Dynamic LCA framework for environmental impact assessment of buildings

Accepted Manuscript Title: Dynamic LCA Framework for Environmental Impact Assessment of Buildings Author: Shu Su Xiaodong Li Yimin Zhu Borong Lin PII: DOI: Reference:

S0378-7788(16)32089-8 http://dx.doi.org/doi:10.1016/j.enbuild.2017.05.042 ENB 7623

To appear in:

ENB

Received date: Accepted date:

31-12-2016 16-5-2017

Please cite this article as: S. Su, X. Li, Y. Zhu, B. Lin, Dynamic LCA Framework for Environmental Impact Assessment of Buildings, Energy and Buildings (2017), http://dx.doi.org/10.1016/j.enbuild.2017.05.042 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Dynamic LCA Framework for Environmental Impact Assessment of Buildings Shu Sua, Xiaodong Lia∗, Yimin Zhub, Borong Linc a

Department of Construction Management, School of Civil Engineering, Tsinghua University, 100084,

b

ip t

Beijing, China. [email protected], [email protected].

Department of Construction Management, Louisiana State University, 70803, Baton Rouge, LA, USA,

Department of Building Science, School of Architecture, Tsinghua University, 100084 Beijing, China.

us

c

cr

[email protected]

an

[email protected]

M

Abstract： ：Traditional Life Cycle Assessment (LCA) methods are used to conduct building environmental impact assessment (EIA) with little consideration of

Ac ce pt e

d

influential factors that vary in time and of variation in occupancy behaviors. Because the life cycle of a building is quite long, such details have significant influence on the accuracy of evaluation results. To fill in this gap and extend the LCA system, this paper developed a dynamic assessment framework based on LCA principles after reviewing the research progress of DLCA (dynamic LCA). The new framework identified four dynamic building properties (i.e., technological progress, variation in occupancy behavior, dynamic characteristic factors, and dynamic weighting factors) and considered them in corresponding assessment steps to realize real-time EIA. In addition, residential occupancy profiles were described at personal level, family level, and social level; and three potential quantification methods were introduced to ∗

Corresponding Author: [email protected]. (86)010-62784957; (86)13911750029.

Page 1 of 45

explore the relationship between occupancy profiles and household energy consumption. The DLCA framework expands the connotation of the LCA system from a dynamic perspective, making it possible to present time-varying EIs of

ip t

buildings over their long life cycle and guide occupancy behavior in time. This framework has the potential to be base for developing a useful tool for conducting

cr

forecast evaluation and promoting sustainability.

M

1. Introduction

an

(DLCA); technological progress; occupancy behaviors

us

Keywords: building; environmental impact assessment (EIA); Dynamic LCA

d

Construction is a high consumption, high pollution industry that makes a

Ac ce pt e

considerable contribution to total energy consumption (30–40% in developed countries [1] and 20–25% in China [2]). With some increasingly serious global issues arising, such as the energy dilemma, climate warming, and so on; buildings have become a major focus of attention from the perspective of sustainability. To scientifically quantify the influences of major environmental impact (EI) factors and to effectively propose improvement measures, many Life Cycle Assessment (LCA)-based building environmental impact assessment (EIA) models were developed in recent decades. Building for Environmental and Economic Sustainability (BEES) in the U.S.A. measures building EIs during raw material acquisition, manufacturing, transportation, installation, use, and waste management stages by

Page 2 of 45

synthesizing the environmental and economic results into a single score [3]. “Optimization of global demands in terms of costs, energy and environment within an integrated planning process” (OGIP) calculates EI in the construction and operation

ip t

phases, and helps designers portray the complex relationships between costs, energy and EI [4,5]. EcoQuantum classifies the environmental performance of a building

cr

during its life span into 11 environmental impact categories, and then aggregates them

us

into four environmental scores: resources, emissions, energy, and waste [4,6]. The

an

Building Environmental Performance Analysis System (BEPAS) [7,8] in China allocates building EIs into ecological damage and resource depletion, and then into

M

many sub-categories. It has been widely applied and tested in engineering practices

d

[9,10]. The Building Health Impact Assessment System (BHIAS) [11,12] establishes

Ac ce pt e

the links between emissions and consequent health damages and aims to quantify damage in four categories (i.e., global-warming-related diseases, respiratory effects, circulatory effects, and carcinogenesis) due to construction activities in China. Although these building EIA models are widely used in the world and have

offered guidance for improvement in engineering practice, they still have some limitations in theory and practice:

(1) Traditional LCA methods have two theory defects: there is little consideration of the time-variance of parameters over the long life cycle of a building, and little attention paid to the behaviors of occupants. Buildings are different from many other products because they have very long life cycles (usually 40 to 70 years), and during

Page 3 of 45

this long interval, the environment, economy, and society may change greatly and affect the building environmental performance significantly [13-15]. The time points when pollutants are emitted strongly affect their fate factors and then their impacts on

ip t

the receptors [16]. Many impact categories (such as water consumption, greenhouse gas (GHG) emissions, etc.) are time sensitive and might vary significantly as a

cr

function of time [17]. It is common sense that ignoring time-varying influences

us

decreases the accuracy of assessment results [18-22] and blocks extending

an

applications of LCA [23-25]. The ISO also addressed this limitation by pointing out that static LCA decreases the environmental association of results to some extent

M

[26,27]. Moreover, buildings are typical examples of situations in which significant

d

uncertainties due to occupancy behaviors may prevent green engineering solutions

Ac ce pt e

from realizing their full potential. Occupancy behaviors significantly affect a building’s EI over long periods of time, and should be considered in the assessment. (2) On the other hand, the traditional LCA methods have apparent limitations in

giving full play to application values. Most of the traditional LCA methods were intended to serve various involved entities: some methods aimed at guiding constructors toward green construction [8]; some aimed at guiding designers to make green plans [28]. Others provide suggestions to governments for establishing environmental policies [29], and some reflect how much attention the public focuses on environmental problems [7]. However, the influence of occupancy behaviors has not been assessed systematically nor improved scientifically, even though it plays a

Page 4 of 45

significant role [30-32]. Proper occupancy behaviors would help green designs work better to alleviate the building EIs [33]. Otherwise, the building and equipment systems with higher efficiency and more advanced technologies, may exhibit larger

ip t

energy consumption and worse performance than conventional buildings [34]. The lack of understanding and management of occupancy behaviors has been a significant

cr

obstacle to building energy efficiency [35]. Thus, conducting effective assessment,

us

feedback, management, and improvement of occupancy behaviors in buildings could

an

do greatly help to make the evaluation results more valuable in actual application and could better promote sustainable development.

M

To address the gaps between the requirement of dynamic consideration and

d

current static implementation of LCA methodology, this paper tries to develop a

Ac ce pt e

dynamic LCA (DLCA) with time-varying factors and occupancy behaviors into consideration. Four dynamic properties (i.e., technological progress, occupancy behaviors variance, characteristic factors, and weighting factors) are brought into a static LCA model to develop a dynamic one that could be used to quantify building EIs over time. The new DLCA framework could help improve the building LCA theoretical base, extend the connotation of LCA system from a new perspective, assess the influences of occupancy behaviors, and promote sustainable buildings.

2. Literature review 2.1 Static LCA model

Page 5 of 45

The LCA concept was first proposed in the 1990s and its framework involves four steps: goal & scope definition, life cycle inventory (LCI) analysis, life cycle impact assessment, and interpretation [27], as shown in Fig. 1. Over the past several

ip t

decades, many mature LCA-based evaluation systems were developed and applied in various industries and for various products. DLCA model was developed abased on

us

cr

the static LCA framework and follows its evaluation paradigm and steps.

Goal & Scope Definition

an

Life Cycle Inventory Analysis

Application

Interpretation

M

Impact category definition Classification Characterization

Decision

d

Weighting

Ac ce pt e

Life Cycle Impact Assessment

Fig. 1. LCA framework.

