rl-Hneer

olehigimgintruessounted at the end of each jib to improve the performance of crane operations. This

ction imong trations

have a signicant impact on the overall scheduling, cost, and safety of cles. Similar data are needed to quantify the CO2 emission generated

Automation in Construction 31 (2013) 255264

Contents lists available at SciVerse ScienceDirect

Automation in

j ourna l homepage: www.e lsthese projects. It is therefore important to analyze the capacity andcapability of key resources to improve the productivity of on-siteoperations. Thomas and Yiakoumis [1] separated factors affectingproductivity into four categories: environmental, site, management,and design. The construction industry has also become increasinglyaware of the sector's contribution to climate change. The constructionindustry's energy use causes signicant environmental impacts [2].Carbon footprint quantication, analysis, and reduction cannot beignored. Cranes maintain a central role in construction projects, andtower cranes in particular have dominated high-rise buildingprojects. Construction cranes consume great amounts of energy and

by on-site crane engines. At a time when the environment is ofutmost importance, every opportunity to minimize a project's carbonfootprint is vital. Minimizing crane operations on construction sites willlead to faster construction and eliminate unnecessary CO2 emissions.

Current research in the domain of construction cranes focusesprimarily on developing tools to assist practitioners in the craneselection process and visualization of their operations. A basic 4DComputer AidedDesign (CAD) simulationmodel allows users to visualizethe expected evolution of building structures during a given period ofconstruction based on the schedule of activities. Specically, these draw-ings showwhere cranes are expected to be located at different periods ofemit signicant volumes of CO2 on site. Sincof Canada has been introducing regulation

Corresponding author. Tel.: +1 780 492 0370; fax:E-mail addresses: mdshav@ualberta.ca (S. Hasan), a

(A. Bouferguene), malhussein@ualberta.ca (M. Al-Hussein(P. Gillis), atelyas@kullman.com (A. Telyas).

0926-5805/$ see front matter 2012 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.autcon.2012.11.044building involves liftingient crane operation can

in non-road equipment [5]. However, the NONROAD model datawere only intended to predict average emissions for a eet of vehi-are complex, as erection of a typical high-risea number of heavy prefabricated objects. EfcCarbon footprint

1. Introduction

The use of machinery in the construa major cost element, and cranes are aof construction equipment. Crane opewith the tower crane swing operation. A case study is presented to demonstrate the effectiveness of craneproductivity and CO2 emission analysis for building construction projects.

2012 Elsevier B.V. All rights reserved.

ndustry has always beenhe most expensive typesin construction projects

from key sources [3]. Cranes were listed in the U.S. EnvironmentalProtection Agency's (EPA) non-road vehicles and equipment categoryas one of the main sources of emissions [4]. Due to the large amountof diesel used on site, the U.S. EPA published a Non-road DieselProgram to advance emission control technologies for engines usedCO2 emissionsDouble-jib tower cranepaper presents a methodology to quantify and assess the environmental footprint (CO2 emissions) associatedCrane selectionProductivity operates using propellers mProductivity and CO2 emission analysis fohigh-rise building projects

Shaul Hasan a,, Ahmed Bouferguene b, Mohamed Aa Construction Engineering and Management, Department of Civil and Environmental Engib Campus Saint-Jean, University of Alberta, Edmonton, Alberta, Canada T6C-4G9c Koningin Astridlaan 14-2830 Willebroek Belgiumd Kullman Building Corp, Lebanon, NJ 08833, USA

a b s t r a c ta r t i c l e i n f o

Article history:Accepted 25 November 2012Available online 9 January 2013

Keywords:Tower cranes

Cranes maintain a central rused in the construction ofproper crane becomes morebuildings. To provide lift enfor high-rise building consimpact and simulation proce 2006, the Governments to reduce emissions

+1 780 492 0249.hmed.bouferguene@ualberta.ca), patrick.gillis@ggcrane.com

rights reserved.tower crane utilization on

ussein a, Patrick Gillis c, Avi Telyas d

ing, University of Alberta, Edmonton, Alberta, Canada T6G-2G7

in construction projects, and tower cranes are one of the major equipmenth-rise buildings. As the scale of construction projects increases, selecting aportant for the successful completion of projects, especially in regards to tallereers with a planning tool, this paper presents a crane selection methodologyction projects based on crane productivity performance, carbon footprint. The crane productivity analysis considers a tower crane with two jibs that

