- Email: [email protected]

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Incorporating life cycle assessments into building project decision-making: An energy consumption and CO2 emission perspective Wen-Hsien Tsai a, *, Sin-Jin Lin a, Jau-Yang Liu b, Wan-Rung Lin a, Kuen-Chang Lee a a b

Department of Business Administration, National Central University, 300, Jhongda Road, Jhongli, Taoyuan 32001, Taiwan Department of Accounting, Chinese Culture University, Yang-Ming-Shan, Taipei 11192, Taiwan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 September 2010 Received in revised form 20 February 2011 Accepted 23 February 2011 Available online 31 March 2011

In the past two decades, the globalization of ﬁnancial markets and multinational trade has intensiﬁed internationally, and become increasingly competitive. In the construction industry, critical changes are initiated to reduce operating costs for achieving sustainable operation. Conventional cost pricing for building projects no longer apply as energy shortage and environmental pollution are new challenges faced by construction companies. Many countries have attempted to solve the CO2 emission problems by levying a carbon tax, which leads to a higher cost for construction companies. Therefore, this study aims to adopt life cycle assessment (LCA) in order to assess CO2 emission costs and apply a mathematical programming approach to allocate limited resources to maximize proﬁts for construction companies. 2011 Elsevier Ltd. All rights reserved.

Keywords: CO2 emission cost Life cycle assessment Project lifecycle Project management Mathematical programming

1. Introduction In the past two decades, the globalization of ﬁnancial markets and multinational trade has intensiﬁed internationally, and become increasingly competitive. In the construction industry, critical changes are initiated to eliminate operating costs for achieving sustainable operation. The emission of greenhouse gases (GHGs), particularly CO2 emission [1], has posed a serious problem on the global climate system. The construction industry is responsible for 40% of the primary energy use, and 36% of energy related CO2 emissions in industrialized countries [2], because the production of materials and construction process can signiﬁcantly increase atmospheric concentrations of GHGs [3]. Many countries have attempted to solve the CO2 emission problems by levying a “Carbon tax” in order to increase the operating cost of high-energy intensive ﬁrms. Zhang and Baranzini [4] argued that a carbon or energy tax produces “winners” and “losers”, as the different relative impacts on production costs are imposed on both low- and high-energy intensive ﬁrms. Sathre and Gustavsson [5] suggested that environmental taxation may act as an economic incentive to overcome organizational inertia, encouraging ﬁrms to adopt innovations that result in both lower environmental impact and increased economic

* Corresponding author. Tel.: þ886 3 4250 860; fax: þ886 3 4222 891. E-mail address: [email protected] (W.-H. Tsai). 0360-5442/$ e see front matter 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.energy.2011.02.046

beneﬁts. However, as opposed to other well-examined ﬁelds, such as cost pricing of building projects [6], research on the considerations of carbon tax incorporated with project management is minimal. Therefore, this paper takes CO2 emission costs into consideration in order to help construction companies maximizing their proﬁts. Many previous studies suggest that the life cycle assessment (LCA) is a powerful and internationally accepted system analysis tool that measures energy efﬁciency and energy conservation assessments throughout material life cycles [7,8], as acknowledged by a growing number of studies [1,9e11]. Hence, this study applied the LCA method to measure energy consumption and CO2 emissions, in order to gain a better understanding of energy use during construction process, and determine CO2 emissions over the life cycle of building projects. Hoinka and Ziebik [12] applied a mathematical approach to assess the energy management of complex buildings, and pointed out that energy management is an essential problem of complex buildings. Very few researches have simultaneously adopted a mathematical programming approach and LCA method for construction companies to maximize proﬁts on building projects. The contribution of this study is that it incorporates CO2 emission costs into mathematical programming, thus allowing construction companies to evaluate CO2 emissions of building projects. The remainder of this paper is organized as follows: Section 2 introduces the background of the carbon tax, CO2 emissions, and

W.-H. Tsai et al. / Energy 36 (2011) 3022e3029

the LCA method used in this study. Section 3 presents the proposed model adopted mathematical programming methods. Section 4 discusses a numerical example to demonstrate the proposed model. Finally, the conclusions and future developments are addressed in Section 5. 2. Research background 2.1. Carbon tax and CO2 emission costs Energy shortages and environmental pollutions have become major technological, societal, and political challenges around the globe [8,13], and several researches have addressed these problems. Lee et al. [14] proposed that applications of price mechanisms are important instruments for carbon reduction, among which the carbon tax has been frequently advocated as a cost-effective economic tool. They also showed that some European countries, such as the Netherlands, Denmark, Sweden, Finland, and Norway, have implemented carbon taxes for over 10 years, while Italy, Germany, and the UK also began to levy carbon taxes between 1999 and 2001. Herber and Raga [15], and Sathre and Gustavsson [5] proposed that carbon and energy taxation would increase ﬁnancial incentives to reduce emissions of carbon into the atmosphere, thus, combating global warming. With increasing concerns about ecological preservation since the late 1980s, building energy efﬁciency has been under serious considerations [8,16,17]. The construction industry, as one of the fastest growing industry in terms of energy consumption, is responsible for 40% of the primary energy use, and 36% of the energy related CO2 emissions in industrialized countries [2,18]. Thus, building energy is an important issue, as energy is one of the most critical resources used over the lifetime of a building [19]. For construction companies, higher CO2 emissions would lead to higher carbon taxes, which would result in higher building costs. Related research focuses on two areas, namely, the exploration of the reduction of energy consumption in building environments [20e23], and the examination of the effects of carbon taxes on building competitiveness [5,24]. However, as opposed to the above well-examined researches, studies on incorporating CO2 emission costs with building project costs are minimal. Upon this focus, by taking CO2 emission costs into consideration in building projects, the ﬁndings of this study can serve as a reference to construction companies in decision-making. 2.2. Application of LCA in the building project Development of modern evaluation methods that apply the LCA method in building energy conservation assessments have become a trend [8,10,25e28]. Many previous studies suggest that LCA is a powerful and internationally accepted system analysis tool, which studies environmental aspects and potential impacts of a product or service system throughout its life cycle [8,29e32]. This study attempts to apply LCA method in evaluating energy use and carbon emissions created during construction processes. Given the complexities of interactions between construction processes and natural environments, LCA represents a comprehensive approach to examine the environmental impacts of an entire building project. Building projects include the sub-division of phases and terms, all of which cause environmental impacts, such as materials production, transportation, construction wastes, pollutants, and materials consumption [8,11]. Ward and Chapman [33] proposed that a project lifecycle is commonly divided into four phases, namely, conceptualization, planning, execution, and termination, where both the level of resources employed and the rate of

