friday) as defi ned in PCA (2001). Hospitals, hotels, residential accommodation and car parking facilities are assumed to operate for 5,460 hours/annum (equivalent to 8am to 11pm Monday to Sunday), and this translates to an occupancy factor of 2.18 compared to offi ce buildings. Embodied and operating energy data are provided in Table 2. It is important to note that although an implied relationship exists between energy and GFA (i.e. bigger buildings use more energy both to construct and operate), the derivation of energy has no direct relationship to area. Embodied energy is modelled using energy intensities applied to hundreds of work items, and is effectively the true embodied energy of the project using the most comprehensive method available. Alternatively, operating energy is modelled more simplistically, and is a function of operating hours and occupancy profi le in relation to useable fl oor areas. Neither embodied nor operating energy are directly computed from GFA. Capital costs are converted to fourth quarter 2006 prices using a BPI provided by Davis Langdon Australia. Otherwise no adjustment to capital costs is undertaken and all unit rates are taken as correct and refl ective of the project given applicable market conditions at the time. The BPI for fourth quarter 2006 is 175.0 (later indices were not used as they were still forecasts at the time of analysis). Table 2: Case study energy summary Total Embodied Energy Building ID Floor Area (m2 GFA) Initial (GJ) Recurrent (GJ/yr) Other Operating Energy (GJ/yr) 1 1,409 26,741 370 2,448 2 450 9,571 148 979 3 1,791 36,679 429 2,924 4 2,543 60,326 550 2,504 5 528 9,558 103 584 6 6,761 154,157 1,688 12,156 7 328 6,815 119 546 8 1,223 21,397 279 1,299 9 3,278 77,893 785 6,771 10 3,760 96,512 1,003 7,427 11 249 5,991 91 263 12 625 15,696 186 664 13 2,696 49,445 695 2,148 14 2,790 62,048 947 6,115 15 5,677 134,281 2,376 10,751 16 378 9,009 100 740 17 652 10,405 309 1,420 18 5,412 105,937 149 3,551 19 4,281 112,160 1,345 8,657 20 787 18,322 204 821 21 1,159 28,008 436 2,218 22 12,930 286,656 3,943 29,947 23 18,821 260,255 4,354 37,177 24 10,565 263,068 2,839 11,564 25 4,704 121,541 1,226 4,998 26 1,345 24,983 403 2,578 27 5,940 149,738 1,332 11,615 28 2,502 57,218 607 2,593 29 5,223 96,058 1,602 9,568 30 3,649 78,622 775 6,973 Source: Langston (2006) The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 13 Operating costs, on the other hand, are estimated using LIFECOST™ software provided by Computerelation Australia Pty Limited. Maintenance and replacement cycles are determined using personal experience together with a number of useful references (e.g. Dell’Isola and Kirk, 1995), and priced by original unit rates with a suitable allowance for removal and disposal costs where applicable. All costs are adjusted to fourth quarter 2006 as before described. Capital and operating cost data are provided in Table 3. It is important to note that although an implied relationship exists between cost and GFA (i.e. bigger buildings cost more to construct and operate), the derivation of cost has no direct relationship to area. Capital cost is estimated by the quantity surveyor based on hundreds of work items, and is effectively the real cost of the project. Alternatively, operating cost is modelled based also on hundreds of work items, and is a function of component life expectancies, cleaning and repair cycles. Neither capital nor operating cost are directly computed from GFA. For all operating costs and operating energy, including recurrent embodied energy, a one-hundred-year time horizon has been assumed. RESULTS Base data are analysed in Langston and Langston (2007). The relationship between total life cycle energy (embodied + operating) and total life cycle cost (capital + operating) leads to a regression line of y=0.0137x, where y equals total life cycle energy and x equals total life cycle cost. As with all following correlations, the y-intercept has been assumed at zero (i.e. no energy, no cost). Figure 1 shows the regression results for the Melbourne buildings. The correlation between initial embodied energy and capital cost leads to a regression line of y=0.0071x. The Melbourne case studies are shown in Figure 