Construction

ev ie r .com/ locate /autcontime during the construction process [6]. 3D visualization is helpful in theverication and validation of crane operations [7] and can be a useful toolto improve the productivity of crane operation. However, measuring theproductivity of constructionmachinery remains difcult. Park et al. [8] in-troduced a standard construction productivity metrics system (CPMS).The proposed CPMS is a standard construction productivity data collec-tion tool and provides a framework to report industry norms to bench-mark construction productivity. Crane productivity analysis does not

receivemuch attention by practitioners. Lee et al. [9] proposed awirelesstechnology to improve the productivity of the traditional tower crane.In practice, planning for crane operations is performedmostly intuitively

2. Crane selection methodology

Lift planning and crane selection are receiving considerable atten-

Fig. 1. Crane selection methodology.

256 S. Hasan et al. / Automation in Construction 31 (2013) 255264and informally. The construction industry is seeking innovative ap-proaches to improve the productivity and minimize the carbon footprintof crane operations.

This paper presents a methodology that focuses on the selection ofcranes for high-rise building construction projects based on craneproductivity performance and carbon footprint impact. This paper alsopresents a comparison analysis between the use of a single-jib towercrane and a new type of tower crane that operates using two jibs,referred to in this paper as a double-jib tower crane [10]. Theproposed methodology is tested on a case study that involves theconstruction of a 34-story building in Brooklyn, NY. The building isconstructed using modular construction technology. It consists of over950 modules, and the cranes involved in lifting the components of theshear wall will make over 150 additional lifts.Fig. 2. A double-jib ction from practitioners and academics who wish to ensure safety andeconomy within the workplace. Rodriguez-Ramos and Francis [11]proposed a mathematical prescriptive model to establish the optimallocation for a crane in a construction site. Alkass et al. [12] describeda methodology for crane selection for construction projects. The meth-odology is incorporated into an integrated computer system capable ofadvising users on the selection of appropriate cranes for their construc-tion projects. A fuzzy logic approach to selecting the best crane type fora construction task from a list of selected crane types has also beenestablished by Hanna and Lotfallah [13]. Kamat and Martinez [14]demonstrated that process-based simulation could be used to analyzecrane operations by modeling the dynamic movement of cranes aswell as the interaction between the crane and the liftedmaterial duringa given operation. Sawhney and Mund [15] developed a prototyperane operation.

integrated crane selection tool, IntelliCranes, based on adaptive proba-

Fig. 3. Geometrical parameters used for the estimation of mean productivity.

257S. Hasan et al. / Automation in Construction 31 (2013) 255264bilistic neural networks, which assists in both crane type and cranemodel selection. Al-Hussein et al. [16], for instance, developed an opti-mization algorithm for the selection and location of cranes on construc-tion sites. Hasan et al. [17] presented an integrated module to preventcrane accidents due to poor support design practices. In practice, atower crane is selected based on the maximum load that needs to belifted, size of the loads, site layout, and the reach or capacity. In theconstruction planning stage, a particular conguration of the towercrane is chosen to yield the required heights, reach, and capacities.Practitioners focus on simulation tools to select the optimal crane.However, crane productivity and carbon footprint analysis is notconsidered in the selection of cranes.

Practically, cranes are selected by analyzing several parameterswhich are chosen by the practitioner based on a variety of quantitativeand qualitative constraints. This paper proposes a crane selectionmeth-odology that focuses on the productivity analysis and CO2 emissionanalysis of the tower crane. The proposed crane selection methodologycan be broken down into four stages as shown in Fig. 1. The rst stage iscarried out to select an appropriate technically feasible crane to performthe required lifts. In this case, cranes must be selected with knowncongurations, such as capacity, boom-jib lengths, counterweight,crane layout and other crane geometric information. Two or moretechnically feasible cranes are required for the productivity analysis.After identifying themost productive crane, the crane needs to be assessfor the environmental footprints. Once the most productive andFig. 4. Total rotation angle for a double-jib crane when >(/2).environmental friendly top cranes have been identied, the next stageis to determine which is most suitable based on the simulation of thelifting activities. The developed model simulates the lifting operationsbased on priority, schedule and crane congurations and providesinformation about the number of cranes required for each type ofselected crane. If any of the selected crane fails to lift any loads accordingto schedule, then the crane need to be replaced with next suitable crane.Finally cost estimation need to be performed for all the selected cranes.

