Life cycle assessment of pavement: Methodology and case study

Download Life cycle assessment of pavement: Methodology and case study

Post on 13-Sep-2016




3 download


<ul><li><p>Civil and Environmental Engineering, University of South Florida, 4202 E. Fowler Avenue, Tampa, FL 33620, United States</p><p>especially for the pavement structure effect.</p><p>1. Introduction</p><p>Life-cycle analysis (LCA) in pavement aaterit incresud thaen fong et</p><p>candidates of a LCA model. For pavement, this means that they should serve the same trafc over the same analysis spanwith the same performance. To assess the environmental impacts of pavement, a system of LCA model is developed, asshown in Fig. 1. The LCA functionality is fullled by six components, including material module, distribution module, con-struction module, congestion module, usage module, and EOL module, with various supplementary models attached to thecorresponding modules.</p><p>1361-9209/$ - see front matter 2012 Elsevier Ltd. All rights reserved.</p><p> Corresponding author.E-mail address: (B. Yu).</p><p>Transportation Research Part D 17 (2012) 380388</p><p>Contents lists available at SciVerse ScienceDirect</p><p>Transportation Research Part D complete; third, the EOL phase is simply taken as landll while practically, most hot mixture asphalt (HMA) is recycledand old Portland cement concrete (PCC) is crushed to substitute base course aggregates.</p><p>2. Methodology</p><p>We begin by dening a functional unit needed to build the LCA model framework. A functional unit quanties a standardamount to be compared between alternatives that serve this function. Equivalent functionality shall be maintained for allsists of the following components: mHowever, most of the work does nomodels, usage and trafc congestionplexity. Huang et al. (2009) suggesteroadwork periods are signicant. Evimprove (Keoleian et al., 2005; Zha 2012 Elsevier Ltd. All rights reserved.</p><p>ssessment is still at an immature stage. Typically, a LCA model of pavement con-al, construction, use, maintenance and rehabilitation (M&amp;R), and end of life (EOL).orporate all the components (Chan, 2007). Two most important elements in LCAlted from construction and M&amp;R activities are typically ignored due to great com-t additional fuel consumptions and pollutant emissions due to trafc delay duringr studies that incorporate the use and congestion phases, there is still room toal., 2010): rst, data in some studies are outdated; second, the usage phase isa r t i c l e i n f o</p><p>Keywords:Pavement designInfrastructure life cycle assessmentPavement overlay systems</p><p>a b s t r a c t</p><p>A life cycle assessment model is built to estimate the environmental implications ofpavements using material, distribution, construction, congestion, usage, and end of lifemodules. A case study of three overlay systems, Portland cement concrete overlay, hotmixture asphalt overlay, and crack, seat, and overlay, is presented. The case leads to thefollowing conclusions. It is reasonable to expect less environmental burdens from thePortland cement concrete and crack, seat, and overlay options as opposed to hot mixtureasphalt while although the results have a high degree uncertainties. The material,congestion, and particularly usage modules contribute most to energy consumption andair pollutant. Trafc related energy consumption and greenhouse gases are sensitive totrafc growth and fuel economy improvement. Uncertainties exist in the usage module,Life cycle assessment of pavement: Methodology and case study</p><p>Bin Yu , Qing Lu</p><p>journal homepage: www.elsevier .com/ locate / t rd</p></li><li><p>(CH4),</p><p>B. Yu, Q. Lu / Transportation Research Part D 17 (2012) 380388 381The distribution module is closely linked to the material module and the EOL module. All materials, equipment, andwastes are transported by a combination of road, rail, and waterway. Greenhouse Gases, Regulated Emissions, and EnergyUse in Transportation (2010) is used to model greenhouse gas emissions and energy embracing a data for fuel and electricityproduction, truck transportation, tie and dowel bar production, and natural gas burned that may be used during the pave-ment life time. Emission data for all non-road construction and vehicular equipment are obtained from the US Environmen-tal Protection Agencys (EPA) NONROAD 2008 model for construction and maintenance activities. For each piece ofconstruction equipment, an estimate of the engine horsepower is made on the basis of one or two typical machines.NONROAD2008 model provides emission factors for various ranges of horsepower.</p><p>Most prior work has not included a EOL module because the pavement structure is assume to have an indenite life. It isdesired to investigate the role of EOL module on the LCA model. Environmental burdens to dismantle and transport the oldpavement, the environmental savings due to the reuse of old pavement materials, and the potential additional energy con-sumption to process the old pavement materials before they can be used, need to be identied. They can be modeled in asimilar fashion as the way of material, distribution, and construction modules.