3023

expenditures are very different in each phase. In addition, the majority of expenditures occur within the execution phase [33]. Therefore, this study is divided into three phases in order to illustrate the energy consumption and CO2 emission costs of building projects: (1) Design and planning phase (conceptualization and planning): this phase causes little energy consumption (e.g., burdens from electricity used for lighting) [8,11]. (2) Construction phase (execution): the phase causes the majority of energy used (e.g., the production and transportation of building materials; diesel fuel used by heavy equipment; burdens from electricity used for power tools and lighting) [9,23]. (3) Delivery and maintenance phase (termination): this phase causes some end-of-life energy consumption (e.g., disposal the waste treatment by burdens from electricity used for power tools and lighting; diesel fuel used by heavy equipment) [1,7]. 2.3. Summary LCA is internationally acknowledged as a science-based, fairly comprehensive, and standardized environmental assessment methodology, which is used in several sectors, including the construction industry, with a wide range of applications [7]. One of the most important extensions of related research is in the area of energy consumption and CO2 emissions [1,8e11]. The LCA method can acquire a comprehensive view of a project’s entire-life environmental cost, which implies that the environmental and social costs (e.g., CO2 emission costs) of all phases in the building project life cycle are assessed. Therefore, this study applied the LCA method to analyze the energy consumption and CO2 emission costs in a building project. The calculation processes of building project costs are brieﬂy described as follows. 3. Model formulation e assessing building project costs Project costs must be determined in a relatively short time by project managers as a reference for evaluating competitive bids [6]. The information regarding all cost items on project bids must be known and assessed in advance in order for managers to make accurate judgments. Conventional construction costs include material costs, labor, and equipment cost, but exclude value added tax [5], which would lead to improper measurement. The reason is that CO2 emission costs are critical, and have become one of the major cost items in recent years. In order to address related problems and accurately calculate the total costs of a building project, this study classiﬁed these cost items into four categories [2,5,22,34], namely (1) materials costs: including the cost of raw materials and goods purchased from other categories of the industry; (2) labor costs: including the personnel expenses and other added costs; (3) machine costs: including equipment for operating and completing building projects; and (4) environmental and social costs: including the costs of environmental pollution, and paying carbon taxes. This study only considered the above costs, while regarded other costs as unchangeable ﬁxed costs. A ﬂowchart is used to illustrate the costs of a building project, as shown in Fig. 1. This study was divided into two main stages; the ﬁrst focuses on applying the LCA method to assess CO2 emission costs, while the second focuses on incorporating the above costs (e.g., direct material costs, direct labor costs, direct machine costs, and CO2 emission costs) through a mathematical programming approach in order to identify the optimal building project. This study is a pioneer in incorporating CO2 emission costs into building project costs, and offers construction companies comprehensive considerations in decision-making.

3024

W.-H. Tsai et al. / Energy 36 (2011) 3022e3029

Bid priceBuilding project A

Bid priceBuilding project B

Bid priceBuilding project C

Bid priceBuilding project D

Bid priceBuilding project E

Building project costs

Direct material

Direct labor

Machine cost

Direct cost (building costs)

CO2 emission cost

Life cycle assessment (LCA)

Select project with maximize profit for a construction company

Fig. 1. Building project cost calculation stages.

3.1. Assumptions Suppose a construction company considers CO2 emission costs in a building project. The following assumptions are incorporated into the proposed mathematical programming model: (1) the project bidding prices and direct material costs for existing projects remain ﬁxed within a certain relevant range in the short term; for some materials, a purchase discount will be offered if the order exceeds a minimum quantity; (2) this model does not consider the substitution of resources in the building project, such as the replacement of labor hours by machine hours; (3) the company has established good relationships with its vendors, thus, renting additional machines from vendors can extend the machine hour resources in the short term; (4) conforming to governmental policies, the direct labor resources can be expanded using overtime work or additional night shifts, as well as hiring workers at a higher wage rate in the short term; (5) management can acquire accurate cost information from the company’s ﬁnancial department in order to formulate policies for maximizing its proﬁts; moreover, the company executes projects based on aggregate market demands and related costs; (6) the potential competitors’ bidding price is not sensitive to temporary price changes in the short term because each company has its own target customers and pricing policy; (7) the increasing CO2 emissions will enhance taxation; (8) direct materials, direct labor, machine costs, and CO2 emission costs are taken into consideration; however, other costs are viewed as unchangeable ﬁxed costs, and are thus excluded from this study. 3.2. A mathematical programming model According to the above, there are four components in building project costs, namely materials costs, labor costs, machine costs, and CO2 emission costs. The following is a discussion on purchase discounts, capacity expansions, capacity constraints, and carbon costs, which are incorporated into the mathematical programming model for determining the optimal project decision. 3.2.1. Total revenue P The terms in the ﬁrst set of parentheses in Eq. (1), i.e., ni¼ 1 Pi Xi , represent the total revenue of bid prices for building projects. 3.2.2. Total direct material cost The terms in the second set of parentheses in Eq. (1), i.e., P P lDr RMr Þ are the total direct material costs, with ð sr ¼ 1 lr Rr þ