2.1. Select feasible cranes

Construction cranes are selected based on themaximum load to belifted, size of the load, site layout, and the reach or capacity. The devel-oped methodology considers the crane's geometrical characteristics,the lift, and the site. A feasible crane can be selected from previouslydeveloped databases such as D-Crane [18], or PCL Crane Database[19]. These databases house information related to cranes, theirgeometric lifting conguration specications, and their lifting capacitiesbased on information provided bymanufacturers in crane lifting capac-ity charts. A crane with the capacity to carry the design load can beselected from the list of available cranes in the database. After identify-ing the most feasible cranes satisfying the safety requirements theselected cranes need to be evaluated for the productivity at the planningstage.

2.2. Productivity analysis

Traditional tower crane operation poses several challenges, namely1) the wind affects the running of the boom during lifting, 2) torsionforces occur in the central tower, and 3) assembling, dismantling andtransporting the crane from one yard to another can be complex,which decrease the productivity of lifting operations. Taking thesedeciencies into consideration, the productivity analysis considers atower crane equipped with two jibs (see Fig. 2). As cranes are the

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 50 100 150 200Swing angle in degrees

Ratio of single to double jib cycle angle

Fig. 5. Ratio of single-jib crane and double-jib crane cycle angle.most expensive equipment in the construction process, the productivityanalysis and improvement of crane operations become importantfactors in the construction industry. Crane productivity analysis doesnot receive much attention by practitioners. This paper provides amethodology to calculate the crane swing productivity of traditional(single-jib) tower cranes and double-jib tower cranes (see Fig. 2),which swings with the assistance of mounted propellers.

To derive the mathematical apparatus needed to obtain meanestimations of productivity, the crane is conveniently located at theorigin of the Cartesian coordinate's frame, the X axis of which isdened as the line connecting the crane to the pick point (see Fig. 3).The rotation angle needed to take the load from the pick point to its des-tination is referred to as and is subjected to the constraint 0bdegree.

Assuming the pick point is xed, the full cycle of a single-jib crane(pick point to destination to pick point) requires a forward and a

backward rotation totaling (2) angle units. On the other hand, for adouble-jib crane two different scenarios should be considered (seeFig. 4):

1. When b(/2), the full rotation cycle is described by (2) in thesame way as its single-jib counterpart.

2. However, for rotations larger than (/2), the total rotation isconstant and equals since in this case both arms are used in sequencefor lifting loads.

Fig. 6. Three different scenarios of daily lifts.

258 S. Hasan et al. / Automation in Construction 31 (2013) 255264Fig. 7. Distribution of swing angles for (a) the worst, (b) the most likely and (c) thebest case lift scenarios.To analyze the crane's productivity, the swing angle can be assumedto be a uniformly distributed random variable, that is X~U(0,) degree,and it becomes possible to quantify the average swing saving thatresults from using a double-jib crane.

1. For a single-jib crane, if the swing angle is then the cycle angle(that is pick point to destination to pick point) is 2. In this case,the average cycle angle can be computed as the expected valueof the random variable 2X, thus leading to,

2Xh isingle

0

2xdx

: 1

2. On the other hand, for a double-jib crane, the cycle angle is 2X for0bXb/2; whereas, for X>/2 such an angle is constant and equals degree. As a result, the expected cycle angle can be computed as,

2Xh idouble =2

0

2xdx

=2

dx

34

: 2

Based on the above expectation values, a double-jib crane is onaverage 25%more efcient in comparison to its single-jib counterpart.However, the distribution of such a saving cannot be uniform since it isessentially achieved only for angles larger than /2. The ratio betweenthe cycle angles of the double- and single-jib cranes can bemodeled bythe function,

s Single jib cycle angleDouble jib cycle angle

1 0 X =2= 2X =2 b X :

3

The above function is represented in Fig. 5.From the above analysis, a recommendation can be made to use a

double-jib crane if the number of lifts for which the swing angle islarger that /2.

Construction is known to be subject to a variety of uncertainties,and requires a model that incorporates these variables. Let us assumethat the number of daily lifts is described by the discrete distributionwhich provides three different scenarios (see Fig. 6):

1. The worst case in which only n1 lifts were performed.2. The most likely case where the operator manages n2 lifts.3. And nally the best case in which n3 lifts were completed.

Associated with each of the above cases are the probabilities p1, p2and p3 on a scale of 0 to 1 (0 cannot occur, 1 certain occur).