2. This is an important relationship for the prediction of embodied energy based on construction cost estimates. Provided costs are expressed in fourth quarter 2006 terms, the embodied energy can be derived in minimal time using the constant (gradient) of 0.0071. The reliability of this constant is investigated in Langston and Langston (2008). Operating energy and operating cost, calculated over a one- hundred-year time horizon, leads to a regression line of y=0.0150x, as shown graphically in Figure 3. In all investigated cases, the relationships between energy and cost are clearly linear, and require no more complex regression method. VARIABLE DEPENDENCY An obvious criticism of this research is that both energy and cost are proxies for building area, and therefore all that has been shown is that bigger buildings use more energy and cost more money. In order to determine the ‘goodness of fi t’ (r2) for each relationship it is fi rst necessary to remove area from the comparison. This can be done in one of two ways. The fi rst way is to express energy and cost as unit rates (i.e. energy/m2 and cost/m2). The infl uence of building size is Table 3: Case study cost summary Operating Cost/yr ($2006) Building ID Floor Area (m2 GFA) Capital Cost ($2006) Recurrent Expenditure Energy Expenditure 1 1,409 2,999,243 75,221 33,654 2 450 1,198,625 44,716 13,462 3 1,791 3,881,208 102,215 43,189 4 2,543 6,321,146 293,590 42,622 5 528 832,897 34,125 8,033 6 6,761 20,213,443 790,532 211,057 7 328 575,722 45,054 8,301 8 1,223 1,984,097 106,580 17,863 9 3,278 10,932,154 308,775 112,286 10 3,760 13,188,297 379,602 119,721 11 249 577,654 27,142 3,622 12 625 1,710,606 65,176 9,129 13 2,696 4,140,979 265,684 29,723 14 2,790 7,901,075 456,106 103,254 15 5,677 16,603,374 750,830 212,207 16 378 1,174,502 60,533 15,183 17 652 1,325,701 107,333 25,804 18 5,412 5,857,972 207,995 60,434 19 4,281 12,754,657 501,092 191,499 20 787 1,806,207 97,070 15,450 21 1,159 3,042,888 213,702 48,087 22 12,930 43,838,966 1,609,751 539,706 23 18,821 41,392,112 1,670,800 792,257 24 10,565 36,750,343 1,043,406 215,991 25 4,704 16,328,325 597,484 90,831 26 1,345 3,143,639 160,966 54,933 27 5,940 19,778,684 612,987 247,515 28 2,502 5,796,121 238,294 47,947 29 5,223 13,119,590 660,692 179,993 30 3,649 9,032,456 393,959 156,952 Source: adapted from Langston (2006) The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 14 therefore limited to issues relating to economies of scale, where bigger buildings will tend to have slightly lower unit rates due to effi ciencies caused by the larger scope of works. But this method has the disadvantage of clustering (bunching) the project data around the average, making any correlation specifi c to the effect of the economies of scale only. For this reason this method is not recommended as leading to a practical and valid outcome. The second way is to compare energy with area (and cost with area) and then compare the residuals of both analyses, thus eliminating the GFA trends in each case. Thus, the underlying relationship between energy and cost is exposed after eliminating all dependency (whether direct or indirect) on building area. Residual (or error) represents unexplained variation after fi tting a regression model. It is the difference (or left over) between the observed value of the variable and the value suggested by the regression model. By comparing energy and GFA (or cost and GFA), the residuals for each represent the unexplained variation, and enable the independent relationship between energy and cost to be tested. It is obvious that energy or cost is positively correlated with building size, and confi rms the simple conclusion that bigger buildings demand more energy and cost more to construct or operate. The analysis of residuals, however, shows that even after building area is eliminated, there is still a strong relationship between energy and cost. The original regression exercises provide a model for predicting energy performance based on cost, while the residual regression exercises demonstrate the validity of the comparisons. The calculations for energy versus GFA residuals are presented in Table 4 and cost versus GFA residuals in Table 5. The energy