Toward the ecological footprint of the use and maintenance phase ofbuildings: utility consumption and cleaning tasks

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ Esta es la versión aceptada del artículo publicado en: This is an accepted manuscript of a paper published in: Ecological Indicators (2016): 10/10/2024 DOI: http://dx.doi.org/10.1016/j.ecolind.2016.04.007 Copyright: © 2016 Elsevier Ltd. All rights reserved. El acceso a la versión publicada del artículo puede requerir la suscripción de la revista. Access to the published version may require subscription. “This is an Accepted Manuscript of an article published by Elsevier in [Ecological Indicators] on [2016], available at: http://dx.doi.org/10.1016/j.ecolind.2016.04.007 ” 1 Towards the Ecological Footprint of the use and 1 maintenance phase of buildings: utility consumption 2 and cleaning tasks 3 Martínez-Rocamora, Alejandro (1,*); Solís-Guzmán, Jaime (1); Marrero, Madelyn (1) 4 (1) ArDiTec Research Group. Department of Architectural Constructions II, Higher 5 Technical School of Building Engineering, University of Seville. Av. Reina Mercedes, 4- 6 a. 41012 – Seville, Spain 7 (*) Corresponding author Alejandro Martínez Rocamora: ArDiTec Research Group. 8 Department of Architectural Constructions II, Higher Technical School of Building 9 Engineering, University of Seville. Av. Reina Mercedes, 4-a. 41012 – Seville, Spain; 10 Tel.: +34 675041010; email: [email protected] 11 12 Abstract 13 Due to poor design of buildings in terms of maintenance, there are a number of buildings 14 today that remain extremely expensive to maintain, both economically and 15 environmentally. In order to mitigate these overheads, the development of a cost 16 database is needed with which the resources required to clean and maintain buildings 17 can be estimated. This paper presents a methodology to estimate these costs and the 18 environmental impact, in terms of Ecological Footprint (EF), associated to the utility 19 consumption and to the cleaning tasks necessary during the service life of buildings. 20 Given the numerous peculiarities identified for this type of activity compared to the 21 construction of buildings, it is necessary to define a new methodology of calculation, with 22 its own assumptions and formulae. This methodology is then applied to the case of a 23 college hall of residence that houses up to 139 residents. The results show that the 24 2 annual EF of cleaning tasks accounts for 11.42% of the EF of utility consumption. 25 Together they total 67.334 global hectares per year (gha/yr), 88% of which corresponds 26 to the carbon footprint. Within the EF of cleaning, about 71% is due to food consumed 27 by labor, while 26% is due to the manufacture of cleaning products and tools, which are 28 equally divided among the six categories of productive land. The development of this 29 methodology is essential for the detailed quantification of the environmental impact of 30 utility consumption and cleaning tasks that occur during the service life of buildings. The 31 use of discount rates on results is included in terms of the EF of a baseline year, as an 32 equivalent to the discount rate in economic terms. 33 Keywords 34 Ecological footprint, maintenance and cleaning of buildings, utility consumption. 35 36 3 1. Introduction 37 The crisis in the construction sector in Spain in recent years has forced a new approach 38 to business models, since the highest priority today is not the construction of new 39 buildings, but the maintenance and renovation of existing buildings, as predicted by 40 Kohler and Hassler (2002). However, during the earlier 'housing bubble' stage, the 41 frenetic activity of construction companies failed to allow time to be taken to consider the 42 maintenance cost that the buildings would impose. Today, cities have several buildings 43 that exemplify absurd maintenance costs mainly due to a thoughtless initial design in this 44 aspect, such as the City of Sciences in Valencia, the Bankia Tower in Madrid, and the 45 Sevilla Tower in Seville. 46 Provoked by this turn of events, the role of the Facility Manager (FM) has become of 47 major importance, especially in tertiary buildings, such as offices, shopping malls, hotels 48 and hospitals (Heng et al., 2005). The FM is responsible for the control and coordination 49 of various tasks, which include the proper functioning of the installations, the 50 maintenance, and the regular cleaning of the building. For the economic management 51 and control of these tasks, the FM usually employs internal calculations as a product of 52 experience, or maintenance and operating cost databases, which require the acquisition 53 of a license that is not freely available to all buyers: these include BCIS On-line (Royal 54 Institute of Chartered Surveyors, 2015), BLCC (National Institute of Standards and 55 Technology, 2013), ECONPACK (U.S. Army Corps of Engineers, 2014), CostLab (CBRE 56 and Whitestone Research, 2015), and CYPE’s decennial maintenance module (CYPE 57 Ingenieros, 2015). The lack of open-access databases imposes additional obstacles to 58 the estimation of these costs from the design phase of buildings. Advance awareness of 59 this information would be invaluable to both the owner and the manager of the building, 60 and would provide a breakthrough in the study of Life-Cycle Costs (LCC) of buildings 61 (ISO 15686-5, 2008). 62 4 In recent years, a number of research studies have been presented on the application of 63 environmental indicators, such as Ecological Footprint and Carbon Footprint, to 64 construction processes (González-Vallejo et al., 2015a; Solís-Guzmán et al., 2013), 65 whereby this analysis is extended to other life-cycle phases of the building, such as the 66 transformation of land use (urbanization previous to construction) (Freire and Marrero, 67 2014), and rehabilitation vs. demolition (Alba Rodríguez et al., 2015), as they are defined 68 in the Life-Cycle Assessment (LCA) methodology for buildings (UNE-EN 15978, 2012). 