In addition it is assumed that the number of lifts is a result of theswing angle, that is the larger the angle the smaller the number oflifts. Although this is a simplication of crane operations, it has theadvantage of providing a framework that takes uncertainty intoaccount. As a result, the following piecewise uniform distributionswill be considered for each of the cases enumerated above. Accordingto Fig. 7, the worst case scenario (corresponding to n1 lifts) is describedby the distribution of the swing angle depicted in Fig. 7a. In this case theprobability q1,3 of wide swinging (here assumed to be between 120and 180) is larger than that corresponding to smaller angles. Theswing angles of the most likely lifting scenario were chosen to beuniformly distributed according to Fig. 7b, and the probabilities writtenas q2,1=q2,2=q2,3. Here the indices are kept for the sake of generalityand uniformity of the notation. As for the best case scenario, it is charac-terized by a large probability of small swing angles (see Fig. 7c), thusmaking the lifting operation faster and more productive.

At this point, using the distribution of swing angles correspondingto the worst, most likely, and best case scenarios, the expected valueof the cycle angle can be derived, from which one can obtain the

productivity ratio between a single and double-jib crane.

1. For a single-jib crane, the expected daily cycle angle (pick point todestination to pick point) can be represented as a dot product of

and 7. However, in this particular case an assumption can be madethat 1+2>/2, which is well-founded as for angles smallerthan /2 there is no benet in using a double-jib crane (see Fig. 5).

2Xh idouble;daily

q11 q12 q13q21 q22 q23q31 q32 q33

24

35

1221 =2 2

= 21

24

35

0@

1A n1 p1n2 p2

n3 p3

24

35:5

To illustrate the usefulness of the above theory, Fig. 8 shows theproductivity ratio (double-jib over single-jib) for varying values of1 and 2. As for the other parameters, they are set as follows,

3 ;q11 q12 q13q21 q22 q23q31 q32 q33

24

35 0:2 0:3 0:51=3 1=3 1=3

0:65 0:20 0:15

24

35 and n1 p1n2 p2

n3 p3

24

35

50 0:2580 0:55110 0:20

24

35: 6

Using the input parameters show above, the following contourmap was obtained for the production,

An interesting application of the contour map depicted in Fig. 8 isthe determination of the angles that will lead to a given goal in termsof productivity. As may be seen, for any given productivity ratio, arange of possible values for 1 and 2 can be easily determined.

Fig. 8. Contours for productivity ratios (double-jib over single jib crane) for varyingvalues of 1 and 2.

259S. Hasan et al. / Automation in Construction 31 (2013) 255264two vectors such that,

2Xh isingle;daily q11 q12 q13q21 q22 q23q31 q32 q33

24

35 11 2

2 3

24

35

0@

1A n1 p1n2 p2

n3 p3

24

35: 4

Here, it is important to mention that 3 is the largest possibleswing angle, which in many practical scenarios can be equated to degree.

2. In a similar manner, the expected daily cycle angle for a double-jibcrane can be determined based on the distributions given in Figs. 6Fig 9. Crane operations simulation moSuch values can, in turn, be used to help locate the position of thecrane that will lead to the selected productivity.

Once the productive cranes have been identied the next step is todetermine which is the most suitable based on the carbon footprints.

2.3. CO2 emission analysis

The CO2 emissions of a tower crane are relatively lower than those ofa mobile crane at the same productivity level. In this section, emphasisis placed on deriving a quantitative estimation for the calculation of CO2for tower cranes during the swing operation. Let Y be the cycle angle fordel from Simphony.NET3.5 [22].

Fig. 10. Typical oor plan and crane location.

260 S. Hasan et al. / Automation in Construction 31 (2013) 255264a given lift operation; in such a case the energy associated to thismotionis dened as,

Ecycle PY= 2

: 7

The ratio [Y/(2)] is a representation of the cycle angle as a fraction

of a full rotation of the jib. and P refer to the speed of the jib (expressedin rotations per second) and the power (expressed in kilowatts) itrequires to perform a single full rotation. Assuming the distribution of

Fig. 11. Lufng jib crane oswing angles is known (see previous section), the average yearly energyconsumption is obtained as,

Eday n EcycleEannual d Eday Pn Y= 2

d 8

where n and d represent the average number of lifts per day and the

number of working days per year. Tower cranes traditionally operateon electricity. Since a primary energy source, such as coal, natural gas,or petroleum, is needed to generate electricity, CO2 emissions are

peration on the site.