versus cost scatter plots and regression details for each of total project, capital and operating data are shown in Figures 4, 5 and 6 respectively. The discovered correlation between energy and cost after eliminating any dependency on GFA is determined, and in each case the relationship has a p-value less than 0.05 (95% confi dence level) and a suffi ciently large t-statistic. Parametric line of best fi t has been used in this analysis. However, a non- parametric line of best fi t yields very similar results. In each case, a strong relationship between energy and cost, independent of GFA, has been proven. The r2 value for total project energy and total project cost is 0.5959. The r2 value for embodied energy and capital cost is 0.7081. The r2 value for operating energy and operating cost is 0.6264. DISCUSSION In all observed instances, whether at project, elemental group or element levels, higher energy values give rise to higher costs and vice versa. This is obvious, but in addition it has been shown that cost is a strong predictor of energy. While at the work item level it is feasible to spend more capital to purchase materials or systems Figure 1: Total project energy v cost correlation Figure 2: Embodied energy v capital cost correlation Figure 3: Operating energy v operating cost correlation The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 15 with lower embodied and/or operating energy demand, and thus introduce effi ciency and value for money, at the higher levels of aggregation this is unlikely to be signifi cant. This fi nding leads to the conclusion that optimisation of energy and optimisation of cost are not mutually exclusive goals. The very strong correlations between energy and cost, and particularly between embodied energy and capital cost, suggest that this bond cannot be upset by individual decisions made at more refi ned levels of detail. It therefore may be more important to look elsewhere for effi ciencies. In the case of embodied energy, savings can be found by building less. This may arise from more effi cient fl oor plans (i.e. less non-functional area), more effi cient plan shape (i.e. lower wall/fl oor ratios), more effi cient internal layouts (i.e. more open plan design), and higher fl oor densities (i.e. people/m2). In the case of operating energy, savings can be found, in addition to building less, by reducing demand through use of natural lighting and ventilation, proper orientation, reduced operating hours (where possible) and intelligent control systems. Having said that, this paper comprises case studies that did not have signifi cant components of recycled or reused materials. Clearly where this is achieved, initial embodied energy can be markedly reduced since the upstream energy impacts would have been previously counted on another project (i.e. double counting is not valid). The cost of recycled and reused materials varies widely (for example, refer Treloar et al., 2003). In some cases the cost can be higher than comparable new materials, and in many other cases the cost can be lower. The introduction of signifi cant amounts of recycled and reused materials may upset the relationships found in this paper. This issue also has implications for energy unit rates. In this paper 73.3% of projects represent new construction, and the remainder represent redevelopment projects. In many cases, redevelopment implies reuse of structure and other elements, so both embodied energy and capital cost are reduced. There is no discernible difference between the results for new and redeveloped projects, indicating that the inherent energy-cost relationship still appears to apply. Current preoccupation with operating energy performance in Australia should be tempered with the understanding that embodied energy is signifi cant, particularly in the context of a building’s economic life, which is much less than the one-hundred- year time horizon adopted in this paper. Energy rating schemes should take embodied energy into account, simply by taking cost into account. Where the cost is high it is assumed the embodied energy is also. High performance in energy will then be dictated, at least in part, by the scale of the development. Large houses, for