69 Other existing studies evaluate buildings through ecological indicators, such as the 70 Ecological Footprint (Bastianoni et al., 2007), Carbon Footprint, and Emergy Analysis 71 (Marchi et al., 2015; Martínez-Rocamora et al., 2014; Pulselli et al., 2014, 2007), which 72 all focus on the construction phase, operation, or the implications of adding Vertical 73 Greenery Systems to improve the envelope behavior. 74 These calculation models attempt to predict the environmental impact of the construction 75 process through the project’s bill of quantities. In the bill of quantities, the various tasks 76 are broken down into three key elements: manpower, materials and machinery (Marrero 77 and Ramirez‐De‐Arellano, 2010). Emission or embodied energy factors are then applied 78 to these elements, which are subsequently converted into environmental impacts using 79 the methodology of the corresponding ecological indicator (González-Vallejo et al., 80 2015b). 81 However, the calculation models above cannot be applied when an attempt to tackle the 82 cleaning of buildings is made, since no cost database quantifies the three elements of 83 manpower, materials and machinery in a disaggregated way. These tasks present so 84 many unique features that it remains impossible to apply the same methodology of the 85 construction process. Moreover, it is essential to treat cleaning separately from 86 maintenance because, despite pertaining to the same life-cycle phase, the nature of the 87 activities differs so widely that the assumptions and formulation to be applied cannot be 88 the same. 89 5 Environmental impact studies focused on the use and maintenance phase of buildings 90 generally consider that the impact sources corresponding to this stage can be divided 91 into three distinct branches: utility consumption (electricity, water, and fuel); regular 92 cleaning tasks; and periodic maintenance and renewal of elements (Fuller, 2010). Due 93 to the high complexity of calculating the resources necessary for the maintenance and 94 cleaning of buildings, the vast majority of studies calculate solely the impact of the utility 95 consumption. In certain cases, the renovation of materials is also included, whereby the 96 materials are assigned a specific service life (Adalberth, 1997a, 1997b; Pulselli et al., 97 2007); manpower and machinery, however, are not taken into account. 98 Given the problematic above, this paper focuses on the utility consumption and cleaning 99 tasks. To this end, in the following section, the system boundaries for this study are 100 defined. Secondly, the peculiarities of cleaning tasks are determined and the method to 101 structure these tasks is defined, thereby justifying the need for the separate analysis 102 applied in this model. The methodology for the application of the EF indicator to utility 103 consumption and cleaning tasks is then described. In Section 4, the case study of a 104 college hall of residence is presented for the application of the methodology for economic 105 and environmental costs. Finally, results obtained from this case study are shown and 106 discussed, and a number of conclusions are drawn from the present study. 107 108 2. System boundaries 109 The standard UNE-EN 15978 (2012) establishes, in the first place, that the use phase 110 extends from the end of the construction works until the building is about to be 111 deconstructed/demolished. 112 In order to set the transversal system boundaries, the possible impact drivers during this 113 phase of the building’s life cycle have been studied. These have been classified into 114 three main sectors: industry, building, and the occupants (Figure 1). Impacts from the 115 6 manufacture of furniture, appliances, decoration and other objects are associated to the 116 industrial sector, while the consumables and occupants’ food consumption, mobility and 117 Municipal Solid Waste (MSW) all belong to the occupants’ EF and not to that of the 118 building. All these impact drivers remain outside the system boundaries of the present 119 study. The remaining elements, as can be observed in Figure 1, are attributed to the 120 building, and are divided into those corresponding to its use and those referring to 121 maintenance and cleaning tasks. 122 123 124 Figure 1. System boundaries 125 126 This paper focuses on the utility consumption and cleaning tasks. In the following section, 127 a methodology is proposed to structure the tasks pertaining to this phase of the building’s 128 7 life cycle, and a number of peculiarities of the cleaning tasks are determined. The 129 methodology for the application of the EF indicator to the use and cleaning of buildings 130 is then defined. 131 132 3. Methodology 133 3.1. Cleaning costs singularities 134 The actions carried out during this phase are divided into cleaning, maintenance 135 (predictive, preventative and corrective), and renovation. In order to classify the tasks 136 related to these five types of action, the structure defined in previous advances of the 137 study (Martínez-Rocamora et al., 2015) is used. This structure relies on the Andalusian 138 Construction Costs Database (ACCD) (Andalusia Government, 2014) as a reference of 139 structure and of a classification system of information in the construction sector. 140 Focusing on cleaning tasks, these feature certain peculiarities compared to construction 141 and maintenance of constructive elements. Firstly, there is a multitude of different 142 products and tools with which a surface or piece of furniture can be cleaned. 