</p><p>Trafc delay induced by construction and rehabilitation activities has signicant inuences on energy consumption andpollutant emissions compared with those under normal vehicular operations. The changes in trafc ow, trafc delay, andqueue length are estimated using the QuickZone model. Outputs include detour rate, queue length and speed reductionwithin work zones. Once vehicle delays due to construction and maintenance events are determined, they are coupled withfuel consumption and vehicle emissions to measure their environmental impacts. CO2 is calculated by the fuel consumptions(Emission Facts, 2005). Other vehicle emissions are calculated using US EPAs MOBILE 6.2 model, which supplies the tailpipeemissions and evaporative emissions at varying trafc speeds on a per year basis through 2050.</p><p>Fuel consumptions and environmental burdens are calculated as the differences between those of construction and reha-bilitation periods and those of normal operations:</p><p>Ytotal VMTqueue Yqueue VMTworkzone Yworkzone VMTdet our Ydet our VMTnormal Ynormal 1where Yi is the value of different environmental indicators, such as fuel usage (L/km) or emission values (g/km), VMTi is themiles traveled by vehicles (km or mile), i is scenario index, representing the total, waiting in queue, passing through workzone, taking detour, or operating under normal conditions.Thperiod</p><p>Throughand tr</p><p>Incand proughsuremfrom 2lioratevehicl2</p><p>nitrogen oxide (NOx), sulfur oxide (SOx), volatile organic compound (VOC), particulate matter (</p></li><li><p>3.1. Th</p><p>direction, the widths of the inner paved shoulder, main lanes, and outsider paved shoulder are 1.2 m, 3.6 2 m, and2.7 m</p><p>HM</p><p>382 B. Yu, Q. Lu / Transportation Research Part D 17 (2012) 380388and Muench, 2010). Crack, seat, and overlay (the CSOL option). Crack and seat the existing PCC pavement and then overlay with 125 mmHMA.Use the same mill-and-ll plan as the periodic rehabilitation strategy every 16 years (Weiland and Muench, 2010).</p><p>The pavement overlay designs follow the Florida Department of Transportation (FDOT) pavement design manual as ver-ied by the Mechanistic-empirical Pavement Design Guide (MEPDG) software using Florida local weather data. Thus thefunctional unit is a one kilometer overlay system over an existing PCC pavement with four lanes in two directions that wouldprovide satisfactory performance over a 40-year period.</p><p>For material module, cement concrete production uses data from the Portland Cement Association (Marceau et al., 2007)while HMA production uses data from the Swedish Environmental Research Institute (Stripple, 2001). Distribution and con-struction modules are estimated based on the quantities of construction and maintenance activities.</p><p>For EOL module, reclaimed concrete material (RCM) is frequently used to substitute aggregates in base course. Reclaimedasphalt pavement (RAP) is now routinely accepted in asphalt paving mixtures with substitution rates ranging from 10% toA surface and replace the same depth of new HMA) plan every 16 years as a periodic rehabilitation strategy (Weilanda year. Three replacement options frequently adopted in Florida are considered:</p><p> Remove and replace the 225 mm PCC pavement with 250 mm new PCC (the PCC option). Diamond grinding is frequentlyused to restore surface smoothness and reported to be viable for 1617 years (Stubstand et al., 2005) and thus is per-formed every 16 years as a periodic rehabilitation strategy.</p><p> Remove and replace the existing pavement with 225 mm HMA (the HMA option). Use a mill-and-ll (remove 45 mm. There is an annual average daily trafc ow (AADT) of 70,000, with 8% being truck that is growing at growth of 4%For the case study we consider an old PCC pavement that is at the end of its service life and requires rehabilitation torestore the serviceability. The existing base course is assumed to perform well and can function without intensive mainte-nances. This pavement has a 225 mm PCC surface with 250 mm crushed aggregate as base course, and subgrade. In eache studyBesides the inuence on fuel economy, increase of IRI reduces the driving speed and thus leads to a reduction of highwaycapacity. The speed-reduced eet may witness signicant fuel consumption increase and pollutant emission changes.Moreover, additional roughness causes increased friction and vertical acceleration of the vehicle body, and thus leads tomore vehicle fuel consumption and pollutant emissions. How these factors contribute to the life cycle inventory will beaddressed by the case study in detail.</p><p>Pavement structures have signicant inuence on the fuel consumptions of vehicles, especially for asphalt compared withPCC and composite pavements (Taylor et al., 2000). Taylor and Pattens (2006) study suggested that PCC and composite pave-ments have signicant fuel economy advantages over HMA pavement.