(C) by a tower crane during full operation can be calculated satisfyingEq. (9):

C E EF 9

where EF is the emission factor.After identifying the best suitable cranes based on the carbon

footprints the next step is to simulate the lifting operations based

Table 1Total rotation and time for swing to complete different module lifting operations.

Moduleno.

Angle fromsource (deg)

Single jib crane 2-jib crane

Total rotation(deg)

Time (min) Total rotation(deg)

Time (min)

1 135 270 1.50 180 1.002 139 278 1.54 180 1.003 140 280 1.56 180 1.00

261S. Hasan et al. / Automation in Construction 31 (2013) 2552644 147 294 1.63 180 1.005 123 246 1.37 180 1.006 142 284 1.58 180 1.007 114 228 1.27 180 1.008 134 268 1.49 180 1.009 105 210 1.17 180 1.0010 120 240 1.33 180 1.0011 81 162 0.90 162 0.9012 79 158 0.88 158 0.8813 78 156 0.87 156 0.8714 98 196 1.09 180 1.0015 77 154 0.86 154 0.8616 60 120 0.67 120 0.6717 47 94 0.52 94 0.5218 57 114 0.63 114 0.6319 35 70 0.39 70 0.3920 50 100 0.56 100 0.5621 29 58 0.32 58 0.3222 44 88 0.49 88 0.4923 26 52 0.29 52 0.2924 38 76 0.42 76 0.4225 21 42 0.23 42 0.2326 35 70 0.39 70 0.3927 19 38 0.21 38 0.21attributed to the process. The Energy Information Administration (EIA)in the United States collected data on CO2 emissions from electricitygeneration [20]. The CO2 emissions are presented on the basis of totalmass (tons), output rate (pounds per kilowatt-hour), and the averageoutput rate, which is the ratio of pounds of carbon dioxide emittedper kilowatt-hour to electricity produced from all energy sources,both fossil and non-fossil. The GHG (green house gas) emission factorused to estimate GHG reductions is based on the GHG emission intensi-ty of marginal electricity generation in each province. It is assumed thatincremental renewable power generation replaces existing marginalelectricity generation. To obtain a national factor, the emission factorfor the marginal fuel in each province was weighted by the provincialshare of electricity generation and then summed. Thus, across Canada,a GHG emission factor of 465.88 t/GWh was used [3]. Examples of suchgreenhouse gases are methane, peruorocarbons and nitrous oxide.A gas' contribution to the greenhouse effect is affected by both its charac-teristics and its abundance. Equivalent CO2 (CO2E) is the concentration ofCO2 that would cause the same level of radiative forcing as a given typeand concentration of greenhouse gas. The total amount of CO2 emission

28 33 66 0.37 66 0.3729 28 56 0.31 56 0.3130 23 46 0.26 46 0.2631 18 36 0.20 36 0.2032 12 24 0.13 24 0.13Total: 25.41 20.89

Table 2Energy consumptions and CO2 emissions comparison of lufng jib and double-jib crane.

Crane type Crane capacity for swing Swing speed Energy consumpti

Lufng jib crane 20 kW 0.6 rpm 10,140 kWh/yearDouble jib crane 10 kW 1.0 rpm 2280 kWh/yearon priority, schedule and crane congurations.

2.4. Simulation

Crane operations can be broken down into separate activitieswhich are dependent on: (1) source location (the location of the loadwhen delivered to site); (2) destination location (nal design location);(3) weight and size; (4) priority setting; (5) crane location; (6) boom/jib length; (7) rotating/hoisting speed; and (8) reach (required liftingradii). To simulate the lifting operation, travel speeds for hoisting, radial,and horizontal trolleymovements also need to be considered. A simula-tion model has been developed using Simphony.NET3.5 [21] as shownin Fig. 9. The Simphony software allows users to select different kindof distribution model depends on the uncertainty of any activity. Forexample, in the construction sites, if the change of source location isslight, the user can select the uniform distribution for the crane hookswing activity. However, if the change of source location is signif-icant then user need to select a different distribution for hook swingactivity which can proper t the uncertainty of the reallocation (i.e., tri-angular or normal distribution). Based on the project activity schedule,the construction site requires a crane to complete a lifting activity, ifany. The following steps describe the total time required to performan activity:

a. Lifting requirement is described by the activity schedule.b. All activities which require lifting are added to a wait list for

the next available crane.c. The crane operator selects a lift (object) based on the project

schedule.d. The crane hook moves from the last location to the source

location of the load (ts1).e. Time required to hoist-down without any load (th-down1).f. Time required to hook the load (thooking).g. Time required to hoist-up the load (th-up1).h. Time required for radial (swing) movement (ts2).i. Time required for horizontal (trolley) or vertical (lufng)movement of the load (tl).