143 Unfortunately, if single costs for every different combination of products, tools, and 144 surfaces to be cleaned were developed, then an excessive amount of costs would 145 appear for the same task. 146 In order to provide a simplified solution to this issue, the contents of a professional 147 cleaning trolley have been studied (Table 1). For each tool or product its frequency of 148 renewal, cost, weight and main material have been obtained. In the same table, the 149 emission factor for each material, obtained from Ecoinvent, is included, which after an 150 exhaustive study (Martínez-Rocamora et al., 2016) was confirmed as the most suitable 151 Life-Cycle Assessment (LCA) database for use in the construction sector. By using a 152 value of 1515 hours worked per year from the regional collective agreement of the 153 cleaning sector (Andalusia Government, 2013), the hourly cost (0.69 €/h), CO2 emissions 154 8 (0.227 kgCO2/h), and MSW generation (0,057 kg MSW/h) of the cleaning trolley as a 155 whole were obtained (bottom line in Table 1). 156 157 15 3.2.1. Built-up land 218 The EF of the land surface occupied by the building is incurred in the construction phase, 219 although this is a footprint that lasts as a constant value throughout its whole service life, 220 and is not cumulative. This surface is determined in hectares, by taking into account the 221 whole parcel area, and it is transformed into gha through equation 1. For this kind of 222 impact it is assumed that cities are built on land appropriate for the growing of crops 223 (Wackernagel and Rees, 1997), and therefore the equivalence factor of built-up land is 224 the same as in cropland. 225 𝐸𝐹𝑏𝑠 = 𝑆 · 𝐸𝑄𝐹𝑖𝑛𝑓 (1) 226 where 227 EFbs: Partial Ecological Footprint of built-up land (gha/yr) 228 S: Total surface occupied by the building or parcel (wha/yr). 229 EQFinf: Equivalence factor of infrastructure land (2.51 gha/wha) (GFN, 2014) 230 231 3.2.2. Electricity 232 The EF of electricity consumption is determined through two factors: the emission factor 233 of electricity, which depends on the energy mix of the country where the building is 234 located; and the CO2 absorption factor of forests. Thus, equation 2 is comprised of these 235 calculations, and includes the reduction of emissions due to the carbon uptake in oceans, 236 which, according to Borucke et al. (2013), is obtained by dividing the ocean uptake from 237 Khatiwala et al. (2009) by the anthropogenic carbon emissions (both in Pg C yr-1) from 238 Marland et al. (2007). The outcome varies between 28% and 35% over the period 1961- 239 2008, as do the equivalence factors due to variations in the productivities of the six land 240 types. For the purposes of the present study, a cautious value of 28% is used for the 241 absorption of CO2 in oceans. 242 16 𝐸𝐹𝑒𝑙 = 𝐶𝑒𝑙 · 𝐸𝑒𝑙 · (1 − 𝐴𝑜𝑐) 𝐴𝑓 ⁄· 𝐸𝑄𝐹 𝑐𝑎 (2) 243 where 244 EFel: Partial Ecological Footprint of electricity consumption (gha/yr) 245 Cel: Electricity consumption per year (kWh/yr) 246 Eel: Emission factor of electricity (0.000248 tCO2/kWh) (REE, 2014) 247 Aoc: Percentage of CO2 absorption in oceans (Borucke et al., 2013; GFN, 2014) 248 Af: Absorption factor of forests (3.59 tCO2/wha) (GFN, 2014) 249 EQFca: Equivalence factor of carbon absorption land (1.26 gha/wha) (GFN, 2014) 250 251 3.2.3. Fuel 252 The impact of fuel consumption for the heating of spaces or water is calculated in the 253 same way as electricity (equation 2), except for the substitution of the consumption and 254 the emission factor with those corresponding to each fuel used, being the latter 0.0028 255 tCO2/l of C Diesel and 0.0015 tCO2/l of propane, among others (IDAE, 2011). 256 In the case of agricultural biomass or forest industry, which in this table have a neutral 257 emission factor, their impact is transformed into forest land EF through equation 3. The 258 low values used for the yield factor of wood fuel as well as those for wood materials by 259 the GFN methodology lead to a high EF of forest land, and is high even when compared 260 to that of the carbon footprint, which usually controls the EF of buildings. 261 𝐸𝐹𝑤𝑓 = 𝐶𝑤𝑓/ 𝑌 𝑤𝑓 · 𝐸𝑄𝐹 𝑓𝑜 (3) 262 where 263 EFwf: Partial Ecological Footprint of wood fuel (gha/yr) 264 Cwf: Consumption of wood fuel per year (m3/yr) 265 17 Ywf: Yield of wood fuel (3.35 m3/wha) (GFN, 2014) 266 EQFfo: Equivalence factor of forest land (1.26 gha/wha) (GFN, 2014) 267 268 3.2.4. Water 269 The EF of water consumption was assigned in previous models (González-Vallejo et al., 270 2015a; Solís-Guzmán et al., 2013) to the forest land category, whereby it was considered 271 the forest’s capacity to produce water (Domenech Quesada, 2007). However, this 272 assumption is not recognized in the official indicator’s methodology (Borucke et al., 273 2013), in which the forest land category is specifically dedicated to the impacts generated 274 by wood fuel and products. One possible solution to this issue would be the addition of 275 a new productive area of water, where a cautious percentage of the total production 276 could be reserved for human consumption. 277 In this model water consumption is computed as the energy needed to bring it to the 278 point of consumption, including treatments both previous and subsequent to its use, as 279 specified in the standard UNE-EN 15978 (2012). It is assumed that the only energy 280 needed in the whole process is that of electricity. 