example, would be less likely to achieve high-energy performance than small houses, even though on a per-square-metre-basis the former are more effi cient. This is a challenge for government authorities and policy makers. Table 4: Case study energy residual summary Building ID Floor Area (m2 GFA) Total Project Energy Residual (energy v GFA) Embodied Energy Residual (energy v GFA) Operating Energy Residual (energy v GFA) 1 1,409 42,542 -10,763 53,305 2 450 82,437 -10,662 93,099 3 1,791 15,902 -7,704 23,606 4 2,543 -167,582 2,400 -169,983 5 528 20,023 -12,079 32,102 6 6,761 10,668 20,269 -9,601 7 328 62,258 -11,221 73,479 8 1,223 -42,899 -12,757 -30,142 9 3,278 126,832 6,730 120,102 10 3,760 119,224 16,669 102,555 11 249 49,003 -10,622 59,626 12 625 19,554 -7,689 27,243 13 2,696 -235,624 -11,236 -224,388 14 2,790 176,623 -326 176,950 15 5,677 174,622 19,915 154,707 16 378 70,110 -9,928 80,038 17 652 95,857 -13,466 109,323 18 5,412 -733,922 -3,657 -730,265 19 4,281 169,177 22,934 146,243 20 787 1,518 -7,980 9,498 21 1,159 86,414 -4,993 91,407 22 12,930 693,162 41,671 651,491 23 18,821 41,787 -90,821 132,608 24 10,565 -721,508 60,674 -782,183 25 4,704 -299,000 24,698 -323,698 26 1,345 72,178 -11,368 83,547 27 5,940 110,107 30,636 79,471 28 2,502 -146,454 31 -146,485 29 5,223 47,731 -10,132 57,864 30 3,649 59,259 778 58,481 sum 0 0 0 Source: Microsoft Excel Data Analysis Module The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 16 Table 5: Case study cost residual summary Figure 4: Total project energy v cost residual correlation Figure 5: Embodied energy v capital cost residual correlation Figure 6: Operating energy v operating cost residual correlation Building ID Floor Area (m2 GFA) Total Project Cost Residual (cost v GFA) Capital Cost Residual (cost v GFA) Operating Cost Residual (cost v GFA) 1 1,409 -9,518,138 -872,695 -8,645,443 2 450 -493,883 -49,275 -444,608 3 1,791 -11,314,548 -1,035,967 -10,278,581 4 2,543 -2,257,512 -653,669 -1,603,843 5 528 -3,754,377 -628,429 -3,125,948 6 6,761 8,263,047 1,697,239 6,565,807 7 328 422,964 -338,359 761,323 8 1,223 -5,893,653 -1,378,904 -4,514,749 9 3,278 -1,343,587 1,946,215 -3,289,802 10 3,760 750,046 2,883,499 -2,133,453 11 249 -524,859 -120,265 -404,594 12 625 -1,269,675 -16,133 -1,253,542 13 2,696 -11,053,979 -3,252,478 -7,801,502 14 2,790 17,543,422 250,413 17,293,009 15 5,677 18,764,069 1,053,236 17,710,833 16 378 2,429,078 123,610 2,305,468 17 652 3,781,111 -474,917 4,256,027 18 5,412 -57,050,080 -8,967,066 -48,083,013 19 4,281 11,008,144 1,024,287 9,983,857 20 787 -37,609 -363,800 326,192 21 1,159 9,960,442 -144,994 10,105,436 22 12,930 44,429,824 8,443,003 35,986,821 23 18,821 -24,294,944 -10,122,940 -14,172,004 24 10,565 -12,467,150 7,825,548 -20,292,697 25 4,704 7,143,334 3,440,533 3,702,802 26 1,345 2,389,482 -553,181 2,942,662 27 5,940 7,326,774 3,508,920 3,817,854 28 2,502 -7,100,107 -1,066,509 -6,033,598 29 5,223 10,569,645 -1,188,302 11,757,947 30 3,649 3,592,718 -968,621 4,561,339 sum 0 0 0 Source: Microsoft Excel Data Analysis Module The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 17 Sustainable development is a balance between progress and conservation, and most popularly defi ned as meeting the needs of the present generation without disadvantaging the ability of future generations to meet their own needs. But it can never be achieved. It is more likely to be approached by building less than building more. It is also important to maximise building life through adaptive reuse and recycling. Initial embodied energy, just like capital cost, is an indicator of project scope, and can be used therefore to gauge the impact of developments on the natural environment. The lack of public knowledge about embodied energy has led to its absence from design decision-making and government policies. This situation has arisen as a direct result of the complexity of the calculations and the lack of reliable data upon which to base them. Over time, embodied energy models have become even more complex. This translates into specialist knowledge requirements and a considerable time commitment. The way forward is to develop models that practitioners can use and apply routinely to their projects. Only by