j. Time required to hoist-down the load (th-down2).k. Time required to position and unhook the load (tun-hooking).l. Time required to hoist-up without the load (th-up2).

m. The crane hook moves another source location to pick a loadbased on the lifting schedule.

n. If no activities are waiting for a crane, hook moves to predenedparking location and simulation stops.

Most of the activities vary based on the complexity of the workand the crew's performance. Swing activity depends on the radialdistance between the source location of the load to the destinationlocation, and the crane's swinging speed. Similarly, hoist up and

on per year (E) Emission factor (EF) in Canada CO2 emissions per year (C)

465.88 g/kWh 4724 kg/year465.88 g/kWh 1062 kg/year

262 S. Hasan et al. / Automation in Construction 31 (2013) 255264hoist down activities depend on the hoisting speed of the crane withand without the load, and the height of the destination location. Thedeveloped simulation model provides the possible time for eachlifting cycle. The total time spend by a crane for a full lifting cycleoperation (Tf) can be calculate satisfying Eq. (10)

Tf ts1 ts2 thdown1 thooking thup1 tl thdown2 tunhooking thup2:10

The number of cranes required for the project (N) can be calculatedsatisfying Eq. (11):

N Tf nATw 11

Fig. 12. Different cran

Table 3Simulation results.

Crane type No. of cranerequired

No. of module can be installed per day

Floor 2 Floor 3 Floor 33 Floor 34

Lufng jib crane 1 21 21 16 16Double jib crane 1 23 23 18 18where the notation xmeans that x is rounded up to the closest integervalue, nA is the number of lifting activities/cycles per day and Tw is thetotal working duration of the project on a typical day. The idle time ofcrane is waste and thus, it is not considered in determining number ofcrane requirements.

Most of the critical planning for crane operation needs to benalizedduring the crane selection process. If it is found that the selected crane isremains idle in most of the working duration (non-productive), thenthe planning engineer needs to select different crane to avoid scheduledelays or increase in budget.

3. Case study

This analysis is based on the installation of 950 structural steelmodules for a 34-story building. The main oor will be constructedonsite, and the remaining oors will be lifted into place by a towercrane. The typical oor plan and the optimum position of the craneare shown in Fig. 10. In order to keep the installation sequenceowing, it will be necessary to have a minimum of two (2) trailers(loaded with ready modules) onsite to move the delivered modulesto the lifting zone. The following assumptions have been consideredto analyze the project:

1) Hoisting and lufng speeds are based on the proposed tower cranefeatures.

2) Day starts at 7.30 am and will nish around 4.30 pm.

e work processes.

263S. Hasan et al. / Automation in Construction 31 (2013) 2552643) Criticalmoduleswill require 50% additional time for bolting/welding:

a. Modules next to shear wallsb. Modules next to stairs/elevator shaftsc. Final corner modules for every oor.

4) Modules will be unhooked at their nal location once they arealigned. Welding/bolting will continue immediately thereafter.

Step 1 Crane Selection: Based on expert analysis without consideringa double-jib crane, the best crane to be used on site is a 32 toncapacity tower crane with a 164-ft lufng jib. The operation ofthe proposed 32 ton capacity tower crane is shown in Fig. 11.The management wanted to compare the lifting productivityand the environmental impact of the selected crane with asame capacity double-jib crane [23].

Step 2 Productivity analysis: based on the source location (modulepick point), module placing position, and the crane positionas shown in Fig. 10, the angle of each module from the sourceis calculated. The total rotation of the jib from pick point toplacing module and to pick point is calculated for both typesof crane as shown in Table 1. Assuming the same swingspeed for both types of crane, the total time for the swing ofa jib to complete the different module lifting operations iscalculated and presented in Table 1.The total time to place the 32 modules in each oor will takeapproximately 25.41 min for the lufng-jib crane and approxi-mately 20.89 min for the double-jib crane (see Table 1). Thusthe double-jib crane can complete the swing activity 17.8%faster than the lufng-jib crane.