281 𝐸𝐹 𝑤𝑎 = 𝐶𝑤𝑎 ·𝐸𝐼𝑤𝑎 · 𝐸𝑒𝑙 · (1 − 𝐴𝑜𝑐) 𝐴𝑓 ⁄· 𝐸𝑄𝐹 𝑐𝑎 (4) 282 where 283 EFwa: Partial Ecological Footprint of water consumption (gha/yr) 284 Cwa: Water consumption per year (m3/yr) 285 EIwa: Energy intensity of drinking water (0.44 kWh/m3) (EMASESA, 2005) 286 Eel: Emission factor of electricity (0.000248 tCO2/kWh) (REE, 2014) 287 288 18 289 3.2.5. Materials 290 The second branch established in the methodology flowchart (Figure 3) starts with the 291 consumption of materials. As mentioned above, for this branch, the data comes from the 292 project’s bill of quantities. The consumption of materials implies an environmental impact 293 due to three different aspects: manufacturing, transport and waste. In the case of the 294 present study (the assessment of use and of cleaning tasks) the materials employed are 295 those contained in the cleaning trolley (Table 1) or in the domestic cleaning kit. In contrast 296 to that in the construction phase, the use of these two sets of materials designed for 297 cleaning tasks is accounted for in hours. The calculation methodology (equation 5) 298 therefore uses the values shown in Table 1 for CO2 emissions, weight of materials and 299 weight of MSW per hour as auxiliary data. 300 𝐸𝐹 𝑚𝑎 =∑(𝐻𝑚𝑎𝑖· 𝐸𝑚𝑎𝑖) · (1 − 𝐴𝑜𝑐)/𝐴𝑓· 𝐸𝑄𝐹 𝑐𝑎 (5) 301 where 302 EFma: Partial Ecological Footprint of consumption of materials (gha/yr) 303 Hmai: Hours of use of material i per year (h/yr) 304 Emai: Emission factor of material i (tCO2/h) 305 i: cleaning trolley or domestic cleaning kit 306 307 Wooden materials must also use equation 6 since these cause an EF of forest land. 308 Nevertheless, it is important to point out that although two land categories are used for 309 emissions and wood, since most of the CO2 uptake occurs in forests, the CO2 absorption 310 land is considered a subcategory of forest land (Borucke et al., 2013). Further discussion 311 19 on this issue has been provided by several authors (Kitzes et al., 2008; Lenzen and 312 Murray, 2001; Van Den Bergh and Verbruggen, 1999). 313 𝐸𝐹 𝑤𝑜 =∑(𝐶𝑤𝑜𝑖/𝑌 𝑤𝑜𝑖) · 𝐸𝑄𝐹 𝑓𝑜 (6) 314 where 315 EFwo: Partial Ecological Footprint of wooden materials (gha/yr) 316 Cwoi: Consumption of wooden material i per year (t or m3/yr) 317 Ywoi: Yield of wooden material i (t or m3/wha) (GFN, 2014) 318 319 Additionally, the environmental impact of the transport of cleaning products from the 320 factory to the building must also be taken into account, since LCA databases use a 321 cradle-to-gate assessment (i.e. from the extraction of raw materials until the final product 322 is ready for delivery) which does not include this process. For the calculation of the EF 323 of transport (equation 7) the following assumptions are made: 324  A single purchase of cleaning products is made per year 325  Average distance from factory to the building (back and forth): 30 km 326  Transport vehicle: 327 o Maximum load: 2.1 t 328 o Fuel consumption: 7.9 l/100 km 329 o Type of fuel: Diesel 330  Emission factor of fuel: 0.0022 tCO2/l of petrol or 0.0026 tCO2/l of diesel (IDAE, 331 2011) 332 𝐸𝐹𝑡𝑟 =∑(𝐻𝑚𝑎𝑖·𝑊𝑚𝑎𝑖 𝑇𝑐𝑎𝑝 · 𝐷𝑚𝑎) · 𝑇𝑐𝑜𝑛 · 𝐸𝑓· (1 − 𝐴𝑜𝑐)/𝐴𝑓· 𝐸𝑄𝐹 𝑐𝑎 (7) 333 where 334 EFtr: Partial Ecological Footprint of the transport of materials (gha/yr) 335 20 Hmai: Hours of use of material i per year (h/yr) 336 Wmai: Annual Weight of material i (t/yr) 337 i: cleaning trolley or domestic cleaning kit 338 Tcap: Truck capacity (t) 339 Dma: Average distance (km) 340 Tcon: Truck consumption (l/100 km) 341 Ef: Emission factor of fuel (tCO2/l) 342 343 In contrast to the construction phase, waste generated from products contained in the 344 cleaning trolley or the domestic cleaning kit are considered as MSW. According to Almasi 345 and Milios (2013), the net emission factor in Spain, taking the benefits of recycling and 346 energy recovery into account, is 0.244 tCO2 per ton of MSW. The hourly MSW generation 347 from cleaning products and tools are obtained from Table 1, assuming that the complete 348 weight of a tool becomes waste when its service life comes to an end, while for cleaning 349 products, only the bottle containing it becomes waste when it is empty. 350 𝐸𝐹𝑀𝑆𝑊 = 𝐻𝑚𝑎𝑖· 𝐺𝑚𝑎𝑖· 𝐸𝑀𝑆𝑊 · (1 − 𝐴𝑜𝑐) 𝐴𝑓 ⁄· 𝐸𝑄𝐹 𝑐𝑎 (8) 351 where 352 EFMSW: Partial Ecological Footprint of MSW from materials (gha/yr) 353 Hmai: Hours of use of material i per year (h/yr) 354 Gmai: Hourly MSW generation of material i (t/h) 355 i: cleaning trolley or domestic cleaning kit 356 EMSW: Emission factor of MSW (0.244 tCO2/t) (Almasi and Milios, 2013) 357 358 21 3.2.6. Machinery 359 Machinery, accounted for in hours of utilization, is always electric in the case of cleaning 360 tasks. It is therefore necessary to transform the hours of use into electricity consumption 361 through its capacity (Freire and Marrero, 2014). Once the energy consumption is 362 obtained, the procedure explained in subsection 3.2.2 must be applied in order to obtain 363 the EF of machinery. 364 365 3.2.7. Manpower 366 Manpower is too significant in a project’s bill of quantities to ignore the impact it causes. 367 It is therefore considered here that two impact sources are involved in manpower: food 368 and MSW. Although the mobility of workers was also taken into account in previous 369 calculation models (González-Vallejo et al., 2015a; Solís-Guzmán et al., 2013), the 370 standard UNE-EN 15978 (2012) establishes that, for the construction phase, transport 371 of workers back and forth to the worksite must not be included, and therefore neither 372 must this mobility be included in the case of workers of any company. 373 In order to determine the EF of food, its consumption by workers that occurs during 374 working hours (breakfast and lunch) is considered to cause an impact attributable to the 375 building, as was applied by Domenech Quesada (2007). Moreover, it could be assumed 376 that food provides the energy needed to do the work involved, thereby becoming the 377 ‘fuel’ of workers. 