example, the vendor of a material offers a purchase discount of 10% for purchases that exceed Rr. Hence, the total material cost function would be a piecewise linear function, as shown in Fig. 2. The available normal material quantity is Rr and the material quantity can be expanded to RMr; the total material cost is lr and lDr at Rr and RMr, respectively. Eqs. (2)e(9) are the constraints associated with various kinds of materials. For a material with a purchase discount condition of ðr˛DÞ, Eq. (4) describes that the quantity of this material either qualiﬁes or disqualiﬁes for a purchase discount, and should satisfy the necessary amount for building each project. Eqs. (5) and (6) describe the conditions that a purchase discount either qualiﬁes or disqualiﬁes. Eq. (7) sets a maximum quantity of a material with a purchase discount that can be ordered. Finally, Eq. (8) ensures that one, and only one, of the conditions described by Eqs. (5) and (6) is in effect for each material. 3.2.3. Total direct labor cost This paper assumes that direct labor resources can be expanded using overtime work or additional night shifts, and hiring workers at a higher wage rate. Thus, the direct labor hours are separated into two parts, namely, non-discretionary labor hours and discretionary labor hours, as shown in Fig. 3. The available normal direct labor hours are LH1, and the direct labor hours can be expanded to LH2 and LH3; the total direct labor cost is LC1, LC2, and LC3 at LH1, LH2, and LH3, respectively. The terms in the third set of parentheses in Eq. (1), i.e.,LC1 þ ðLC2 LC1 Þa1 þ ðLC3 LC1 Þa2 , represent the total direct labor costs of building projects, where LC1 represents the cost of non-discretionary labor hours, and ðLC2 LC1 Þa1 þ ðLC3 LC1 Þa2 represents the sum of the cost of discretionary labor hours and overtime work. Eqs. (10)e(15) are the constraints associated with direct labor. TL, in Eq. (10), is the total direct labor hours required. In Eqs. (11)(15), ðh1 ; h2 Þ is an SOS1 set of 0e1 variables, within which exactly one variable must be non-zero; ða0 ; a1 ; a2 Þ is an SOS2 set of nonnegative variables, within which there are at most two adjacent variables in the order given to the set, that can be non-zero [35,36]. If h1 ¼ 1, then h2 ¼ 0 [from Eq. (15)], a2 ¼ 0 [from Eq. (13)], a0 ; a1 1 [from Eqs. (11) and (12)], and a0 þ a1 ¼ 1 [from Eq. (14)]. Thus, the total direct labor hours required, and the total labor costs are LH1 þ ðLH2 LH1 Þa1 and LC1 þ ðLC2 LC1 Þa1 , respectively, indicating that there will be overtime work. On the other hand, if h2 ¼ 1, then h1 ¼ 0 [from Eq. (15)], a0 ¼ 0 [from Eq. (11)], a1 ; a2 1 [from Eqs. (12) and (13)], and a1 þ a2 ¼ 1 [from Eq. (14)]. Then, the total direct labor hours required, and total labor costs are LH1 þ ðLH2 LH1 Þa1 þ ðLH3 LH1 Þa2 and LC1 þ ðLC2 LC1 Þa1 þ ðLC3 LC1 Þa2 , respectively, indicating that there will be overtime work and some additional workers will be hired. 3.2.4. Total direct machine cost In this paper, the total machine cost is regarded as a common ﬁxed cost, and it is assumed that its cost function is a stepwise function (as show in Fig. 4), which varies with machine hours, as Cost

lDr lr

0

Rr

RM r

r˛D

ðr˛DÞ and without ðr;DÞ purchase discount, respectively. For

Fig. 2. Direct material costs.

Material quantity

W.-H. Tsai et al. / Energy 36 (2011) 3022e3029

3025

Cost

Cost

COP 3

LC 3

COP 2

LC 2 LC1

COP1

0

LH 1

LH2 LH3

0

Labor hour

COQ 1

observed from prior cost behavior analysis. The total machine cost is MC0 under the current capacity of MH0 machine hours. If the capacity is successively expanded to MH1, MH2, ., MHt machine hours, then the total machine cost increases to MC1, MC2, ., MCt, respectively. The terms in the fourth set of parentheses in Eq. (1), Pt i.e., k ¼ 0 MCk qk , represent the total direct machine costs of building projects. Eqs. (16) and (17) are the constraints associated with direct machine costs. Let lih be the requirement for machine hours for project i, where ðq0 ; q1 ; .; qt Þ is an SOS1 set of 0e1 variables, within which exactly one variable must be non-zero [35,36]. When qZ ¼ 1ðzs0Þ, the capacity must be expanded to the zth level, i.e. MHz machine hours. 3.2.5. Environmental and social cost d total CO2 emission cost In order to obtain the CO2 emission costs, it is necessary to quantize the amount of CO2 emissions in a building project. The CO2 emissions are measured by the LCA method. Eq. (18) is used to quantize the CO2 emissions [8,11,33,37].