putting useful tools into the hands of building designers and managers can a wider understanding of the importance of embodied energy be gained. This also enables life cycle energy and GGE calculation to be incorporated in the early design stages to reduce unnecessary environmental degradation. CONCLUSION The research objective was to discover the nature of the energy- cost relationship and other related heuristic rules by performing a thorough statistical analysis of the created dataset. This analysis included both initial and recurrent energy and cost considerations. Through a series of regression analyses, energy and cost are shown to have a high correlation, to the extent where an inherent relationship can be confi dently claimed. While a dependency between energy and area, or between cost and area, has been shown to exist, after elimination of GFA using energy and cost residuals a strong correlation remains. The r2 value for total project energy and total project cost is 0.5959. The r2 value for embodied energy and capital cost is 0.7081. The r2 value for operating energy and operating cost is 0.6264. The overall relationship between embodied energy and capital cost can lead to the development of predictive models for energy based on cost. For example, embodied energy is given by the regression line y=0.0071x, where energy (y) is expressed in GJ and cost (x) is expressed in fourth quarter 2006 dollars. In this case it is therefore possible, by calculating construction cost (excluding contingencies, professional fees, land acquisition costs and goods and services tax), to derive embodied energy (including direct and indirect energy fl ows) at the project level. But it should be remembered that this research was based on thirty ‘ordinary’ buildings. They had no special signifi cance in terms of environmental performance or effi ciency. They are representative of probably 90% of the existing commercial building stock in our cities. The challenge from this research, therefore, lies in trying to break the inherent energy-cost relationship by fi nding ways to minimise energy for new development even though the cost may have to rise. The heuristics promoted in this research apply where we make no effort to introduce improvements through innovation and better environmental design but rather follow traditional paradigms. Our ability to redefi ne the energy-cost relationship will ultimately be our success in realising more sustainable development. In the developed world, where wants are routinely placed ahead of needs, the goal of sustainable development seems forlorn. The marketplace will to some extent limit the use of energy as long as energy is appropriately priced. While it might be argued that development is important to meet present needs, this must be tempered with a regard for future generations and the legacy we leave them. ACKNOWLEDGEMENTS The authors of this paper would like to thank the kind assistance of the late Dr Graham Treloar for the use of his input-output- based hybrid method for embodied energy analyses and for his supervisory advice and guidance in this research, Computerelation Australia Pty Limited for making available LIFECOST™ software for operating cost analyses, and Davis Langdon Australia for permission to use their data in the case studies. Without their cooperation this research would not have been possible. Finally, the expert input from Dr John Boland (University of South Australia) and Professor Gerry Quinn (Deakin University) in relation to residual regression analysis is gratefully appreciated and acknowledged. REFERENCES Arnold F. (1993) ‘Life cycle doesn’t work’, The Environmental Forum, September/October, 19-23. ASEC (2001) Australia 2001: State of the environment, Independent Report to the Commonwealth Minister for the Environment and Heritage, Australian State of the Environment Committee. Ashworth A. (1988) ‘Making life cycle costing work’, Chartered Quantity Surveyor, April, 17-18. Beggs C. (2002) Energy: Management, supply and conservation, Butterworth-Heinemann. Brown R.J. and Yanuck R.R. (1985) Introduction to life lycle costing, Prentice Hall. Bull J.W. (ed.) (1992) Life cycle costing for construction, Spon Press. Costanza