378 According to Grunewald et al. (2015), the EF of food in Spain is approximately 1.15 379 gha/person/yr. Its broken-down composition into the six EF categories is presented in 380 Table 2 following the proportions for a Spanish resident. The equivalence factors are 381 implicit in the EF values for each category, and are expressed in gha/person/yr. 382 383 22 Table 2. Ecological Footprint of food in Spain 384 Category i Percentage EFi (gha/person/yr) Cropland 49.65 % 0.571 Grazing land 11.00 % 0.127 Forest land 1.00 % 0.012 Fishing grounds 29.00 % 0.334 Built-up land 0.35 % 0.004 Carbon 9.00 % 0.104 TOTAL 100.00 % 1.150 385 386 The EF of food consumed by workers is determined through equation 9 and it is taken 387 into account that, according to a recent study on the daily distribution of nutrients in the 388 standard Spanish diet (Moreno Rojas et al., 2015), breakfast and lunch represent 15% 389 and 46% respectively of the daily diet of an adult (> 25 years old). The total number of 390 hours worked is obtained from the bill of quantities, and eight hours is assumed to be the 391 standard duration of a working day. 392 𝐸𝐹 𝑓𝑜𝑖=𝐻𝑤 𝐻𝑑·𝐹𝐼𝑓·𝐸𝐹𝑖 365 (9) 393 where 394 EFfoi: Partial Ecological Footprint of food consumption in EF category i (gha/yr) 395 Hw: Total number of hours worked per year (h/year) 396 Hd: Number of hours worked per day (8 h/day) 397 FIf: Food intake fraction for breakfast and lunch of a Spanish adult (61%) (Moreno 398 Rojas et al., 2015) 399 23 EFi: Footprint of food consumption in EF category i (gha/person) (Table 2) 400 365: Days in a year 401 402 Regarding MSW generation by manpower, in 2013 each person in Spain generated 449 403 kilograms of MSW (EUROSTAT, 2015). If it is assumed that each day a person sleeps 404 for 8 hours (not generating any waste), then a MSW quantity of 0.077 kg/h is obtained. 405 By considering the previously cited emission factor of 0.244 tCO2 per ton of MSW, the 406 calculation of the EF of MSW from manpower can be made using equation 8, being in 407 this case Hw the total number of hours worked per year, and Gw the hourly waste 408 generation. 409 410 3.3. EF assessment in long-term scenarios 411 With the calculation methodology defined, the last remaining detail consists of the 412 ecological update of the results obtained. As pointed out by Riahi et al. (2007), the 413 longevity of buildings forces us to face the challenge of predicting the changes in the 414 quantity of greenhouse gas emissions, the advances in energy-generating technology, 415 and the rise of new policies to mitigate climate change. 416 In a recent study, Moore et al. (2012) perform a projection of the EF until 2050 by 417 following several scenarios. In order to choose a conservative (while optimistic) scenario 418 for this study, the BLUE map scenario from the International Energy Agency is used, 419 wherein it is assumed that, by 2050, emissions will be stabilized at 50% of the levels 420 from 2005. Food consumption, population growth and climate change follow hypothetical 421 projections from international organizations such as the Food and Agriculture 422 Organization (FAO), United Nations Population Division (UNPD), and the 423 Intergovernmental Panel on Climate Change (IPCC). 424 24 Reduction rates are extracted from this scenario for each of the six categories of EF by 425 setting the current year as baseline year, which enables impacts caused in other years 426 to be translated into terms of current EF (equation 10). These conversion factors reflect, 427 apart from the aforementioned assumptions, the variations throughout the history of the 428 productivities and equivalence factors used by the Global Footprint Network (Moore et 429 al., 2012). 430 𝐸𝐹𝑖=𝐸𝐹𝑖,𝑏 ·∑𝐸𝐹𝑤,𝑖,𝑗 𝐸𝐹𝑤,𝑖,𝑏 𝑗 (10) 431 where 432 EFi: Total EF of land type i in the building’s service life (gha) 433 EFi,b: EF of land type i for the baseline year b (gha) 434 EFw,i,j: World’s EF of land type i for year j (gha) 435 EFw,i,b: World’s EF of land type i for the baseline year b (gha) 436 437 In the next section, the case study of a building is presented in order to test the feasibility 438 and utility of the designed methodology, followed by the presentation and discussion of 439 the results. 440 441 4. Case study 442 The building under study, located in the south sector of Seville (Spain), holds the 443 Hernando Colón college hall of residence. It was built in 1948 on the site of the 444 Conqueror’s Square from the Ibero-American exposition of 1929. For the purposes of 445 this study, a service life of 100 years is assumed from the construction date. It is a 446 detached building with 7,187 m2 of built-up surface in a 10,501 m2 parcel with garden 447 spaces, parking lots, sports facilities and a swimming pool (Figure 4). 448 31 512 As can be observed in Figure 6 (right-hand-side), the food consumption by workers 513 represents more than 2/3 of the EF of cleaning tasks, which renders the distribution 514 among the six EF categories as more uniform (Figure 6, left-hand-side). The second 515 main impact driver is the manufacturing of materials, with almost 26%. Finally, MSW and 516 transport of materials remain insignificant in comparison to the other factors. 517 518 Table 5. Annual Ecological Footprint of Use and Cleaning 519 IMPACT DRIVER CROPLAND PASTURE LAND FISHING GROUNDS FOREST LAND CARBON FOOTPRINT BUILT SURFACE TOTAL (gha/year) (gha/yr) USE 57.800 2.636 60.436 CLEANING 2.454 0.544 1.433 0.741 1.701 0.017 6.890 TOTAL 2.454 0.544 1.433 0.741 59.501 2.653 67.326 520 Total results (Table 5) show that the EF of cleaning represents around 11.42% of that of 521 the ‘use’. These results justify the necessity of a detailed study of cleaning tasks, as was 522 performed in the present study, since cleaning tasks trigger a relatively major impact, 523 especially in non-residential buildings. 524 Assuming 139 residents, the EF per person due to the use and cleaning of the building 525 is 0.484 gha/yr/person, approximately 12.4% of the EF per capita for Spain in 2011 526 (GFN, 2016). On the other hand, the EF of cleaning in the director’s residence is 0.074 527 gha/yr. Assuming a total of 4 members of the director’s family live in this residence, the 528 EF per family member is 0.0185 gha/yr/person. This change in the order of magnitude is 529 due to the lower periodicity of cleaning tasks in domestic environments and to the 530 absence of a task coordinator. 