Step 3 CO2 emission analysis: CO2 emissions per year for the lufng-jibtower crane and a double-jib crane of the same capacity can becalculated using Eqs. (7) to (9) as presented earlier in thispaper. Assuming the power consumption and the swing ratefor the proposed lufng-jib crane are 20 kW and 0.6 rpm,respectively, and for the double-jib crane with propellers 10 kWand 1 rpm, respectively, the energy consumption and CO2 emis-sions per year are calculated and presented in Table 2.The results show that a double-jib crane emits 77.5% less CO2 thanthe lufng-jib tower crane for the swing operation only.

Step 4 Simulation: the developed simulation model as illustrated inFig. 9 has been used in this case study. In this model, all activitydurations are analyzed within a triangular distribution, and vehundred minutes of working time a day are considered. Theobjective of the simulation model is to provide number of cranerequires and minute-by-minute schedules for crane operations.Most of the crane activities vary based on the complexity of thework and the crew's performance and productivity. For a 34-story building, the lifting height varies from 10 m to 110 m.Thus hoisting the module to the 34th oor can take up to sevenminutes. Given that accuracy is an issue, the probability distribu-tion which describes the ascending hoisting is updated for eachoor.

Considering crane activities as shown in Fig. 12, utilizing thelufng-jib crane 21 modules can be installed per day at oor 2. Foroor 34, utilizing the same crane operation sequence, 16 modulescan be installed per day due to the increase of oor height. Thus,using the proposed lufng-jib crane to install 950 modules will takeat least 52 days. While satisfying capacity and other constraints,using the double-jib tower crane 23 modules can be installed perday at oor 2. For oor 34, utilizing the same sequence of craneoperation, 18 modules can be installed per day. Thus, using adouble-jib tower crane to install 950 modules will take a minimumof 47 days, which improves the productivity of the lifting operation

by about 10%. Table 3 illustrates the simulation results.The results show that the double-jib tower crane operation requiresa shorter duration than the option of a single-jib crane operation, whilesatisfying capacity and other constraints. From the above four steps it isvisible that the selected double jib crane is more feasible, productiveand environmental friendly crane option than the selected lufng jibcrane. However, installation of the double jib crane in the downtownof a busiest city creates a big challenge and increase the cost of installa-tion. Thus a costbenet analysis is recommended to the projectmanager before nalize the crane selection process. Again, the nalselection of the crane also depends on the management policy, craneownership and availability of technical support.

4. Conclusion

The crane selection process for a high-rise building constructionproject is complex. Optimization or proper crane selection requiresnot only extensive analysis of crane conguration, but also productivityand CO2 emission analyses. This research has been motivated by thelarge number of tower cranes used in the construction industry andthe consequent need to improve productivity and safety. A metho-dology was developed to identify the productivity of traditional towercranes and double jib cranes with propellers. The results show that adouble-jib tower crane can improve productivity an average of 25%for the swing operation only. This paper also presented a methodologyto calculate the CO2 emissions for tower crane swing operations. Theanalysis shows that the double-jib crane operation emits less CO2 thana traditional tower crane operation. Finally, the developed simulationmodel narrowed the list of feasible cranes by considering differentuncertainties in crane operations. The case study demonstrated thepotential capabilities of the proposed crane selection method. Thedeveloped simulation model currently focuses on the crane activitiesonly, which one of the limitations of the model. The model can beexpanded in future by integrating other internal activities in theconstruction sites or external factors such as weather impact whicheffect the crane operation. Also carbon footprint analyses for mobilecranes need to be developed. The proposed crane selection criteria fora building construction project are proven to be effective in reducingthe time and environmental footprints associated with the towercrane operation on construction sites.

References

[1] H.R. Thomas, I. Yiakoumis, Factor model of construction productivity, Journal ofConstruction Engineering and Management 113 (4) (1987) 623639.

[2] C. Hendrickson, A. Horvath, Resource use and environmental emissions of U.S.construction sectors, Journal of Construction Engineering and Management 126(1) (2000) 3844.

[3] Environment Canada, A climate change plan for the purposes of the Kyoto Protocolimplementation Act, http://www.climatechange.gc.ca/Content/4/0/4/4044AEA7-3ED0-4897-A73E-D11C62D954FD/KPIA_2010.pdf 2010.

[4] EPA, Emission facts: average carbon dioxide emissions resulting from gasolineand diesel fuel, http://www.epa.gov/OTAQ/climate/420f05001.htm 2005.