X

COQ 3

CO2 emission quantity (metric ton)

Fig. 5. CO2 emission costs.

Fig. 3. Direct labor costs.

TLCCO2i ¼

COQ 2

ðLCCO2D þ LCCO2C þ LCCO2M Þ

(18)

where TLCCO2i, is the amount of CO2 emission for project i. LCCO2D, LCCO2C, and LCCO2M are used to measure the quantity of CO2 emissions in the design and planning phase, construction phase, and delivery and maintenance phase, respectively. Thus, Eq. (18) is adopted to measure the amount of CO2 emissions within the entire lifecycle of a building project. With increasing CO2 emissions, taxation will be enhanced. Thus, the total CO2 emissions cost function will be a piecewise linear function, as shown in Fig. 5. The quantity of CO2 emissions can be increased from COQ1 to COQ2 and COQ3. Therefore, the total CO2 emission cost is COP1, COP2, and COP3 at COQ1, COQ2 and COQ3, respectively. The terms in the ﬁfth set of parentheses in Eq. (1), i.e., COP1 b1 þ COP2 b2 þ COP3 b3 , represent the total CO2 emission costs of building projects.

Eqs. (19)e(25) are the constraints associated with CO2 emissions. TCOC, in Eq. (19), is total CO2 emission quantities from executing building projects. In Eqs. (20)e(25), ðm1 ; m2 ; m3 Þ is an SOS1 set of 0e1 variables, within which exactly one variable must be non-zero; ðb0 ; b1 ; b2 ; b3 Þ is an SOS2 set of non-negative variables, within which at most two adjacent variables, in the order given to the set, can be non-zero [35,36]. If m1 ¼ 1, then m2 ; m3 ¼ 0 [from Eq. (25)], b2 ; b3 ¼ 0 [from Eqs. (22) and (23)], b0 ; b1 1 [from Eqs. (20) and (21)], and b0 þ b1 ¼ 1 [from Eq. (24)]. Thus, the total quantities of CO2 emissions from executing building projects, and the total CO2 emissions cost are ðCOQ 1 b1 Þ and ðCOP1 b1 Þ, respectively. This means that (1) the point ðCOQ 1 b1 ; COP1 b1 Þ is on the ﬁrst segment of the piecewise linear CO2 emission cost function, and (2) ðCOQ 1 b1 ; COP1 b1 Þ is the linear combination of (0,0) and ðCOQ 1 ; COP1 Þ. In other words, the construction company does not exceed the allowed CO2 emissions to increase the carbon tax rate. On the other hand, if m2 ¼ 1, then m1 ; m3 ¼ 0 [from Eq. (25)], b0 ; b3 ¼ 0 [from Eqs. (20) and (23)], b1 ; b2 1 [from Eqs. (21) and (22)], and b1 þ b2 ¼ 1 [from Eq. (24)]. Thus, the total quantities of CO2 emissions from executing building projects, and the total CO2 emission costs are ðCOQ 1 b1 þ COQ 2 b2 Þ and ðCOP1 b1 þ COP2 b2 Þ, respectively. This means that (1) the point ðCOQ 1 b1 þ COQ 2 b2 ; COP1 b1 þ COP2 b2 Þ is on the second segment of the piecewise linear CO2 emission cost function, and (2) ðCOQ 1 b1 þ COQ 2 b2 ; COP1 b1 þ COP2 b2 Þ is the linear combination of ðCOQ 1 ; COP1 Þ and ðCOQ 2 ; COP2 Þ. To sum up, the construction company produces excessive CO2 emissions, which would lead to a higher carbon tax rate. 3.2.6. Integrated cost models The model for building project decisions, with green costs, is as follows. Maximize:

p ¼ Total Revenue Total Direct Material Cost Total Direct Labor Cost Total Direct Machine Cost Total CO2 Emission Cost

Cost

p¼

MC t

n X

Pi Xi

i¼1

s X

lr Rr þ

r¼1

MC t −1

þðLC3 LC1 Þa2

X r˛D

t X

! lDr RMr ½LC1 þðLC2 LC1 Þa1 !

MCk qk ðCOP1 b1 þCOP2 b2

k¼0

þCOP3 b3 Þ

MC2 MC1

Subject to: (Direct material quantity constraints):

MC0

0

(1)

MH0

MH1 MH2

MHt −2 MHt −1 MHt

Fig. 4. Machine costs.

Machine hour

n X i¼1

bir Xi Rr ;

r ¼ 1; .; s; r;D

(2)

3026

W.-H. Tsai et al. / Energy 36 (2011) 3022e3029

Rr Wr ; n X

r ¼ 1; .; s; r;D

bir Xi Rr þ RMr ;

r ¼ 1; .; s; r˛D

(3)

(4)

i¼1

RMr TDr SDr ; Rr < TDr NDr ;

r ¼ 1; .; s; r˛D R ¼ 1; .; s; r˛D

RMr Wr SDr ;

r ¼ 1; .; s; r˛D

NDr þ SDr ¼ 1;

r ¼ 1; .; s

Rr 0

(5) (6) (7) (8) (9)

(Direct labor hour constraints):

TL ¼ LH1 þ ðLH2 LH1 Þa1 þ ðLH3 LH1 Þa2

(10)

a0 h1 0

(11)

a1 h1 h2 0

(12)

a2 h2 0

(13)

a0 þ a1 þ a2 ¼ 1

(14)

h1 þ h2 ¼ 1

(15)