2.2 Progress in DLCA research The concept and expression of DLCA came up a few years ago and soon

attracted lots of attention. The first paper that mentioned “dynamic LCA” was published in 2003 (from a search using Google Scholar and Web of Science). It considered the changes caused by continuous generation of new products, by-products, and material wastes, and then proposed a dynamic LCA model to assess the impact of lead free solder [36]. In recent years, DLCA related research has received more and

Page 6 of 45

more attention, and has made remarkable progress in different industries and fields. These studies mainly emphasized two aspects of “time”: economic and social progress, and dynamic characterization factors (CFs), as summarized in Table 1. In

Ac ce pt e

d

M

an

us

cr

categories, respectively, and then of DLCA studies for buildings.

ip t

the following section, a review is presented of studies considering these dynamic

Page 7 of 45

cr us an

Table 1 Summary of partial DLCA research

Pehnt

Published time 2006

Evaluated objective

Dynamic categories

Indicators

Reference

Three cases: photovoltaics,

Economic and social

steel production, aluminum production, electricity

[37]

steam turbine power plant,

progress

production, module efficiency, optimized manure

M

Authors

ed

and central heating with

production, et al.

forest timber

Levasseur

2010

Ac

et al.

2009

Collinge

Mechanical products

pt

Wei

2007

2011

Dynamic CFs

recovery rate of propene polymer, silica gel, polyvinyl copper; annual growth rate of the heating time

Economic and social

growth rate of non-agricultural population, growth rate of

management

progress

per capita disposable income, Engel coefficient, central

Corn ethanol

[38]

chloride, poly carbonate, stainless steel, aluminum, glass,

Municipal solid waste

ce

Qi et al.

[39]

heating ratio Economic and social

yearly emissions of CO2, CH4, N2O and land-use change;

progress, dynamic

global warming CF

[20]

CFs Buildings

et al.

Economic and social

electricity usage, steam usage (heating, hot water), chilled

progress

water usage (cooling, refrigeration), CO2 concentration,

[40]

VOC concentration, particulate concentration Levasseur

2012

An afforestation project

et al.

Economic and social

yearly emissions of CO2, global warming CF

[41]

yearly emissions of CO2 and CH4, global warming CF

[42]

progress, dynamic CFs

Levasseur et al.

2013

A wooden chair

Economic and social progress, dynamic CFs

Page 8 of 45

cr us 2013

an

Collinge

Buildings

Collinge

2013

Buildings

Yang and

2014

Chen

Economic and social

pollutant data, operating data, inventory analysis, exposure

progress

estimates

A crop residue gasification

Economic and social

electricity mix, steel manufacturing technological progress,

project

progress, dynamic

global warming CF

2014

Filleti et al.

al.

Fouquet et al.

2014

[43] [44] [45]

CFs timely input-outputs of materials, energy and wastes

[46]

Economic and social

real time emission of CO2-eq and Sb-eq, real time

[19]

progress

consumption of water

Mechanical manufacturing

Economic and social

system

progress

Manufacturing processes

2014

Metals

Dynamic CFs

time-horizon-dependent fate factors

[47]

2015

Three single family houses

Economic and social

global warming CF, technological breakthrough of

[48]

progress, dynamic

electricity mix, improvements in cement and expanded

CFs

polystyrene production

Ac

Lebailly et

Buildings

[15]

photochemical ozone, et al. estimates

ce

Li

time-adjusted global warming potentials, seasonal CFs for

progress

ed

al.

progress

pollutant data, operating data, inventory analysis, exposure

pt

2014

electric power generation, GHG emissions, air pollutants,

Economic and social

et al. Collinge et

M

et al.

Economic and social

Page 9 of 45

(1) DLCA researches considering dynamic socio-economic parameters The time horizon determines the maturity and performance of a society and economy, and some dynamic socio-economic parameters have been considered in

ip t

DLCA researches. Pehnt [37] considered the time-variation of electricity mix, some materials and production processes in the assessment of GHGs and acidification

cr

impacts. Qi et al. [38] analyzed the dynamics of energy property of mechanical

us

products in different stages of their life cycle. Wei [39] carried out a time series study on characteristics of Tianjin’s municipal solid waste management from 2000 to 2007,

an

considering the growth rate of the non-agricultural population, disposable income,

M

Engel coefficient, and so on. Filleti et al. [19] collected real-time inventory data directly from the productive processes; then performed a web-based dynamic LCI. Li

d

[46] used timely input-outputs of materials, energy, and wastes in assessment of a

Ac ce pt e

mechanical manufacturing system.

(2) DLCA research considering dynamic CFs The following studies considered dynamic characterization factors and

demonstrated its necessity and importance in EI assessment over a finite time horizon. Lebailly et al. [47] addressed the time-horizon-dependent CFs of freshwater ecotoxicity impacts of 18 metals. Levasseur et al. [20] developed a DLCA method involving a time function of global warming CFs. This method has been applied to assess the global warming impacts of corn ethanol [20], a wooden chair [42], and an afforestation project [41]. Yang and Chen [45] assessed and compared the global warming impacts of a crop residue gasification project with both static and dynamic

Page 10 of 45

global warming CFs. (3) DLCA research in building In October 2015, scholars from all over the world attended a workshop focusing

ip t

on “Dynamic Life Cycle Assessment - A Human and Business Perspective”, which implied that more and more researchers have realized the importance of building

cr

DLCA and started work on this hot topic.

us

In past years, William Collinge put a lot of efforts on building DLCA research covering environmental protection, human health damage evaluation, and productivity.

an

He pointed out that a dynamic building LCA framework should consider temporal

M

variations in the operation phase, and rapidly update results according to different design and operation scenarios [40]. His paper in 2013 [15] identified major

d

time-related changes of variables (i.e., dynamic modeling of unit processes, temporal

Ac ce pt e

variations in industrial systems, temporal variations in emissions/resources, and temporal variations in environmental systems); and then developed a dynamic model based on the general LCA equation. Then he compared the EIs of an institutional building by static LCA and DLCA methods using a 1971 perspective and a 2009 perspective. In another study [43], indoor environmental quality was incorporated into a DLCA model with three types of impacts included: internal chemical impacts, internal non-chemical health impacts, and performance/productivity impacts. Based on this research, a new paper in 2014 [44] added productivity and non-chemical health impacts to outline a framework. Collinge’s researches provided good exploration of building DLCA focused on integrating indoor environmental quality

Page 11 of 45

into assessment, offering a foundation and reference for further in-depth study. Fouquet et al. [48] used a dynamic LCA method [20] to assess the global warming impact of three low-energy houses over time with the technological breakthrough of

ip t

electricity mix, and some innovation for refurbished materials (cement and expanded

cr

polystyrene) into consideration.

us

2.3 Limitations

Although some progresses have been made in this field, DLCA research is at the

an

premature stage for the following two limitations:

M

First, most of the previous DLCA frameworks were applied in industries, such as crop residue gasification projects [45], mechanical manufacturing systems [46],

d

manufacturing processes [19], and afforestation projects [41], which are not suitable

Ac ce pt e

for buildings for the uniqueness and longtime life span. Every building is different with its own character, and the inputs-outputs (consumed resources and emitted pollutants) of different constructions are quite diverse. In addition, the life cycle of a building is usually dozens of times longer than that of other products, and the operation process changes with time. This implies that dynamic factors have longer and more complicated influence on the evaluation results. Thus, it becomes necessary to develop a specific DLCA model for buildings with various processes and complicated activities taken into account. Second, the available building DLCA models paid much attention to dynamic socio-economic parameters and dynamic CFs, but ignored the influence of dynamic

Page 12 of 45

occupancy behaviors. These incomplete assessment models have difficulties in offering accurate evaluation results and in mitigating environmental burdens in the operation stage. Moreover, these DLCA modes were developed according to foreign

ip t

considerations and cannot be directly applied to Chinese buildings because in various countries and regions, the energy consumption patterns, occupancy behavior habits,

cr

and user responses to changing environments are quite different. The influential

us

factors, function routes, analysis emphasis, and evaluation results are space-varying. Thus, a comprehensive DLCA framework involving a variety of kinds of dynamic

M

an

properties, and of Chinese user behaviors, is in great demand.