531 32 Considering the distribution by productive land type (Figure 7), the carbon footprint 532 controls the final results due to the utility consumption. This dominion by the carbon 533 footprint frequently leads to discussion about the accounting methodology of the EF 534 indicator, as mentioned before. Nonetheless, the considerable percentage represented 535 by cleaning tasks and thus by manpower provides better equilibrium among the six land 536 categories. 537 538 539 Figure 7. Annual EF of Use and Cleaning by land type 540 541 Based on the long-term scenario explained in subsection 3.3, update rates have been 542 applied to the EF of the six land categories. These factors are determined from 1948 (the 543 year of construction of the building) to 2047 (100 years of service life are assumed), 544 taking 2015 as the baseline year in equation 10. The total EF result for these 100 years 545 according to the BLUE map scenario (Figure 8) becomes 4,226.472 gha, with 87.46% 546 due to the ‘use’ branch and 12.54% caused by cleaning tasks. It must be mentioned that 547 when the whole service life is observed, the EF of built-up land is reduced from 4 to 548 0.06%, since while the rest of the EFs increase, this EF remains constant. 549 33 550 551 Figure 8. BLUE map scenario applied to the case study for 100 years of service life 552 553 Although the methodology has suffered substantial changes, as has been stated 554 throughout this paper, results for this phase have been compared to those for other 555 phases of the building’s life cycle. The EF of utility consumption and cleaning tasks for 556 this building is 0.588 gha/m2, which, compared to typology 5 in González-Vallejo et al. 557 (2015b), is 2.48 times that of the construction phase. When comparing the EF ratio per 558 parcel area (0.402 gha/m2) to that from Bastianoni et al. (2007) for the construction 559 phase of a 4-story apartment building (0.006 gha/m2), the EF of the use and cleaning 560 tasks is almost 35 times greater. This considerable variation of proportions is assumed 561 to be due to differences in the methodology used to apply the EF indicator to the building 562 sector, since this still constitutes a new approach in constant evolution. 563 In order to illustrate the sensitivity of the model, 35 different scenarios have been studied 564 and sorted from the lowest to the highest EF (Figure 9). These scenarios are based on 565 34 the possible distributions of the total area dedicated to bedrooms, en-suite and shared 566 bathrooms, and shower rooms (~1840 m2) among four different types of room: single (11 567 m2), single en-suite (13.5 m2), twin (15 m2), and twin en-suite (18 m2). One shared 568 bathroom and one shared shower room (22.5 m2 each) is assumed necessary for each 569 20 and 40 bedrooms without private bathrooms, respectively. Variations of the area 570 dedicated to each of these four types of room have been studied in intervals of 25%. 571 572 573 Figure 9. Sensitivity analysis of the EF of 35 cleaning scenarios 574 575 The results of this analysis show that the EF of cleaning tasks is certainly sensitive to 576 changes in the distribution of spaces. Distributions with only en-suite rooms showed 577 lower EF results. In contrast, those with a majority of rooms with shared bathroom show 578 a larger EF. This is due to the higher cleaning frequency of shared bathrooms and 579 shower rooms compared to private bathrooms. Moreover, an inflexion zone is produced 580 where the number of shared bathrooms and shower rooms increase at the same time 581 35 due to the assumptions made above. Those cases remain thus unclassifiable, since they 582 are not representative of the set of scenarios taken. 583 584 6. Conclusions 585 In this paper, a structure is first defined for the classification of cleaning tasks carried out 586 in a building and for the quantification of the resources needed, broken down into 587 materials, manpower, and machinery, which simplifies its implantation and 588 understanding within the building sector. The method of organization is proven robust 589 and user-friendly thanks to its highly logical approach, which allows users of any profile 590 to develop a detailed economic and environmental budget. 591 Secondly, a methodology to determine the EF of utility consumption and cleaning tasks 592 in a building is developed. The methodology presented in previous papers for the 593 construction phase is revised, and, given the numerous peculiarities of the tasks herein 594 studied, the formulae to be used must be adjusted accordingly. Despite all these 595 setbacks, the EF methodology of the construction phase is adapted to a new phase of 596 the life cycle of buildings: the use and maintenance phase. The EF indicator shows high 597 flexibility in its adaptation to productive sectors instead of to societies. Nonetheless, it 598 still contains certain flaws, such as the usual dominion by the EF of carbon absorption 599 land, and the accounting methodology for the impact caused by water consumption. 600 Although the present study is only focused on determining the EF of utility consumption 601 and cleaning tasks, the implementation of maintenance and renovation of constructive 602 elements should pose no great difficulty in future advances of this research due to their 603 similitude to the construction process. 604 Since this is a first proposal for modeling the issue under study, it has yet to be applied 605 to a representative set of cases. However, the case studied leads to discussion focused 606 36 on the order of magnitude of results. Once the whole model is defined and improved, it 607 will be validated through its application to numerous cases. 