[5] EPA, Final regulatory analysis: control of emissions from nonroad diesel engines,EPA420-R-04-007, Assessment and Standards Division, Ofce of Transportationand Air Quality, Washington, D.C., 2004.

[6] B. Akinci, K. Tantisevi, E. Ergen, Assessment of the capability of a commercial 4DCAD system to visualize equipment space requirements on construction sites,Construction Research Congress: Winds of Changes: Integration and Innovationin Construction, ASCE, 2003..

[7] M. Al-Hussein, M.A. Niaz, H.T. Yu, H. Kim, Integrating 3D visualization and simulationfor tower crane operations on construction sites, Automation in Construction 15 (5)(2006) 554562.

[8] H.S. Park, S.R. Thomas, R.L. Tucker, Benchmarking of construction productivity,Journal of Construction Engineering and Management, ASCE 131 (7) (2005)772778.

[9] U.K. Lee, K.I. Kang, G.H. Kim, Improving tower crane productivity using wirelesstechnology, Computer-Aided Civil and Infrastructure Engineering 21 (2006)594604.

[10] S. Hasan, M. Al-Hussein, P. Gillis, Advanced simulation of tower crane operationutilizing system dynamics modeling and lean principles, Paper, 302, WinterSimulation Conference, Baltimore, Maryland, U.S., 2010.

[11] W.E. Rodriguez-Ramos, R.L. Francis, Single crane location optimization, Journal of

Construction Engineering and Management 109 (4) (1983) 387397.

[12] S. Alkass, M. Al-Hussein, O. Moselhi, Computerized crane selections for constructionprojects, 13th Annual ARCOM Conference, Association of Research in ConstructionManagement 2 (1997) 427436.

[13] A.S. Hanna, W.B. Lotfallah, A fuzzy logic approach to the selection of cranes,Automation in Construction 8 (5) (1999) 597608.

[14] V. Kamat, J. Martines, Visualization simulated construction operations in 3D,Journal of Computing in Civil Engineering 15 (4) (2001) 329337.

[15] A. Sawhney, A. Mund, Adaptive probabilistic neural network-based crane typeselection system, Journal of Construction Engineering and Management, ASCE128 (3) (2002) 265273.

[16] M. Al-Hussein, S. Alkass, O. Moselhi, Optimization algorithm for selection andon-site location of mobile cranes, Journal of Construction Engineering andManagement, ASCE 131 (5) (2005) 579590.

[17] S. Hasan, M. Al-Hussein, U.H. Hermann, H. Safouhi, Interactive and dynamicintegrated module for mobile cranes supporting system design, Journal ofConstruction Engineering and Management, ASCE 136 (2) (2010) 179186.

[18] M. Al-Hussein, S. Alkass, O. Moselhi, D-CRANE: database system for utilization ofcranes, Canadian Journal of Civil Engineering 27 (2000) 11301138.

[19] S. Hasan, M. Al-Hussein, U.H. Hermann, H. Safouhi, An automated system formobile crane selection, swing control and ground pressure calculation, the twelfthinternational conference on civil, structural and environmental engineering com-puting, in: Paper, 292, Civil-Comp Press, Stirlingshire, United Kingdom, 2009.

[20] EPA, Carbon dioxide emissions from the generation of electric power in theUnited States, http://tonto.eia.doe.gov/ftproot/environment/co2emiss00.pdf 2000.

[21] D. Hajjar, S.M. AbouRizk, Simphony: An Environment for Building Special PurposeConstruction Simulation Tools, in: Proceedings of the 1999 Winter SimulationConference, Phoenix, AZ, 1999, pp. 9981006.

[22] S. Han, S. Hasan, M. Al-Hussein, K.U. Gokce, A. Bouferguene, Simulation of MobileCraneOperations in 3D Space, Proceedings of the 2012Winter Simulation Conference,Berlin, Germany, 2012.

[23] GG Crane Group, GG Tower Crane of the 21st Century, http://www.ggcrane.com2010.

264 S. Hasan et al. / Automation in Construction 31 (2013) 255264

Productivity and CO2 emission analysis for tower crane utilization on high-rise building projects1. Introduction2. Crane selection methodology2.1. Select feasible cranes2.2. Productivity analysis2.3. CO2 emission analysis2.4. Simulation

3. Case study4. ConclusionReferences