3. Development of building DLCA framework

d

3.1 Identification of dynamic properties

Ac ce pt e

Based on the above analysis and previous references, this study classifies the

time-dependency factors in building EIAs into four types: technological progress, variation in occupancy behavior, dynamic characteristic factors, and dynamic weighting factors; as shown in Fig. 2. •

Technological progress: With the development of economy and science,

construction technology and machine efficiency improve markedly. As a result, material consumption quantities, input-output flows, energy mix vary over different time horizons. •

Variation in occupancy behavior: The behavior of occupants has direct

influence

on

household

energy

consumption,

and

consequently,

building

Page 13 of 45

environmental performance. Occupancy behaviors are complicated and affected by lots of factors, such as income, age, and energy-saving awareness. With progress in the economy and society, more green appliances will enter the market; residents

ip t

become older, and may pay more attention to environmental protection. Thus, occupancy behaviors will change over time.

Dynamic characteristic factors: A unit pollutant emission released today

cr

•

us

has different consequences from one that might be released decades later, considering the different background concentrations and atmospheric composition. CFs of

an

pollutants vary significantly in different regions and time periods, and local dynamic

•

M

CFs is suggested in the DLCA research.

Dynamic weighting factors: Weighting factors (WFs) are time-sensitive for

d

the indexes commonly used to calculate them (i.e. public concerns, green taxation

Ac ce pt e

systems, pollution charge fees, government environmental policies, and expert opinions) are all time-dependent and influenced by the developing economy and society. Thus, appropriate dynamic WF values could be identified according to the assessed objective, building location, and the time horizon of the evaluation, to present the actual weightings.

Technological progress Variation in occupancy behavior

Building DLCA

Dynamic characteristic factors Dynamic weighting factors

Fig. 2. The dynamic properties in a building DLCA

Page 14 of 45

Among these four dynamic properties, technological progress and variation in occupancy behaviors are closely linked to building processes, while the dynamic CFs

ip t

and WFs are integral with the LCA method itself. These four properties will be introduced into the new LCA model to develop a dynamic framework in Section 3.2.

cr

In Section 4 and Section 5, we will explain how to conduct dynamic assessment with

us

technological progress and variation of occupancy behaviors into consideration in some detail by reviewing literature, putting forward potential assessment methods,

an

and analyzing influence factors. It is noted here that dynamic CFs and WFs are not

M

introduced in this paper due to their weak links with construction and insufficient research basis. The time variations of CFs and WFs are independent on the assessed

d

building consumptions and inventory flows, and their dynamic values in different

Ac ce pt e

building EIA assessment applications in one time horizon are the same. Besides, there were researches focusing on the time variance of CFs for acidification [49] and global warming [48], but mature dynamic CFs and WFs of other categories are still inadequate. They could be incorporated into the DLCA model once they are mature enough, because the DLCA system is open and extensible.

3.2 DLCA framework

Based on LCA principles and mature static building EIA models [7,11,12], four dynamic properties were applied to develop a DLCA framework for buildings, as shown in Fig. 3. Related time-varying factors are shown in the assessment steps, and the dynamic flows marked in red are building-related while the flows marked in green

Page 15 of 45

are LCA model related. In the new framework, the inventory flows involved are time functions, and the evaluation steps are introduced below. •

The scope definition step provides specific boundaries for the evaluation, and

ip t

the general scope in LCAs is the whole building life cycle, spanning “from the cradle to grave”. Considering the pre-use phase is quite short while the operation and

cr

maintenance phases take up almost the whole life cycle, variations over time mainly

us

impact the activities and environmental performance in operation and maintenance phases. This study takes the operation and maintenance stages as its research scope,

an

and the household daily activities and maintenance constructions as research

•

M

objectives.

The consumption data identification step makes clear how many resources

d

and equipment are consumed. The energy and resource consumption in the operation

Ac ce pt e

stage, and building materials and mechanical equipment running in maintenance stages are common data types. Technological progress and variation in occupancy behaviors impact the types and quantities of consumption, and play significant roles in consumption identification. •

The inventory data identification step transforms the consumptions into

inputs (resources and energy) and outputs (pollutions). The current mature widely-used inventory databases, such as ecoinvent [55], European Reference Life Cycle Database, and Chinese Life Cycle Database [56,57] offer little temporal information. A time-varying inventory database considering the influences of technological progress is critical in this step.

Page 16 of 45

•

The classification & characterization step classifies the impacts into

categories, and then quantifies the relative contributions of each to its assigned impact category. This framework divides inputs and outputs into three common

ip t

corresponding categories (i.e., ecological damage, resource depletion, and human health damage) and into several subcategories referring to BEPAS and BHIAS.

cr

Dynamic CFs in corresponding time horizons will be taken in this step to replace

•

us

constant ones.

The weighting step integrates the relative severity of EIs expressed in

an

different units into a single value. Monetization, distance-to-target, and expert panels

M

are three common quantitative-weighting approaches, and time-varying factors are recommended for consideration as a way to offer dynamic WFs at the time points. The interpretation step is used to analyze the calculation results, to identify

d

•

Ac ce pt e

major impact categories and burdening substances, and to put forward effective improvement suggestions. Sensitivity analyses are encouraged in this step. The assessment results in this dynamic model are time series.

Page 17 of 45

cr us Energy mix (t)

Household energy consumption (t)

Inventory Data Identification

Inputs outputs (t)

ce

Building material consumption: Concrete (t) Cement (t) Steel (t) ……

Ac

Maintenance stage

Technological progress

Variation of occupancy behavior

Classification & Characterization

Energy consumption: Electric (t) Diesel (t) ……

Characteristic factor (t)

Pollutant emission : CO2 (t) SO2 (t) TSP (t) ……

Dynamic Characteristic Factor

Dynamic Weighting Factor

Weighting & Interpretation

Ecological Damage: Global warming (t) Acidification (t) Solid waste (t) ……

Resource consumption: Water (t) Timer (t) ……

pt

Operation stage

Energy consumption: Electric (t) Natural gas (t) Coal (t) ……

ed

Resource and energy consumption (t)

an

Consumption Data Identification

M

Scope Definition

Resource Depletion: Water depletion (t) Fossil energy depletion (t) ……

Weighting factor (t)

Health Damage: Global warming related diseases (t) Carcinogenesis (t) Respiratory effect (t) …… Building related dynamic flows are in red LCA method related dynamic flows are in green

Fig. 3. The building DLCA framework

Page 18 of 45

Environmental & Health Performance (t)

4. Technological progress Technological progress in construction is embodied in many aspects. For example, more renewable energy will be incorporated into the energy mix, aging

ip t

equipment may bring about energy conversion efficiency reduction, inputs and outputs of materials will change owing to the manufacturing efficiency improvement,

cr

pollutant emissions from energy consumptions might be different, and so on. Pan [60]

us

classified construction-technology improvement into four types: product improvement, building materials improvement, construction technology improvement and

an

construction-enterprises management-level improvement. Collinge [11] thought that

M

improving technology levels would bring about dynamic building operations, dynamic supply chain, and dynamic inventory. Because this study aims to assess

d

technological progress in the LCA model, it is critical to make clear how

Ac ce pt e

improvements affect inventory data over time. To address this issue, three questions should be answered.