608 In this case study, the impact of cleaning tasks represents 11.42% of that of utility 609 consumption: this is a non-negligible percentage, which, along with those corresponding 610 to maintenance and renovation, must be quantified. Results show that manpower 611 represents approximately 73% of the impact of cleaning tasks. It is extremely important 612 to bear in mind that manpower is a necessary impact. In fact, it remains highly 613 complicated to reduce manpower in cleaning tasks by substitution with autonomous 614 machinery. However, this result provides encouragement for future research into the 615 influence of an easier-to-clean design of buildings upon the economic and environmental 616 costs of cleaning. Also the influence of the untidiness of users on the performance of 617 cleaning tasks remains difficult to be included in the calculation model, since it is only 618 noticeable by laborers on-site. 619 A uniform share across the six categories of EF is identified when cleaning tasks are 620 studied. The indicator is thus proven more useful in the use and maintenance phase of 621 buildings than in the construction phase, in which the carbon footprint clearly dominated 622 the results due to the intense application of materials. 623 The analysis of 35 different scenarios of possible distributions of the space dedicated to 624 bedrooms illustrates the model’s sensitivity to changes. Moreover, the selection of 625 scenarios is oriented towards determining which configurations of the area dedicated to 626 bedrooms allow a reduction in the annual EF of cleaning tasks. It is concluded that, with 627 the assumptions taken, distributions dominated by twin en-suite bedrooms provoke a 628 lower EF of cleaning than the existent distribution in the building. 629 The calculation methodology presented is considered useful for the economic and 630 environmental impact assessment of both residential and tertiary buildings (hotels, office 631 buildings, etc.), especially for the latter, whose results attain highly significant values. 632 37 This methodology can prove itself valuable to Facility Managers and companies in the 633 establishment of a possible marketing strategy based on sustainability. The methodology 634 also presents certain weaknesses, such as the high representation of carbon footprint 635 when utility consumption is considered, and the difficulty in predicting the evolution of 636 the six EF categories in the long term. 637 638 Acknowledgements 639 The authors gratefully acknowledge the University of Seville for funding this research 640 through its 5th Research Program, as well as Mrs. Lesley Burridge for her invaluable help 641 to improve the use of English language in this paper. 642 643 References 644 Adalberth, K., 1997a. Energy use during the life cycle of single-unit dwellings: a 645 method. Build. Environ. 32, 321–329. doi:10.1016/S0360-1323(96)00069-8 646 Adalberth, K., 1997b. Energy use during the life cycle of single-unit dwellings: 647 examples. Build. Environ. 32, 321–329. doi:10.1016/S0360-1323(96)00069-8 648 Alba Rodríguez, M.D., Marrero, M., Solís-Guzmán, J., Llácer Pantión, R., Lucas Ruíz, 649 R., 2015. Ecological Footprint of Building Rehabilitation: Economic and 650 Environmental Assessment, in: 6th International Congress of Energy and 651 Environment Engineering and Management. Paris. 652 Almasi, A.M., Milios, L., 2013. Municipal waste management in Spain. 653 Andalusia Government, 2014. Andalusia Construction Cost Database (ACCD). 654 Andalusia Government, 2013. Convenio Colectivo del Sector de Limpieza de Edificios 655 y Locales de Sevilla y Provincia [WWW Document]. URL 656 http://www.dipusevilla.es/export/bop/201310/25.pdf (accessed 1.6.16). 657 Bastianoni, S., Galli, A., Pulselli, R.M., Niccolucci, V., 2007. Environmental and 658 economic evaluation of natural capital appropriation through building construction: 659 practical case study in the Italian context. Ambio 36, 559–565. doi:10.1579/0044- 660 7447(2007)36{[}559:EAEEON]2.0.CO;2 661 Borucke, M., Moore, D., Cranston, G., Gracey, K., Iha, K., Larson, J., Lazarus, E., 662 Morales, J.C., Wackernagel, M., Galli, A., 2013. Accounting for demand and 663 supply of the biosphere’s regenerative capacity: The National Footprint Accounts' 664 underlying methodology and framework. Ecol. Indic. 24, 518–533. 665 doi:10.1016/j.ecolind.2012.08.005 666 CBRE, Whitestone Research, 2015. CostLab Pro. 667 38 CYPE Ingenieros, 2015. Mantenimiento decenal de CYPE Ingenieros. Calendario de 668 mantenimiento y Valoración de mantenimiento decenal del edificio. 669 Domenech Quesada, J.L., 2007. Huella ecológica y desarrollo sostenible (Ecological 670 Footprint and Sustainable Development). AENOR, Madrid. 671 EMASESA, 2005. Sostenibilidad y gestión. Así éramos, así somos. 1975-2005 / 672 Sustainability and management. How we were, how we are. 1975-2005. 673 EUROSTAT, 2015. Municipal waste generated by country in selected years (kg per 674 capita) [WWW Document]. URL http://ec.europa.eu/eurostat/statistics- 675 explained/index.php/File:Municipal_waste_generated_by_country_in_selected_ye 676 ars_%28kg_per_capita%29_new.png (accessed 8.6.15). 677 Freire, A., Marrero, M., 2014. Analysis of the ecological footprint produced by 678 machinery in construction, in: World Sustainable Building 14. Barcelona. 679 Fuller, S., 2010. Life-Cycle Cost Analysis (LCCA). 680 GFN, 2016. EF per capita of Spain [WWW Document]. URL 681 http://www.footprintnetwork.org/en/index.php/GFN/page/trends/spain/ (accessed 682 1.12.16). 683 GFN, 2014. Learning Package of National Footprint Accounts 2014 edition. 684 González-Vallejo, P., Marrero, M., Solís-Guzmán, J., 2015a. The ecological footprint of 685 dwelling construction in Spain. Ecol. Indic. 52, 75–84. 686 doi:10.1016/j.ecolind.2014.11.016 687 González-Vallejo, P., Solís-Guzmán, J., Llácer, R., Marrero, M., 2015b. La 688 construcción de edificios residenciales en España en el período 2007-2010 y su 689 impacto según el indicador Huella Ecológica. Inf. la Construcción 67, e111. 690 doi:10.3989/ic.14.017 691 Grunewald, N., Galli, A., Katsunori, I., Halle, M., Gressot, M., 2015. The Ecological 692 Footprint of Mediterranean Diets. 693 Heng, H.K.S., McGeorge, W.D., Loosemore, M., 2005. Beyond strategy: exploring the 694 brokerage role of facilities manager in hospitals. J. Health Organ. Manag. 