(Machine hour constraints): n X

lih

i¼1 t X

t X

MHk qk

(16)

k¼0

qk ¼ 1

(17)

k¼0

(CO2 emission quantity constraints):

TCOC ¼ COQ 1 b1 þ COQ 2 b2 þ COQ 3 b3

(19)

b0 m1 0

(20)

b1 m1 m2 0

(21)

b2 m2 m3 0

(22)

b3 m3 0

(23)

b0 þ b1 þ b2 þ b3 ¼ 1

(24)

m1 þ m2 þ m3 ¼ 1

(25)

lDr: the unit cost of the rth materials, with a purchase discount used; Rr: the quantities of the rth materials, without a purchase discount used; RMr: the quantities of the rth materials, with a purchase discount used; bir: the requirement of the rth materials for project i; Wr: the available quantity of the rth material; TDr: the quantities of the rth materials that an order must satisfy to obtain a purchase discount; SDr: a 0e1 variable. SDr ¼ 1 means that the quantities of the rth materials satisfy the threshold of discount, otherwise, SDr ¼ 0; NDr: a 0e1 variable. NDr ¼ 1 means that the quantities of the rth materials dissatisfy the threshold of discount, otherwise, NDr ¼ 0; LC1: total direct labor cost in LH1; LC2: total direct labor cost in LH2; LC3: total direct labor cost in LH3; TL: total direct labor hours; LH1: the available normal direct labor hours; LH2: total direct labor hours, under overtime work situations (must add night shifts); LH3: total direct labor hours, under overtime work situations (must hire additional workers); h1 ; h2 : an SOS1 (special ordered set of type 1) set of 0e1 variables, within which exactly one variable must be non-zero [35,36]; a0 ; a1 ; a2 : an SOS2 (special ordered set of type 2) set of nonnegative variables, within which at most two adjacent variables, in the order given to the set, can be non-zero [35,36]; MCk: total machine cost in MHk; MHk: the available direct machine hours; lih: the requirement of machine hours for project i; qk: an SOS1 (special ordered set of type 1) set of 0e1 variables, within which exactly one variable must be non-zero [35,36], qk ¼ 1 ðks0Þ means that the capacity needs to be expanded to the kth level, i.e., MHk machine hours; COP1: total CO2 emission cost in COQ1; COP2: total CO2 emission cost in COQ2; COP3: total CO2 emission cost in COQ3; TCOC: the total quantities of CO2 emissions; COQ1: the total quantities of CO2 emissions from executing building projects; COQ2: the total quantities of CO2 emissions from executing building projects (must increase the carbon tax rate); COQ3: the total quantities of CO2 emissions from executing building projects (must increase the carbon tax rate); m1 ; m2 ; m3 : an SOS1 (special ordered set of type 1) set of 0e1 variables, within which exactly one variable must be non-zero [35,36]; b0 ; b1 ; b2 ; b3 : an SOS2 (special ordered set of type 2) set of nonnegative variables, within which at most two adjacent variables, in the order given to the set, can be non-zero [35,36]. 4. A numerical example

The following notations are used in this paper:

p: the construction company’s proﬁt Pi: the bid price of project i; Xi: a 0e1 variable. Xi ¼ 1 means that the construction company execute ith project; lr: the unit cost of the rth materials, without a purchase discount used;

In this section, a numerical example is used to test the proposed model, with details described as follows. 4.1. Data and description of a numerical example e building projects This study assumes that a construction company is considering executing project 1, project 2, project 3, project 4, and project 5 (i ¼ 1, 2, 3, 4, 5), and these projects consume the same direct

W.-H. Tsai et al. / Energy 36 (2011) 3022e3029

3027

Table 1 Example data (in thousands). Projects (i)

Bid price Direct material constraint Cost/unit lr ¼ $2.5/unit Material quantities TD1 ¼ 10,000

lDr ¼ $2.0/unit

Direct labor constraint Cost LC1 ¼ $6000 Labor hours LH1 ¼ 3000 Wage rate A1 ¼ $2/h

LC2 ¼ $11,000 LH2 ¼ 5000 A1 ¼ $2.5/h

Direct machine constraint Cost MC1 ¼ $8000 Machine hours MH1 ¼ 2000

MC2 ¼ $15,000 MH2 ¼ 5000

CO2 emission constraint Cost COP1 ¼ $100,000 Emission quantities COQ1 ¼ 2000 Tax rate T1 ¼ $50/ton

COP2 ¼ $220,000 COQ2 ¼ 4000 T2 ¼ $60/ton

LC3 ¼ $20,000 LH3 ¼ 8000 A1 ¼ $3/h

COP3 ¼ $360,000 COQ3 ¼ 6000 T2 ¼ $70/ton

materials. The vendor of materials offers a purchase discount of 20% if the amount of purchase exceeds 10,000,000. It is also assumed that these projects are each being followed up during a 24-month period. The related data for this example are presented in Table 1, where MH1 denotes the current machine capacity of 2,000,000 machine hours, and MC1 denotes the cost of $8,000,000 at this capacity level. To increase the machine capacity from MH1 to MH2, the company must lease machines from the vendor, which increases the cost to $15,000,000 (MC2). Normal direct labor hour LH1 is 3,000,000 hours, and the wage rate is $2 per hour. The direct labor hours can be expanded to LH2 ¼ 5,000,000 or LH3 ¼ 8,000,000 using overtime work or additional night shifts (the wage rate will be $2.5), and hire temporary workers at a higher wage rate in the short term (the wage rate will be $3). The carbon taxes vary depending on whether the energy is used by constructing a building project. To increase the emission quantities from COQ1 to COQ2 or COQ3, the company must pay more carbon tax, which increases the CO2 emission cost from $100,000,000 (the tax rate will be $50) to $220,000,000 (the tax rate will be $60) or $360,000,000 (the tax rate will be $70). 4.2. Three-step analysis Step 1: Determining the CO2 emission quantities of building project. Based on the assumption that management can acquire accurate energy information from the company, this step applies the LCA