(1) What resources and energies are consumed? In the operation stage, common household activities such as cooling, heating,

cooking, bathing, and lighting consume significant amounts of water and heating energy (electricity and fuels). The main consumptions in the maintenance stage are building materials, energy used in residential maintenance, and updating of household devices. Technological progress in the future optimizes the energy mix and more renewable energies will be involved in the assessment. Thus, the present composition and component of one unit consumed energy may be different from a future unit.

Page 19 of 45

Many researchers worked on predicting trends in energy structure optimization, and offered some available methods and potential scenarios for further research. Matsumoto [50] used a computable general equilibrium model to investigate the

ip t

energy structure in the following 100 years under climate-mitigation policy scenarios. Wang et al. [51] simulated China's energy structure in future 40 years based on energy

cr

demand and supply equilibrium. Ren [51] applied complex adaptive system theory

us

and discussed the mechanism of technological innovation; then forecasted energy supply structure. These studies and methods help to determine the trend in energy

an

structure development, and what energies will be consumed.

M

(2) What are the quantities of consumptions?

Technology improvements have impacts on consumption quantities in activities.

d

Pehnt [37] considered energy converter optimization, electricity supply improvement,

Ac ce pt e

and ecological transport system optimization for fuel in the LCA of renewable energies. Li [46] proposed a multi-layer dynamic model to assess the variable carbon emissions in mechanical manufacturing systems. Qi [38] analyzed many energy-consuming characteristics of mechanical products (e.g., production efficiency improvement, fabrication process improvement, and performance degradation). For the buildings evaluated, building material consumption mainly comes from maintenance and updates of some household appliances. The construction technology improvements may decrease building material consumptions and improve their mechanical efficiency. In addition, the recovery rates and renewable rates of various resources will increase with the development of science and technology. These

Page 20 of 45

changes will influence the actual consumption quantities. (3) What are the timely inputs-outputs of resources and energies? The raw materials and energy required, and the consequential pollutant emitted

ip t

in one functional unit of products, vary along with the scientific and technological improvements. For instance, the standard coal consumption for power supply in China

cr

has decreased nearly 30% in the last 35 years (from 448 to 315 g/kWh), meaning a

us

significant mitigation of environmental burdens [52]. Thus, it is suggested that inputs-outputs in corresponding time horizons are applied in assessment. Gu [53]

an

analyzed the time-varying life cycle inventory of energy and some common building

d

for this study.

M

materials due to technology development from 1990 to 2005, providing a reference

Ac ce pt e

According to the above analysis, there categories are considered in technological

progress: energy mix, consumption, and inventory data. In each category, many factors are involved during the assessment, as listed in Fig. 4. In addition, Table 2 summarizes some available data acquisition methods for these factors, indicating that it is possible to bring these factors into dynamic assessment to make it work.

Page 21 of 45

Traditional energy structure Energy mix Renewable energy structure

Material consumptions in maintenance and household devices production Energy consumption in maintenance and household devices production

ip t

Technology progress

Consumption

cr

Recovery rate of materials and resources

us

Renewable rate of resources

Inputs-outputs of materials Inventory data

an

Inputs-outputs of energies

M

Fig. 4. Categories and sub-categories of technological progress in buildings

Table 2 The data acquisition methods available for factors Factor

Available methods

Investigation and estimation [54]

Recovery rate of materials and resources,

Government planning, such as [55]

Ac ce pt e

d

Material and energy consumptions in production of maintenance and household devices

renewable rate of resources

Traditional energy structure, renewable energy structure

Inputs-outputs of materials and energies

Government planning, such as [56]; computable general equilibrium model and scenario analysis [50,57]; long-range energy alternatives planning model and information entropy model [58]; energy demand-supply model [51]; autoregressive integrated moving average model [59]; complex adaptive system [60]; Markov chain [61]. Literature investigation, interpolation method and extrapolation method [53].

5. Variation in occupancy behaviors As an important stakeholder with long lifetimes, occupants have significant influences on the building energy consumptions [31,62-65]. A number of studies have

Page 22 of 45

demonstrated that in buildings with the same structural system and external environment, the difference of consumption levels influenced by occupancy behaviors reached several times or even dozens of times. A survey conducted in Beijing showed

ip t

that 20% families consumed about 70% of the total energy for cooling, and the highest consumption was nearly 5-times higher than the average level [66]. The

cr

Architectural Institute of Japan investigated four typical households and found that

us

the electricity consumption for annual heating varied from 0.4 to 9.0 MWh [67]. Surveys in Japan and Kuwait concluded that energy consumption is closely related to

an

the life styles of residents [68, 69]. Thus, better understanding and management of

M

occupancy behavior assist building performance assessment, behavioral guidance, and energy efficiency improvement [32,35,70].

d

The assessment of the variation in occupancy behaviors can be classified into

Ac ce pt e

four sub-steps: occupancy profiles identification, occupancy behaviors quantification, occupancy profiles forecast, and energy consumption forecast. The first two are important ones in establishing the relationships between occupancy profiles, behaviors and energy consumption, and they are described and analyzed in detail in the following sections. The occupancy profiles forecast describes the changing trend and can be conducted by sociological analysis and tracking surveys. Abased on the first three sub-steps, energy consumptions in the whole life cycle can be predicted easily. It is noted that the occupancy behaviors and profiles in residential buildings and public buildings are quite different regarding their different energy use characteristics

Page 23 of 45

and influential factors. Due to limitations of space, this study focuses on occupancy behaviors in residential buildings for the large amount of new residences every year.

5.1 Occupancy profiles identification

ip t

The common behaviors in residences can be classified into two categories: event behaviors (e.g., cooking, bathing) and appliance usage behaviors (e.g., ventilation

cr

facility, lighting facility, air conditioning, heating equipment, TV, computer) [71, 72].

us

These behaviors vary by individuals and are affected by many factors, which are summarized in the occupancy profiles. There are many classifications of occupancy

an

profiles from different perspectives (e.g., architectural features, lifestyle, awareness of

M

energy conservation, and equipment characteristics [73]; economy, geographical location, education, and demographic factors [74]; social property, natural property,

d

and spiritual property [75]; policy, life style, energy knowledge, social psychological

Ac ce pt e

variables, and other factors [76]). Taking Pang’s research [77] as reference, we describe the occupancy profiles from three levels: person, family, and society, as shown in Fig. 5. At the personal level, physiological and psychological individual characteristics are considered. Different people have distinct histories, attitudes, and sociocultural demographics that affect their energy-use behavior and thereby affect household energy consumption [78-80]. Profiles at the family level focus on the household size, area, and two special groups (the elderly and children). Researches [73, 81] show that the elderly spend more time at home and consume more energy, while the parents of children conduct more energy-saving practices [77]. Moreover, social, moral, and normative concerns are also involved in the profiles, because they

Page 24 of 45

play a role in the resident energy-use patterns. The factors at three levels of occupancy profiles as well as the sample literature sources are summarized in Table 3.