19, 16– 695 31. doi:10.1108/14777260510592112 696 IDAE, 2011. Factores de emisión de CO2 / CO2 emission factors [WWW Document]. 697 URL 698 http://www.idae.es/uploads/documentos/documentos_Factores_Conversion_Ener 699 gia_y_CO2_2011_0a9cb734.pdf (accessed 8.4.15). 700 ISO 12006-2, 2015. Building construction -- Organization of information about 701 construction works -- Part 2: Framework for classification. 702 ISO 15686-5, 2008. Buildings and constructed assets -- Service-life planning -- Part 5: 703 Life-cycle costing. 704 Khatiwala, S., Primeau, F., Hall, T., 2009. Reconstruction of the history of 705 anthropogenic CO2 concentrations in the ocean. Nature 462, 346–349. 706 doi:10.1038/nature08526 707 Kitzes, J., Galli, A., Bagliani, M., Barrett, J., Dige, G., Ede, S., Erb, K., Giljum, S., 708 Haberl, H., Hails, C., Jolia-ferrier, L., Jungwirth, S., Lenzen, M., Lewis, K., Loh, J., 709 Marchettini, N., Messinger, H., Milne, K., Moles, R., Monfreda, C., Moran, D., 710 Nakano, K., Pyhälä, A., Rees, W., Simmons, C., Wackernagel, M., Wada, Y., 711 Walsh, C., Wiedmann, T., 2008. A research agenda for improving national 712 Ecological Footprint accounts. Ecol. Econ. 68, 1991–2007. 713 39 doi:10.1016/j.ecolecon.2008.06.022 714 Kohler, N., Hassler, U., 2002. The building stock as a research object. Build. Res. Inf. 715 30, 226–236. doi:10.1080/09613210110102238 716 Lenzen, M., Murray, S.A., 2001. A modified ecological footprint method and its 717 application to Australia 37, 229–255. 718 Marchi, M., Pulselli, R.M., Marchettini, N., Pulselli, F.M., Bastianoni, S., 2015. Carbon 719 dioxide sequestration model of a vertical greenery system. Ecol. Modell. 306, 46– 720 56. doi:10.1016/j.ecolmodel.2014.08.013 721 Marland, G., Boden, T.A., Andres, R.J., 2007. Global, regional, and national fossil fuel 722 CO2 emissions. Oak Ridge, TN. 723 Marrero, M., Ramirez‐De‐Arellano, A., 2010. The building cost system in Andalusia: 724 application to construction and demolition waste management. Constr. Manag. 725 Econ. 28, 495–507. doi:10.1080/01446191003735500 726 Martínez-Rocamora, A., Pulselli, R.M., Solís-Guzmán, J., Marrero, M., Bastianoni, S., 727 2014. Vertical Greenery Systems as sustainable solutions for building retrofitting: 728 a case study, in: Eco-Architecture V - 5th International Conference on 729 Harmonisation between Architecture and Nature. Siena, pp. 127–135. 730 Martínez-Rocamora, A., Solís-Guzmán, J., Marrero, M., 2016. LCA databases focused 731 on construction materials: A review. Renew. Sustain. Energy Rev. 58, 565–573. 732 doi:10.1016/j.rser.2015.12.243 733 Martínez-Rocamora, A., Solís-Guzmán, J., Marrero, M., 2015. A structure for the 734 quantity surveillance of costs and environmental impact of cleaning and 735 maintenance in buildings, in: The Sustainable Renovation of Buildings and 736 Neighbourhoods. Bentham Science Publishers, pp. 103–118. 737 Moore, D., Cranston, G., Reed, A., Galli, A., 2012. Projecting future human demand on 738 the Earth’s regenerative capacity. Ecol. Indic. 16, 3–10. 739 doi:10.1016/j.ecolind.2011.03.013 740 Moreno Rojas, R., Fernández Torres, A., García Pereda, J., Cámara Martos, F., Amaro 741 López, M.A., Ros Berruezo, G., Martínez de Victoria Muñoz, E., Martínez de 742 Victoria Carazo, I., 2015. Estimación de la distribución diaria de nutrientes en la 743 dieta estándar en España. Nutr. Hosp. 31, 2660–2667. 744 doi:10.3305/nh.2015.31.6.8896 745 National Institute of Standards and Technology, 2013. Building Life Cycle Cost 746 Program (BLCC). 747 Omniclass, 2012. Omniclass: A strategy for classifying the built environment - Table 748 13: Spaces by function. 749 Pulselli, R.M., Pulselli, F.M., Mazzali, U., Peron, F., Bastianoni, S., 2014. Emergy 750 based evaluation of environmental performances of Living Wall and Grass Wall 751 systems. Energy Build. 73, 200–211. doi:10.1016/j.enbuild.2014.01.034 752 Pulselli, R.M., Simoncini, E., Pulselli, F.M., Bastianoni, S., 2007. Emergy analysis of 753 building manufacturing, maintenance and use: Em-building indices to evaluate 754 housing sustainability. Energy Build. 39, 620–628. 755 doi:10.1016/j.enbuild.2006.10.004 756 REE, 2014. El sistema eléctrico español / The Spanish electric system. 757 Revuelta Marchena, M.P., Ramírez-de-Arellano Agudo, A., García Vázquez, P., 2015. 758 Control de costes por anticipación. Editorial Universidad de Sevilla, Sevilla. 759 40 Riahi, K., Grübler, A., Nakicenovic, N., 2007. Scenarios of long-term socio-economic 760 and environmental development under climate stabilization. Technol. Forecast. 761 Soc. Change 74, 887–935. doi:10.1016/j.techfore.2006.05.026 762 Royal Institute of Chartered Surveyors, 2015. BCIS Building Running Costs Online. 763 Solís-Guzmán, J., González-Vallejo, P., Martínez-Rocamora, A., Marrero, M., 2015. 764 The Carbon Footprint of Dwelling Construction in Spain, in: The Carbon Footprint 765 Handbook. CRC Press - Taylor & Francis Group, pp. 261–283. 766 Solís-Guzmán, J., Marrero, M., Ramírez-de-Arellano, A., 2013. Methodology for 767 determining the ecological footprint of the construction of residential buildings in 768 Andalusia (Spain). Ecol. Indic. 25, 239–249. doi:10.1016/j.ecolind.2012.10.008 769 U.S. Army Corps of Engineers, 2014. Economic Analysis Package (ECONPACK). 770 UNE-EN 15978, 2012. Sustainability of construction works. Assessment of 771 environmental performance of buildings. Calculation Method. 772 Van Den Bergh, J.C.J.M., Verbruggen, H., 1999. Spatial sustainability, trade and 773 indicators: An evaluation of the “ecological footprint.” Ecol. Econ. 29, 61–72. 774 doi:10.1016/S0921-8009(99)00032-4 775 Wackernagel, M., Rees, W., 1997. Our Ecological Footprint: Reducing Human Impact 776 on the Earth. New Catalyst Bioregional Series. 777 WWF, 2014. Living Planet Report 2014: species and spaces, people and places. 778 WWF, Gland, Switzerland. 779 780