R. (1980) ‘Embodied energy and economic valuation’, Science, 210, 12 December, 1219-1224. Costanza R. (1984) ‘Embodied energy and economic value in the United States economy: 1963, 1967 and 1972’, Resources and Energy, 6, 129-163. Crawford R.H. (2004) ‘Using input-output data in life cycle inventory analysis’, Doctor of Philosophy Thesis, Deakin University, Geelong, Australia. Dell’Isola A.J. and Kirk S.J. (1995) Life cycle cost data, McGraw- Hill. Ding G. (2004) ‘The development of a multi-criteria approach for the measurement of sustainable performance for built projects and facilities’, Doctor of Philosophy Thesis, University of Technology, Sydney, Australia. Evans R., Haste N., Jones A. and Haryott R. (1998) The long term costs of owning and using buildings, London: Royal Academy of Engineering. The Australasian Journal of Construction Economics and Building [Vol 9, No 1] 18 Feagin J., Orum A. and Sjoberg G. eds. (1991) A case for case study, University of North Carolina Press. Flanagan R. and Norman G. (1983) Life cycle costing for construction, Surveyors Publications. Fuller R. (1982), ‘Life cycle costing: A comprehensive approach’, in Building cost techniques: New directions, P.S. Brandon ed., E. & F.N. Spon, 439-446. Irurah D.K. and Holm D. (1999) ‘Energy impacts of building construction as applied to South Africa’, Construction Management and Economics, 17, 363-374. Kirk S.J. and Dell’Isola A.J. (1995) Life cycle costing for design professionals, McGraw-Hill. Langston C.A. (1991) The measurement of life-costs, Sydney: Public Works Department of New South Wales. Langston C.A. (2005) Life-cost approach to building evaluation, Elsevier. Langston Y.L. (2006) ‘Embodied energy modelling of individual buildings in Melbourne: The inherent energy-cost relationship’, Doctor of Philosophy thesis, Deakin University, Geelong, Australia. Langston Y.L and Langston C.A. (2007) ‘Building energy and cost performance: An analysis of thirty Melbourne case studies’, Australasian Journal of Construction Economics and Building, 7(1), 1-18. Langston Y.L. and Langston, C.A. (2008) ‘Reliability of building embodied energy modelling: An analysis of thirty Melbourne case studies’, Construction Management and Economics, 26(2), 147- 160. Langston C.A. and Lauge-Kristensen R. (2002) Strategic management of built facilities, Butterworth-Heinemann. Lavine M.J. and Butler T.J. (1982) Use of embodied energy values to price environmental factors: Examining the embodied energy/dollar relationship, report no.NSF/PRA-82046, Ithaca, New York. Oka T., Suzuki M. and Konnya T. (1993) ‘The estimation of energy consumption and amount pollutants due to the construction of buildings’, Energy and Buildings, 19, 303-311. PCA (2001) Energy guidelines, Brisbane: Property Council of Australia. Robertson G. (1991) ‘Energy effi cient commercial buildings: A realistic market objective’, Architectural Science Review, 34, 139-142. Stone P.A. (1960) ‘The economics of building design’, Journal of the Royal Society, 123(3), 237-273. Tellis W (1997) ‘Application of a case study’, The Qualitative Report, 3(3), available at http://www.nova.edu/ssss/QR/QR3-3/ tellis2.html. Treloar G.J. (1998) ‘A comprehensive embodied energy analysis framework’, Doctor of Philosophy Thesis, Deakin University, Geelong, Australia. Treloar G.J., Gupta H., Love P.E.D. and Nguyen B. (2003) ‘An analysis of factors infl uencing waste minimisation and use of recycled materials for the construction of residential buildings’, Management of Environmental Quality, 14(1), 134-145. Treloar G.J., Ilozor B.D. and Crawford R.H. (2002) ‘Modelling energy use and greenhouse gas emissions associated with commercial building construction’, in proceedings of 36th Annual Conference of the Australian and New Zealand Architectural Science Association (ANZAScA): The modern practice of architectural ccience from pedagogy to andragogy?, Geelong, Victoria, Australia, November, 533-540. van Pelt M.J.F., Kuyvenhoven A. and Nijkamp P. (1990) ‘Project appraisal and sustainability: Methodological challenges’, Project Appraisal, 5, 139-158. Verbeek I. and Wibberley L. (1996) ‘Data collection’, in proceedings of 1st National Conference on LCA, 29 September-1 March.