Family level

Personal Level • Physiological •

Social level

cr

Psychological

ip t

Occupancy Profiles

us

Fig. 5. The residential occupancy profiles

an

Table 3 The factors in occupancy profiles

Sample literature sources

Factor

Personal level (Physiological)

Age Gender Income Occupation Health

Personal level (Psychological)

Thermal sensation Cognition Attitude Preferences

[32,84] [70,78,84] [77,78,85] [85]

Family level

Residence area Household size Number of old people Number of children

[80,82] [82] [70] [70,77]

Social level

Environmental protection related moral concerns Environmental protection related normative concerns

[77,86] [86]

Ac ce pt e

d

M

Category

[80,82] [70,82] [80,82] [32] [32,83,84]

5.2 Potential Methods for quantifying occupancy behaviors Occupancy profiles affect occupancy behaviors, and result in specific energy and resource consumptions (the outputs in LCA). It is important to make clear the functional mechanism between behaviors and consumptions, from which future

Page 25 of 45

consumptions can be well estimated and assessed. Regression analysis, mathematical simulation, and environmental simulation are three potential methods available to quantify the relationship, and they are compared in Table 4.

ip t

Regression analysis: The multiple linear regression method is often used to examine the relationship between a group of independent variables and a single

cr

dependent variable. It estimates the regression coefficients by minimizing the squares

us

sum of deviations to the proposed regression model. This method is widely used to model the relationship between influential factors and building energy consumption

describes the expected change in Y in response to a

M

Xi is a dependent variable,

an

[65,87,88]. Blow shows a regression equation, in which Y is the independent variable,

is the random

d

unitary change in Xi when the rest of predictors remain constant, and

Ac ce pt e

error. Multiple linear regression is easy to develop, but is also inaccurate and inflexible to some extent [88]. In the assessment of building energy consumptions, occupancy profile factors are the dependent variables (such as age, gender, income, etc.) and the energy consumption is the independent variable. The input data can be empirical statistical data, or derived from field survey.

Mathematical simulation: With the development of computer technology, more and more mathematical simulation tools have been developed to simulate and analyze building energy consumptions (e.g., DOE-2 and EnergyPlus in the USA [89], ESP in England [90], DeST in China [91]). Many researchers have developed new energy

Page 26 of 45

consumption simulations based on models, such as the stochastic Markov [92], agent-based simulation model [93], and Block Configuration Model [93]. According to the US DOE’s BTS Directory (www.energytoolsdirectory.gov), there are around

ip t

350 building energy-simulation tools currently in use all over the world. Generally, performing the simulation involves understanding the nature of the issue to be solved,

cr

choosing a suitable simulation program, using the simulation program, and then

us

interpreting the simulation results and making decisions [30]. This method also has its limitations: it is a little difficult to perform in practice because it is highly complex,

an

and often there is a lack of input information [88].

M

Environmental simulation: In recent years, immersive virtual environment (IVE) simulation is maturing and now provides a new way to study occupancy

d

behavior and measure performance [94]. IVE immerses participants in an

Ac ce pt e

experimental, environmental scene to visualize and simulate possible user interactions, track

routes

within

a

designed

environment

[95],

and

simulate

user

energy-consumption behavior [96]. Previous research has indicated that human performance, feelings, perceptions, and behavior in an immersive virtual environment are similar to those in an actual physical environment [94,97]. Environmental simulation has already been used to study occupant lighting-use behavior in buildings [94]. Table 4 Comparison of three methods

Common data source

Regression analysis

Mathematical simulation

Environment simulation

Empirical statistical data

Empirical statistical data and field survey

Field measurement

Page 27 of 45

General

Good

No No Mainly used in lighting setting Good

[65,87,88]

[98,99]

[94,100]

ip t

No Yes Widely

cr

Simplicity Maturity Being used in behavior assessment Accordance with practical situatio n Sample applications

and field survey Yes Yes Widely

According to the above analysis, mathematical simulation is the optimal choice

us

and suggested to be applied in DLCA to quantify the occupancy behaviors. Unlike the

an

environment simulation method, mathematical simulation is relatively mature and widely used. Besides, its results have good accordance with practical situations. The

d

M

related input data can be derived from field measurement and questionnaire survey.

Ac ce pt e

6. Considerations for application The significant difference between the application of DLCA and static LCA is

the input data are time varying, and the values in different years are different. Figure 3 illuminates what time-varying factors are considered in this framework and when they will be inputted into the model. Being limited by immature data and insufficient in-depth studies, it’s difficult to apply the DLCA framework into a real case. The following sections and Table 5 will introduce the calculation sub-steps of these time-varying factors, and their potential calculation methods and data sources, showing how the model work and its potential to be a useful tool. Dynamic resource and energy consumption includes two parts: 1) dynamic material and energy consumption in maintenance stage; and 2) dynamic recovery rate

Page 28 of 45

and renewable rate of resources. Maintenance standards of residences could help identify the time and contents of building maintenance, and then the material and energy consumptions can be estimated by referring engineering quotas. Besides, the

ip t

recovery rate and renewable rate of resources are available in government planning and reports [55], and these targets are the balanced results of economic conditions,

cr

technological conditions and political conditions.

us

Dynamic energy mix is the composition changes of energy structure in building’s life cycle. The energy composition is improving and the share of renewable

The

short-term predictions

can

be

conducted

by

some

M

development.

an

energy will grow up with the increasing emphasis on environment and economic

mathematical methods, such as demand-supply model [51], complex adaptive system

d

[60], Markov chain [61], etc. and the long-term energy mix is suggested to be

Ac ce pt e

predicted by scenario analysis [50, 57].

The inputs-outputs (consumed resources and discharged emissions) vary with

time and should be considered. In this step, we only focus on some major building materials and common energies, and their inputs-outputs at different time in the past years could be derived from previous studies. Based on these history data and the development tendency of technology, inputs-outputs in the future years could be predicted [53]. Dynamic household energy consumption includes four sub-steps: occupancy profiles identification, occupancy behaviors quantification, occupancy profiles forecast, and energy consumption forecast, which are stated in Chapter 5 in detail and

Page 29 of 45

not repeated here. Dynamic characteristic factors have been studied by some researches, and can be applied directly or calculated by the methods referring to the literatures [16, 45,

ip t

101-104]. Dynamic weighting factors can be calculated according to shadow prices of

cr

various resources and green taxes of pollutants in different time horizons, which is

us

monetization weighting method [105]. It can also be calculated according to the

M

distance-to-target weighting method [29].

an

environmental protection policy targets in different time horizons, which is the

It should be noted that, environmental climate, economic development, envelope

d

structure of buildings, occupancy behavior habit, environmental planning, etc. are

Ac ce pt e

different from region to region. So, the assessment data related to technological progress, dynamic characteristic factors, and dynamic weighting factors are national data or reginal data. The data about occupancy behaviors could be regional ones to forecast the average EIs of residences in one city in the following years, or be a family case to forecast their EIs according to the family development.

Page 30 of 45

cr us an

Table 5 The time-varying factors and their related calculation sub-steps and potential methods Time-varying factors

Calculation sub-steps

Potential methods

Data source

Current energy mix

Document query

Statistics

Influence factors identification

Literature review

Statistics and related researches

M

Dynamic property

ed

Relationship establishment Energy mix (t)

between energy mix and influence factors

pt

Influence factors prediction

Energy mix prediction

Technological

dynamic material and energy

Resource and

consumption in maintenance

energy

stage

consumption (t)

dynamic recovery rate and

Ac

ce

progress

Inputs-outputs (t)

renewable rate of resources Inputs-outputs collection in history Inputs-outputs prediction Occupancy profiles identification

Variation of occupancy behavior

Household energy consumption (t)

1) Short-term predictions: mathematical methods, such as demand-supply model,

Statistics, reports, government

complex adaptive system, Markov chain, etc.

planning, etc.

2) Long-term predictions: scenario analysis. Document query, investigation and

maintenance standards, statistics,

estimation

field measurement, etc. government planning,

Document query

environmental programs, etc.

Literature review

Statistics and related researches

Mathematical simulation Literature review

Related researches

Occupancy behaviors

Mathematical simulation, such as DeST,

quantification

EnergyPlus, etc.

Occupancy profiles prediction

Sociological analysis and tracking surveys

Energy consumption prediction

Calculation

Occupancy event behaviors and appliance usage behaviors, local climate, building envelopes, etc.

Page 31 of 45

Statistics and related researches

cr us characteristic factor

factor (t) Weighting

factor

factor (t)

Characteristic factor prediction

Weighting factor prediction

Literature review

Related researches

1) Monetization method: shadow price and green tax prediction; 2) Distance-to-target

Statistics, reports, government

method: applying environmental protection

planning, etc.

targets in future years.