Available capacity

1

2

3

4

5

Museum

School

University

Housing

Shopping center

Pi

80,000

85,000

91,500

80,000

89,000

bi1

5500

6000

9000

5000

8500

800

1100

1200

1000

1300

700

900

1000

800

1100

650

850

1200

900

1000

W1 ¼ 20,000

(from Table 2)

method to calculate the CO2 emission quantities of building projects. In addition, total CO2 emission quantities can be depicted by Eq. (18). The amount of CO2 emission that results from energy consumption, according to the LCA method, is calculated as listed in Table 2, which shows the CO2 emission quantities resulting from the ﬁve building projects. These values are 650,000 ton for the museum project, 850,000 ton for the school project, 1,200,000 ton for the university project, 900,000 ton for the housing project, and 1,000,000 ton for the shopping center project. Step 2: Determining the optimal building projects before adding the CO2 emission costs. The objective of this step is to determine the optimal building project that can maximize the company’s proﬁts, prior to adding CO2 emission costs to the mathematical programming model. Based on the data in Table 1, this step applies the proposed mathematical programming model, which is solved using LINGO software, and neglects the carbon tax. Table 3 shows the objective function, related constraints, and optimal solution prior to adding CO2 emission costs into the mathematical programming model. The resulting optimal solutions are shown in Table 3. As seen, projects 1, 3, and 4 have the highest income based on the resources used during construction. According to the results, the company would decide to execute the three building projects to make the overall proﬁt maximization for the company.

Table 2 LCA results on CO2 emission quantities of building projects. Project lifecycle

Energy use activity

Design and planning

Electric power

Construction

Manufacturing materials Transportation Construction activity Electric power

Delivery and maintain

Transportation Maintain activity Electric power

Building projects (CO2 emission quantities) Museum

Total

University

Housing

50,000

School 70,000

90,000

80,000

Shopping center 85,000

90,000 110,000 150,000 70,000

100,000 170,000 190,000 110,000

150,000 210,000 250,000 190,000

80,000 140,000 180,000 150,000

125,000 165,000 210,000 175,000

50,000 50,000 80,000

60,000 65,000 85,000

90,000 100,000 120,000

75,000 105,000 90,000

70,000 80,000 90,000

650,000

850,000

1,200,000

900,000

1,000,0000

3028

W.-H. Tsai et al. / Energy 36 (2011) 3022e3029

Table 3 Decision analysis prior to adding CO2 emission costs. Maximize p ¼ 80,000,000 X1 þ 85,000,000 X2 þ 91,500,000 X3 þ 80,000,000 X4 þ 89,000,000 X5 2.5 R1 2 RM1 6,000,000 5,000,000 a1 14,000,000 a2 8,000,000 q0 15,000,000 q1 Subject to e direct material 5,500,000 X1 þ 6,000,000 X2 þ 9,000,000 X3 þ 5,000,000 X4 þ 8,500,000 X5 R1 RM1 0 R1 0 R1 < 10,000,000 ND1 RM1 10,000,000 SD1 RM1 20,000,000 SD1 ND1 þ SD1 ¼ 1 Subject to e machine hour 700,000 X1 þ 900,000 X2 þ 1,000,000 X3 þ 800,000 X4 þ 1,100,000 X5 2,000,000 q0 5,000,000 q1 0 q0 þ q1 ¼ 1 Optimal solution is as follows: p ¼ 191,500,000, X1 ¼ 1, X2 ¼ 0, X3 ¼ 1, X4 ¼ 1, X5 ¼ 0, R1 ¼ 0, RM1 ¼ 19,500,000, ND1 ¼ 0, SD1 ¼ 1, a0 ¼ 1, a1 ¼ 0, a2 ¼ 0, q0 ¼ 0, q1 ¼ 1,

Subject to e direct labor 800,000 X1 þ 1,100,000 X2 þ 1,200,000 X3 þ þ 1,000,000 X4 þ 1,300,000 X5 3,000,000 2,000,000 a1 5,000,000 a2 0 a0 h1 0 a1 h1 h2 0 a2 h2 0 a0 þ a1 þ a2 ¼ 1 h1 þ h2 ¼ 1