Ac

ce

pt

ed

Dynamic weighting

an

Characteristic

M

Dynamic

Page 32 of 45

7. Conclusions To improve the theoretical and practical limitations of traditional LCA in building EIA, this paper has developed a DLCA framework after reviewing and

ip t

analyzing existing researches on this topic. Four dynamic properties (technological progress, variation of occupancy behaviors, dynamic characteristic factors, and

cr

dynamic weighting factors) with their time-varying factors are considered within the

us

framework of new model. The dynamic effects of technological progress are

an

classified into three categories (energy mix, consumption, and inventory data) and into many sub-categories. For occupancy behaviors assessment, residential occupancy

M

profiles are described at personal, family, and social levels, and then three methods

d

(regression analysis, mathematical simulation, and environmental simulation) were

Ac ce pt e

briefly introduced as potential ways to quantify the relationships between energy consumption and occupancy profiles. Finally, some considerations for framework application have also been summarized. The new DLCA framework makes great contributions to improve LCA theory

and extend the connotation of LCA from a new perspective; thereby laying foundations for further studies. This is the first study in which the dynamic properties involved in constructions have been analyzed systematically, and in which the impacts from variation in occupancy behaviors have been considered substantially. The use of dynamic assessment will help predict building EIs more accurately and scientifically, acknowledge real-time building performance, and thus offer effective design guidance

Page 33 of 45

before actual construction. Moreover, this framework has the potential to be a useful tool for evaluation, improvement, and management of occupancy behaviors to alleviate environmental burdens.

ip t

However, this study only provided the DLCA framework and some research ideas; large amounts of work remain to be carried out. Future research may proceed

cr

from these perspectives: (1) carrying out the new assessment framework in

us

application cases to verify its operability and feasibility. It is suggested that applying

an

the DLCA framework to some residences and predicting their EI over the whole life cycles to make the DLCA model a useful tool in applications. (2) The research on

M

variation of occupancy behaviors is far from perfect. More work should be done to

d

identify comprehensive occupancy profiles in different types of buildings (e.g., office

Ac ce pt e

buildings, hospitals, teaching buildings), and then better explore their influences on energy consumption. (3) There is a need to pay attention to studies of dynamic CFs and WFs of various impact categories in different time horizons. Once they are mature enough, we could involve dynamic CFs and WFs in our assessment.

Acknowledgements

This work was financially supported by the National Science Foundation of China (No. 51378297 and No. 51078200) and Tsinghua Fudaoyuan Research Fund. The authors would like to thank Prof. Zhihui Zhang, Mr. Xiaomin Yang, and Mr. Xiangqin Kong in the Department of Construction Management at Tsinghua

Page 34 of 45

University for their previous work, and Dr. Nan Li at Tsinghua University for his patient guidance.

ip t

References [1] National Bureau of Statistics of China, China Energy Statistical Yearbook

cr

(2000-2011), Statistics Press, Beijing [In Chinese].

[2] U.S. Department of Energy, 2010 Buildings Energy Data Book, 2010.

us

[3] B.C. Lippiatt, BEESRG 4.0: Building for Environmental and Economic Sustainability Technical Manual and User Guide, National Institute of Standards

an

and Technology, Gaithersburg, MD, 2007.

[4] B. Peuportier, K. Putzeys, J. Anderson, et al., Inter-comparison and

Package 2, 2005.

M

benchmarking of LCA-based environmental assessment and design tools, Work

d

[5] D. Kellenberger, H. Althaus, Relevance of simplifications in LCA of building

Ac ce pt e

components, Build. Environ. 44 (2009) 818-825.

[6] EcoQuantum, Homepage of EcoQuantum, http://www.ivam.uva.nl/index.php?id=2&L=1, accessed on 13th Oct, 2016.

[7] Z. Zhang, X. Wu, X. Yang, et al., BEPAS—a life cycle building environmental performance assessment model, Build. Environ. 41(2006) 669-675.

[8] X. Li, Y. Zhu, Z. Zhang, An LCA-based environmental impact assessment model for construction processes, Build. Environ. 45(2010) 766-775.

[9] S. Su, X. Li, T. Wang, et al., A comparative study of environmental performance between CFST and RC columns under combinations of compression and bending, J. Cleaner Prod. 137 (2016) 10-20. [10] X. Cao, X. Li, Y. Zhu, et al., A comparative study of environmental performance between prefabricated and traditional residential buildings in China, J. Cleaner Prod. 109 (2015) 131-143.

Page 35 of 45

[11] X. Li, S. Su, Z. Zhang, et al., An integrated environmental and health performance quantification model for pre-occupancy phase of buildings in China, Environ. Impact Assess. Rev. 63 (2017) 1-11. [12] X. Kong, Research on the health damage assessment model of building during

ip t

the life cycle, Tsinghua University, Beijing, 2010 [in Chinese]. [13] G. Finnveden, M.Z. Hauschild, T. Ekvall, et al., Recent developments in life

cr

cycle assessment, J. Environ. Manage. 91(2009) 1-21.

[14] J. Reap, F. Roman, S. Duncan, et al., A survey of unresolved problems in life

us

cycle assessment, Int. J. Life Cycle Assess. 13(2008) 374-388.

[15] W.O. Collinge, A.E. Landis, A.K. Jones, et al., Dynamic life cycle assessment:

an

framework and application to an institutional building, Int. J. Life Cycle Assess. 18 (2013) 538-552.

M

[16] V.P. Shah, R.J. Ries, A characterization model with spatial and temporal resolution for life cycle impact assessment of photochemical precursors in the

d

United States, Int. J. Life Cycle Assess. 14 (2009) 313-327.

Ac ce pt e

[17] G. Sonnemann, B. Vigon, M. Rack, et al., Global guidance principles for life cycle assessment databases: development of training material and other implementation activities on the publication, Int. J. Life Cycle Assess. 18 (2013) 1169-1172.

[18] ABBD Faria, M. Spérandio, A. Ahmadi, et al., Evaluation of new alternatives in wastewater treatment plants based on Dynamic Modelling and Life Cycle

Assessment (DM-LCA), Water Res. 84 (2015) 99-111.

[19] R.A.P. Filleti, D.A.L. Silva, E.J. Silva, et al., Dynamic system for life cycle inventory and impact assessment of manufacturing processes, Procedia CIRP, 15 (2014) 531-536. [20] A. Levasseur, P. Lesage, M. Margni, et al., Considering time in LCA: Dynamic LCA and its application to global warming impact assessments, Environ. Sci. Technol. 44 (2010) 3169-3174.

Page 36 of 45

[21] M. Guo, R.J. Murphy, LCA data quality: sensitivity and uncertainty analysis, Sci. Total Environ. 435 (2012) 230-243. [22] B. Laratte, B. Guillaume, Epistemic and methodological challenges of dynamic environmental assessment: A case-study with energy production from solar cells,

ip t

Key Eng. Mater. 572 (2014) 535-538. [23] C. Davis, Integration of life cycle analysis within agent based modeling using a

cr

case study on bio-electricity, Leiden University, Leiden, 2007.

[24] J. Reap, F. Román, T. Guldberg, et al., Integrated Ecosystem Landscape and Modeling

for

Strategic

Environmentally

Conscious

Process

us

Industrial

Technology Selection, Proceedings of the 2006 13th CIRP International

an

Conference on Life Cycle Engineering, Leuven, 2006.

[25] B. Ness, E. Urbel-Piirsalu, S. Anderberg, et al., Categorising tools for

M

sustainability assessment, Ecol. Econ. 60 (2007) 498-508 [26] ISO, Environmental Management: Life Cycle Assessment: Requirements and

d

Guidelines ISO-14044, British Standards Institution, London, 2006.

Ac ce pt e

[27] ISO, Environmental Management: Life Cycle Assessment: Principles and Framework ISO-14040, British Standards Institution, London, 2006.

[28] X. Li, S. Su, Y. Gao, et al., Environmental impacts of residential buildings having different structures, Journal of Tsinghua University, 9 (2013) 1255-1260

[in Chinese].

[29] X. Li, S. Su, J. Shi, et al., An environmental impact assessment framework and index system for the pre-use phase of buildings based on distance-to-target

approach, Build. Environ. 85 (2015) 173-181.

[30] T. Hong, S.K. Chou, T.Y. Bong. Building simulation: an overview of developments and information sources, Build. Environ. 35(2000) 347-361. [31] C. Peng, D. Yan, R. Wu, et al., Quantitative description and simulation of human behavior in residential buildings, Build. Simul. 5 (2012) 85-94. [32] S. Wei, R. Jones, P. de Wilde, Driving factors for occupant-controlled space

Page 37 of 45

heating in residential buildings, Energy Build. 70 (2014) 36-44. [33] Z. Yu, B.C. Fung, F. Haghighat, et al., A systematic procedure to study the influence of occupant behavior on building energy consumption, Energy Build. 43(2011) 1409-1417.

Building, Tsinghua University, Beijing, 2014 [in Chinese].

ip t

[34] C. Wang, Simulation Research on Occupant Energy-related Behaviors in

cr

[35] S. Chen, W. Yang, H. Yoshino, et al., Definition of occupant behavior in

residential buildings and its application to behavior analysis in case studies,

us

Energy Build. 104 (2015) 1-13.

[36] E.V. Verhoef, M.A. Reuter, A. Scholte, A dynamic LCA model for assessing the

an

impact of lead free solder, Yazawa International Symposium on Metallurgical and Materials Processing: Principles and Technologies, (2003) 605-623.

M

[37] M. Pehnt, Dynamic life cycle assessment (LCA) of renewable energy technologies, Renewable energy 31(2006) 55-71.

d

[38] Y. Qi, H. Huang, G. Liu, et al., Energy Analysis Method Based on Dynamic Life

Ac ce pt e

Cycle, Chinese Journal of Mechanical Engineering 43 (2007) 129-134 [in Chinese].

[39] W. Zhao, Sustainable Municipal Solid Waste Management Based on Quasi-dynamic Eco-efficiency, Tianjin University, Tianjin, 2009 [in Chinese].

[40] W.O. Collinge, L. Liao, H. Xu, et al., Enabling dynamic life cycle assessment of buildings with wireless sensor networks, IEEE International Symposium on Sustainable Systems and Technology, IEEE (2011)1-6.

[41] A. Levasseur, P. Lesage, M. Margni, et al., Assessing temporary carbon sequestration and storage projects through land use, land-use change and forestry: comparison of dynamic life cycle assessment with ton-year approaches, Clim. Change 115 (2012) 759-776. [42] A. Levasseur, P. Lesage, M. Margni, et al., Biogenic carbon and temporary storage addressed with dynamic life cycle assessment, J. Ind. Ecol. 17 (2013)

Page 38 of 45

117-128. [43] W.O. Collinge, A.E. Landis, A.K. Jones, et al., Indoor environmental quality in a dynamic life cycle assessment framework for whole buildings: Focus on human health chemical impacts, Build. Environ. 62 (2013)182-190.

ip t

[44] W.O. Collinge, A.E. Landis, A.K. Jones, et al., Productivity metrics in dynamic LCA for whole buildings: using a post-occupancy evaluation of energy and

cr

indoor environmental quality tradeoffs, Build. Environ. 82 (2014) 339-348.

[45] J. Yang, B. Chen, Global warming impact assessment of a crop residue

us

gasification project-A dynamic LCA perspective, Appl. Energy 122 (2014) 269-279.

an

[46] H. Li, Carbon Emissions Dynamic Characteristics of Mechanical Manufacturing System and Its Carbon Efficiency Assessmemt and Optimization Approach

M

Research, Chongqing University, Chongqing, 2014 [in Chinese]. [47] F. Lebailly, A. Levasseur, R. Samson, et al., Development of a dynamic LCA

d

approach for the freshwater ecotoxicity impact of metals and application to a

Ac ce pt e

case study regarding zinc fertilization, Int. J. Life Cycle Assess. 19 (2014) 1745-1754.

[48] M. Fouquet, A. Levasseur, M. Margni, et al., Methodological challenges and developments in LCA of low energy buildings: Application to biogenic carbon and global warming assessment, Build. Environ. 90 (2015) 51-59.

[49] C. Ji, T. Hong, New Internet search volume-based weighting method for integrating various environmental impacts, Environ. Impact Assess. Rev. 56 (2016) 128-138.

[50] M. Ken’Ichi, Energy Structure and Energy Security under Climate Mitigation Scenarios in China, PloS one, 10 (2015) e0144884. [51] Z. Wang, Y. Zhu, Y. Zhu, et al., Energy structure change and carbon emission trends in China, Energy 115 (2016) 369-377. [52] Editorial Committee, China Power Yearbook. China Electric Power Press,

Page 39 of 45

Beijing, 2016. [53] L Gu, Studies on the Environmental Impact of the Building Industry in China based on the Life Cycle Assessment, Tsinghua University, Beijing, 2009 [in Chinese].

ip t

[54] Z. Li, Study on the Life Cycle Consumption of Energy and Resource of Air Conditioning in Urban Residential Buildings in China, Tsinghua University,

[55]

National

Development

and

Reform

cr

Beijing, 2007 [in Chinese]. Commission,

Guidance

on

the

us

comprehensive utilization of resources in the 13th Five-Year, China Environmental Sciences Press, Beijing, 2016 [In Chinese].

an

[56] The State Council of the People’s Republic of China, Energy Development Strategy Plan (2014-2020), 2014 [in Chinese].

M

[57] K. Andriosopoulos, Energy security in East Asia under climate mitigation scenarios in the 21st century, Omega, 59 (2016) 60-71.

d

[58] Q. Wang, P. Liu, X. Yuan, et al., Structural evolution of household energy

Ac ce pt e

consumption: a China study, Sustainability 7(2015) 3919-3932. [59] L. Xue, Y. Hou, X. Yan, et al., Chinese energy consumption structure prediction by application of ARIMA, China Mining Magazine 20 (2011) 24-27 [in Chinese].

[60] R. Yang, L. Wang, Development of multi-agent system for building energy and comfort management based on occupant behaviors, Energy Build. 56 (2013) 1-7.

[61] D. Liu, Y. Yang, K. Yang, et al., Forecasting model and its application of energy structure and pollutant emission based on Markov chain, Electric Power 39(2006) 8-13. [62] R. Yang, L. Wang L, Development of multi-agent system for building energy and comfort management based on occupant behaviors, Energy Build. 56(2013) 1-7. [63] W. Poortinga, L. Steg, C. Vlek, et al., Household preferences for energy-saving

Page 40 of 45

measures: A conjoint analysis, J. Econ. Psychology 24 (2003) 49-64. [64] M. Bonte, F. Thellier, B. Lartigue, Impact of occupant's actions on energy building performance and thermal sensation, Energy Build. 76 (2014) 219-227. [65] T.S. Blight, D.A. Coley, Sensitivity analysis of the effect of occupant behaviour

ip t

on the energy consumption of passive house dwellings, Energy Build. 66 (2013) 183-192.

cr

[66] Z. Li, Y. Jiang, Q. Wei, Survey on energy consumption of air conditioning in summer in a residential building in Beijing, Journal of HV&AC 37 (2007) 46-51

us

[in Chinese].

[67] Architectural Institute of Japan, Japanese residential energy consumption online http://tkkankyo.eng.niigata-.ac.jp/HP/HP/database/japan2/index.htm,

an

database,

accessed on 24th Oct. 2016.