h1 ¼ 1, h2 ¼ 0

Table 4 Decision analysis upon adding CO2 emission costs. Maximize p ¼ 80,000,000 X1 þ 85,000,000 X2 þ 91,500,000 X3 þ 80,000,000 X4 þ 89,000,000 X5 2.5 R1 2 RM1 6,000,000 5,000,000 a1 14,000,000 a2 8,000,000 q0 15,000,000 q1 100,000,000 b1 220,000,000 b2 360,000,000 b3 Subject to e direct material Subject to e machine hour 5,500,000 X1 þ 6,000,000 X2 þ 9,000,000 X3 þ 5,000,000 X4 þ 8,500,000 700,000 X1 þ 900,000 X2 þ 1,000,000 X3 þ 800,000 X5 R1 RM1 0 X4 þ 1,100,000 X5 2,000,000 q0 5,000,000 q1 0 q0 þ q1 ¼ 1 R1 0 R1 < 10,000,000 ND1 Subject to e CO2 emission RM1 10,000,000 SD1 650,000 X1 þ 850,000 X2 þ 1,200,000 X3 þ 900,000 RM1 20,000,000 SD1 X4 þ 1,000,000 X5 2,000,000 b1 4,000,000 ND1 þ SD1 ¼ 1 b2 6,000,000 b3 0 b0 m 1 0 Subject to e direct labor b1 m1 m2 0 800,000 X1 þ 1,100,000 X2 þ 1,200,000 X3 þ 1,000,000 X4 þ 1,300,000 b2 m2 m3 0 X5 3,000,000 2,000,000 a1 5,000,000 a2 0 a0 h1 0 b3 m 3 0 a1 h1 h2 0 b0 þ b1 þ b2 þ b3 ¼ 1 a2 h2 0 m1 þ m2 þ m3 ¼ 1 a0 þ a1 þ a2 ¼ 1 h1 þ h2 ¼ 1 Optimal solution is as follows: p ¼ 31,000,000, X1 ¼ 0, X2 ¼ 1, X3 ¼ 0, X4 ¼ 0, X5 ¼ 1, R1 ¼ 0, RM1 ¼ 14,500,000, ND1 ¼ 0, SD1 ¼ 1, a0 ¼ 1, a1 ¼ 0, a2 ¼ 0, q0 ¼ 1, q1 ¼ 0, h1 ¼ 1, h2 ¼ 0, b1 ¼ 1, b2 ¼ 0, b3 ¼ 0, m1 ¼ 0, m2 ¼ 1, m3 ¼ 0

Step 3: Determining the optimal building projects after adding CO2 emission costs. Generally, CO2 emission costs will increase building project costs, thus, decreasing the proﬁt of the company. From the company’s perspective, it must take CO2 emission costs into consideration and select the optimal building project to maximize the proﬁts. Based on Table 1, this step applies the proposed model, which adds the CO2 emission costs, using Eqs. (1)e(25), and other constraints (i.e., the carbon taxes vary depending on whether the energy is used by constructing a building project, from Table 2), in order to explore the optimal building projects. Table 4 shows the objective functions, related constraints, and optimal solution upon adding CO2 emission costs into the model, which is a 0e1 mixed-integer nonlinear programming model, and can be solved using LINGO software. In Table 4, the optimal building projects are projects 2 and 5. The company can obtain a maximal proﬁt of $31,000,000. Under this solution, machine capacity is not expanded, and neither discretionary labor hours, nor temporary workers, are used in the building projects. Prior to considering CO2 emission costs, the selected projects are 1, 3, and 4, as shown in Table 3; after considering CO2 emission costs, the selected projects are 2 and 5, as shown in Table 4. Comparing Tables 3 and 4, it is found that when considering CO2

emission costs, it leads to very different results. The implication is that the CO2 emission costs are the key factor for a construction company to select building projects. 5. Conclusions In response to the drastic climate changes due to emissions of GHGs, particularly CO2, many countries have levied the carbon tax to address the problem. As the construction industry is responsible for 40% of the primary energy use and 36% of the energy related CO2 emission in industrialized countries, the carbon tax policy imposes heavy ﬁnancial burdens on energy intensive ﬁrms, such as those in the construction industry. Thus, the analysis of building project costs is a matter of serious concern to construction companies. This study considered CO2 emission costs in the construction processes, and measured the CO2 emission cost by the LCA method. It then proposed a mathematical programming approach to determine the proﬁt maximization of building projects. The simulation results indicated that taxation varies depending on energy intensive issues (e.g., CO2 emission), and the results affect the entire proﬁts of the construction company. This ﬁnding implies that the CO2 emission costs are the key factor for construction companies in selecting building projects. Therefore, construction companies

W.-H. Tsai et al. / Energy 36 (2011) 3022e3029

must take CO2 emission costs into consideration in order to render a successful bid that earns appropriate or excessive proﬁts. This study is the ﬁrst research to consider CO2 emissions costs in building project processes, and may provide research directions that apply sustainable development. Moreover, decision makers may ﬁnd valuable tools of objective information that can assist them in environmental assessments. In the future, other environmental factors, such as climate factors and seasonal factors, can be included in related research to analyze building project costs, or relax the limitations in this study to process further measurements.