M

[68] A. Al-Mumin, O. Khattab, G. Sridhar G, Occupants’ behavior and activity patterns influencing the energy consumption in the Kuwaiti residences, Energy

d

Build. 35(2003) 549-559.

Ac ce pt e

[69] L. Lopes, S. Hokoi, H. Miura, et al., Energy efficiency and energy savings in Japanese residential buildings--research methodology and surveyed results, Energy Build. 37(2005) 698-706.

[70] J. Ouyang, K. Hokao, Energy-saving potential by improving occupants’ behavior in urban residential sector in Hangzhou City, China, Energy Build. 41(2009)

711-720.

[71] J. Zhao, B. Lasternas, K.P. Lam, et al., Occupant behavior and schedule modeling for building energy simulation through office appliance power consumption data mining, Energy Build. 82 (2014) 341-355. [72] R.V. Andersen, J.O.R. Toftum, K.K. Andersen, et al., Survey of occupant behaviour and control of indoor environment in Danish dwellings, Energy Build. 41(2009) 11-16. [73] Q. Pu, Research on Prediction Model and Influencing Factors of Urban

Page 41 of 45

Residential Building Energy Consumption, Chongqing University, Chongqing, 2012 [in Chinese]. [74] J. Yao, Domestic Energy Consumption Behavior Study, Energy Research and Utilization 4 (2009) 7-12 [in Chinese].

ip t

[75] X. Peng, Research on Model of Energy Consumption Terminal Constraints in Residential Buildings, Chongqing University, Chongqing, 2009 [in Chinese].

University of Technology, Dalian, 2009 [in Chinese].

cr

[76] L. Chen, Study on Urban Residential Energy Consumption Behavior, Dalian

us

[77] A. Pang, Analysis of Influential Factors of Urban Residents' Energy Use, Dalian University of Technology, Dalian, 2011 [in Chinese].

an

[78] W. Abrahamse, L. Steg, How do socio-demographic and psychological factors relate to households’ direct and indirect energy use and savings? J. Econ.

M

Psychology 30 (2009) 711-720.

[79] A. Druckman, T. Jackson, Household energy consumption in the UK: A highly

Ac ce pt e

(2008) 3177-3192.

d

geographically and socio-economically disaggregated model, Energy Policy 36

[80] J. Chen, X. Wang, K. Steemers, A statistical analysis of a residential energy consumption survey study in Hangzhou, China, Energy Build. 66 (2013)

193-202.

[81] H. Liao, T. Chang, Space-heating and water-heating energy demands of the aged in the US, Energy Econ. 24 (2002) 267-284.

82] D. Mora, C. Carpino, M. De Simone, Behavioral and physical factors influencing energy building performances in Mediterranean climate, Energy Procedia 78 (2015) 603-608. [83] L. Du, T. Prasauskas, V. Leivo, et al., Assessment of indoor environmental quality in existing multi-family buildings in North-East Europe, Environment International 79 (2015) 74-84. [84] V. Fabi, R.V. Andersen, S. Corgnati, et al., Occupants' window opening

Page 42 of 45

behaviour: A literature review of factors influencing occupant behaviour and models, Build. Environ. 58(2012) 188-198. [85] I. Nahmens, A. Joukar, R. Cantrell. Impact of low-income occupant behavior on energy cConsumption in hot-humid climates, J. Archit. Eng. 21(2014)

ip t

B4014006. [86] L. Steg, C. Vlek, Encouraging pro-environmental behaviour: an integrative

cr

review and research agenda, J. Environ. Psychology 29 (2009) 309-317.

[87] Q. Pu, B. Li, Impact factors analysis of residential buildings' energy consumption

us

in Huainan, International Conference on Remote Sensing, Environment and Transportation Engineering. IEEE (2011) 1278-1281.

an

[88] H. Zhao, F. Magoulès, A review on the prediction of building energy consumption, Renewable Sustainable Energy Rev. 16 (2012) 3586-3592.

M

[89] D.B. Crawley, L.K. Lawrie, F.C. Winkelmann, et al., EnergyPlus: creating a new-generation building energy simulation program, Energy Build. 33(2001)

d

319-331.

Ac ce pt e

[90] J.A. Clarke, D. McLean. ESP-A building and plant energy simulation system, Strathclyde: Energy Simulation Research Unit, University of Strathclyde, 1988.

[91] D. Yan, J. Xia, W. Tang, et al., DeST-An integrated building simulation toolkit Part I: Fundamentals, Build. Simul. 1 (2008) 95-110.

[92] J. Virote, R. Neves-Silva, Stochastic models for building energy prediction based on occupant behavior assessment, Energy Build. 53 (2012) 183-193.

[93] J. Chen, R.K. Jain, J.E. Taylor, Block Configuration Modeling: a novel simulation model to emulate building occupant peer networks and their impact on building energy consumption, Appl. Energy 105 (2013) 358-368. [94] A. Heydarian, J.P. Carneiro, D. Gerber, et al., Immersive virtual environments versus physical built environments: A benchmarking study for building design and user-built environment explorations, Autom. Constr. 54 (2015) 116-126. [95] D. Simeone, Y.E. Kalay, D. Schaumann, et al., An event-bBased model to

Page 43 of 45

simulate human behaviour in built environments, Digital Physicality: Proceedings of the 30th eCAADe Conference (2012) 525-532. [96] R. Goldstei, A. Tessier, A. Khan, Space layout in occupant behavior simulation, Conference Proceedings: IBPSA-AIRAH Building Simulation Conference (2011)

ip t

1073-1080. [97] M.F. Shiratuddin, W. Thabet, D. Bowman, Evaluating the effectiveness of virtual

cr

environment displays for reviewing construction 3D models, CONVR 2004, (2004) 87-98.

us

[98] D.J. Sailor. A green roof model for building energy simulation programs, Energy Build. 40 (2008) 1466-1478.

an

[99] Y. Pan, Z. Huang, G. Wu, Calibrated building energy simulation and its application in a high-rise commercial building in Shanghai, Energy Build. 39

M

(2007) 651-657.

[100] A. Heydarian, J.P. Carneiro, D. Gerber, et al., Immersive virtual environments,

d

understanding the impact of design features and occupant choice upon lighting

Ac ce pt e

for building performance, Build. Environ. 89 (2015) 217-228. [101] V. Rosalie, A. Mark, A. Hans, et al., Time horizon dependent characterization factors for acidification in life-cycle assessment based on forest plant species occurrence in Europe, Environ. Sci. Technol. 41 (2007) 922-927.

[102] J. Seppälä, Country-dependent characterisation factors for acidification and terrestrial eutrophication based on accumulated exceedance as an impact category indicator, Int. J. Life Cycle Assess. 11 (2006) 403-416.

[103] J. Struijs, A.V. Dijk, H. Slaper, et al., Spatial- and time-explicit human damage modeling of ozone depleting substances in life cycle impact assessment, Environ. Sci. Technol. 44 (2010) 204-209. [104] A. Meijer, M. Huijbregts, L. Reijnders. Human health damages due to indoor sources of organic compounds and radioactivity in life cycle impact assessment of dwellings - Part 1: characterisation factors, Int. J. Life Cycle Assess. 10 (2005)

Page 44 of 45

309-316. [105] X. Wu, Z. Zhang, Y. Chen, Study of the environmental impacts based on the “green tax”—applied to several types of building materials, Build. Environ. 40

Highlights:

cr

A Dynamic LCA Framework for EIA of Buildings is proposed.

ip t

(2005) 227-237.

us

Progress in DLCA research is reviewed and the limitations are summarized. Four dynamic properties are identified and introduced into the DLCA framework.

an

The impacts on EIA by technological progress are considered from three aspects.

Ac ce pt e

d

M

Occupancy profiles in residences are described from three levels

Page 45 of 45