References [1] Gustavsson L, Joelsson A, Sathre R. Life cycle primary energy use and carbon emission of an eight-storey wood-framed apartment building. Energy and Buildings 2010;42(2):230e42. [2] Nässén J, Holmberg J, Wadeskog A, Nyman M. Direct and indirect energy use and carbon emissions in the production phase of buildings: an inputeoutput analysis. Energy 2007;32(9):1593e602. [3] Radhi H. On the optimal selection of wall cladding system to reduce direct and indirect CO2 emissions. Energy 2010;35(3):1412e24. [4] Zhang Z, Baranzini A. What do we know about carbon taxes? An inquiry into their impacts on competitiveness and distribution of income. Energy Policy 2004;32(4):507e18. [5] Sathre R, Gustavsson L. Effects of energy and carbon taxes on building material competitiveness. Energy and Buildings 2007;39(4):488e94. [6] Wang WC. Supporting project cost threshold decisions via a mathematical cost model. International Journal of Project Management 2004;22(2):99e108. [7] Blengini GA, Carlo TD. The changing role of life cycle phases, subsystems and materials in the LCA of low energy buildings. Energy and Buildings 2010; 42(6):869e80. [8] Zheng G, Jing Y, Huang H, Zhang X, Gao Y. Application of life cycle assessment (LCA) and extenics theory for building energy conservation assessment. Energy 2009;34(11):1870e9. [9] Huberman N, Pearlmutter D. A life-cycle energy analysis of building materials in the Negev desert. Energy and Buildings 2008;40(5):837e48. [10] Kneifel J. Life-cycle carbon and cost analysis of energy efﬁciency measures in new commercial buildings. Energy and Buildings 2010;42(3):333e40. [11] Malmqvist T, Glaumann M, Scarpellini S, Zabalza I, Aranda A, Llera E, et al. Life cycle assessment in buildings: the ENSLIC simpliﬁed method and guidelines. Energy; 2010. doi:10.1016/j.energy.2010.03.026. [12] Hoinka K, Zie˛ bik A. Mathematical model for the choice of an energy management structure of complex buildings. Energy 2010;35(2):1146e56. [13] Fang YP, Zeng Y. Balancing energy and environment: the effect and perspective of management instruments in China. Energy 2007;32(12): 2247e61. [14] Lee CF, Lin SJ, Lewis C. Analysis of the impacts of combining carbon taxation and emission trading on different industry sectors. Energy Policy 2008; 36(2):722e9. [15] Herber BP, Raga JT. An international carbon tax to combat global warming: an economic and political analysis of the European Union proposal. American Journal of Economic and Sociology 1995;54(3):257e67.

3029

[16] Buck J, Young D. The potential for energy efﬁciency gains in the Canadian commercial building sector: a stochastic frontier study. Energy 2007; 32(9):1769e80. [17] Radhi H. Can envelope codes reduce electricity and CO2 emissions in different types of buildings in the hot climate of Bahrain? Energy 2009;34(2):205e15. [18] Intergovernmental Panel on Climate Change (IPCC). Climate change 2001: mitigation, contribution of working group III to the third assessment report of the Intergovernmental Panel on Climate Change. United States of America: Cambridge University Press; 2001. [19] Rai D, Sodagar B, Fieldson R, Hu X. Assessment of CO2 emissions reduction in a distribution warehouse. Energy; 2010. doi:10.1016/j.energy.2010.05.006. [20] Li J, Colombier M. Managing carbon emissions in China through building energy efﬁciency. Journal of Environmental Management 2009;90(8):2436e47. [21] Rolfsman B. CO2 emission consequences of energy measures in buildings. Building and Environment 2002;37(12):1421e30. [22] Suzuki M, Oka T, Okada K. The estimation of energy consumption and CO2 emission due to housing construction in Japan. Energy and Buildings 1995; 22(2):165e9. [23] Weber C, Koyama M, Kraines S. CO2-emissions reduction potential and costs of a decentralized energy system for providing electricity, cooling, and heating in an ofﬁce-building in Tokyo. Energy 2006;31(14):3041e61. [24] Amano Y, Ito K, Yoshida S, Matsuo K, Hashizume T, Favrat D, et al. Impact analysis of carbon tax on the renewal planning of energy supply system for an ofﬁce building. Energy 2010;35(2):1040e6. [25] Raugei M, Bargigli S, Ulgiati S. Life cycle assessment and energy pay-back time of advanced photovoltaic modules: CdTe and CIS compared to poly-Si. Energy 2007;32(8):1310e8. [26] Raugei M, Frankl P. Life cycle impacts and costs of photovoltaic systems: current state of the art and future outlooks. Energy 2009;34(3):392e9. [27] Tarantini M, Loprieno AD, Cucchi E, Frenquellucci F. Life cycle assessment of waste management systems in Italian industrial areas: case study of 1st Macrolotto of Prato. Energy 2009;34(5):613e22. [28] Zhang L, Huang Z. Life cycle study of coal-based dimethyl ether as vehicle fuel for urban buses in China. Energy 2007;32(10):1896e904. [29] González A, Sala JM, Flores I, López LM. Application of thermoeconomics to the allocation of environmental loads in the life cycle assessment of cogeneration plants. Energy 2003;28(6):557e74. [30] Hu Z, Tan P, Yan X, Lou D. Life cycle energy, environment, and economic assessment of soybean-based biodiesel as an alternative automotive fuel in China. Energy 2008;33(11):1654e8. [31] Stoppato A. Life cycle assessment of photovoltaic electricity generation. Energy 2008;33(2):224e32. [32] Yu S, Tao J. Simulation-based life cycle assessment of energy efﬁciency of biomass-based ethanol fuel from different feedstocks in China. Energy 2009;34(4):476e84. [33] Ward SC, Chapman CB. Risk-management perspective on the project lifecycle. International Journal of Project Management 1995;13(3):145e9. [34] Metcalf GE. Corporate tax reform paying the bills with a carbon tax. Public Finance Review 2007;35(3):440e59. [35] Beale EML, Tomlin JA. Special facilities in a general mathematical programming system for non-convex problems using ordered sets of variables. In: Proceedings of the ﬁfth international conference on operational research. London: Tavistock Publications; 1970. p. 447e54. [36] Williams HP. In: Model building in mathematical programming. 2nd ed. New York: Wiley; 1985. p. 173e7. [37] Itoh Y, Kitagawa T. Using CO2 emission quantities in bridge lifecycle analysis. Engineering Structures 2003;25(5):565e77.