</p><p>Albedo directly contributes to global cooling by adjusting the radiative forcing of the earths surface. As a surface covering,pavements can reect a portion of the incoming solar radiation back into space, thus adjusting the global energy balance.Akbari et al. (2008) estimated that for every square meter, 2.55 kg of emitted CO2 is offset for every 0.01 increase in albedodue to increased radiative forcing. Eq. (3) gives the means to calculate the benet:</p><p>DmCO2 100 C A Da 3where DmCO2 is the mass equivalents of CO2 mitigated (kg), C is the CO2 offset constant (kg CO2/m</p><p>2), A is the area of pavement(m2) and Da is the change is albedo.</p><p>Over time, much of the CO2 that was originally liberated from limestone during cement kiln processes will rebind itself tothe cement in the pavement through the carbonation process. The carbonation of concrete can be modeled using a simpli-cation of Ficks second law of diffusion (Lagerblad, 2006):</p><p>dc k</p><p>tp</p><p>4where dc is the depth of carbonation (mm), k is the rate factor (mm/y1/2) and t is time (year).</p><p>Not all of the calcium in the concrete, however, is expected to bind CO2 molecules; the binding efciency is suggested tobe roughly 75% (Stolaroff et al., 2005). The mass of CO2 sequestered is given by:</p><p>mCO2 dc A qconcrete mcement=concrete MCO2MCaO</p><p> e 5</p><p>wheremCO2 is the mass of CO2 sequestered through carbonation (kg), dc is the depth of carbonation (m), A is the surface areaof pavement (m2), qconcrete is the density of concrete (kg/m3),mcement/concrete is the mass ratio of cement in concrete,MCO2 is themolar mass of CO2, MCaO is the molar mass of CaO and e is the binding efciency of CO2 to CaO.</p><p>3. Case study of three pavement overlay systems</p></li><li><p>B. Yu, Q. Lu / Transportation Research Part D 17 (2012) 380388 3830 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20</p><p>1.0</p><p>1.1</p><p>1.2</p><p>1.3</p><p>1.4</p><p>1.5</p><p>1.6</p><p>1.7</p><p>1.8</p><p>1.9</p><p>IRI (</p><p>m/k</p><p>m)</p><p>Age (Year)</p><p> PCC option HMA option CSOL option</p><p>Fig. 2. Development trends of IRI as predicted from MEPDG.</p><p>Table 1Fuel economy comparisons of two pavement structures to PCC.</p><p>Season Winter Spring Fall Winter</p><p>Pavement type HMA CSOL HMA CSOL HMA CSOL HMA CSOL</p><p>For passenger carsComparison (%) 3.1 2.07 0.42 1.2 0.42 1.2 0.42 1.250% or more, depending on state specications. In EOL module, two scenarios are tested. One is to crush PCC pavement anduse 10% and 20% of the crushed materials in base course layer. The other is to recycle the milled asphalt mixture into theasphalt drum plant with a portion of 10% and 20%. RAP are treated as free of any inherent or feedstock energy, whereas,in fact, it retains its feedstock value indenitely. We include the energy used to extract RAP (roadway milling) and the trans-portation to the asphalt plant where it is remixed, but excluded other possible energy demands because milled RAP are veryconsistent and can be used in newmixes without further screening or crushing (Federal Highway Administration, 2011). Theresults suggest that the effect of recycling is very limited compared with the other modules and will not be discussed in de-tail here. One can nd more information in the report by Yu and Lu (2011).</p><p>For construction activities, it is assumed that the two lanes in each direction are both closed so that all trafc takes detour,with a speed reduction from 65 mph to 40 mph, and a longer travel distance of 2.4 km (1.5 mi) (Zhang et al., 2010). For therehabilitation periods, it is assumed that only one lane will be temporarily closed. These assumptions are fed into QuickZonemodel to estimate the parameters like detour rate, queue length, speed reduction within work zones. Trafc delays are thencoupled with fuel consumptions and vehicle emissions to measure their environmental impacts. The fuel consumption andenvironmental burdens are calculated using Eq. (1).</p><p>For the usage module roughens effect, pavement structure effect, albedo, and carbonation, are considered separately. Forroughness effects, the IRI developing trend for the three options are estimated by MEPDG model (Fig. 2).</p><p>It is assumed that IRI is restored to its initial values when rehabilitation activity of every 16 years is performed. The LCAinventory is calculated as the differences between driving on real pavement and on an ideally smooth pavement. Accordingto Chandras research (2004), highway capacity is reduced by approximately 150 vehicles per hour per lane when IRI is in-creased by 1 m/km. Under the IRI development scenarios from Fig. 2, the potential highway capacity reductions are esti-mated accordingly, which are then reected into the QuickZone model to estimate the possible delay and amounts ofdetours. A typical torque curve for an engine (Tunnell and Brewster, 2005) is used to estimate the emission due to additionalfriction and vertical acceleration of the vehicle body, which denes a not-to-exceed zone. Constrained by this zone, a con-stant emission rate is assumed for a typical operation speed (90105 km/h). Any additional emissions produced from engineload increase can be estimated as proportional to the fuel consumption increase calculated by Eq. (2).</p><p>Pavement structure effects are based on Taylor et al. (2000) and applied using Floridas temperature range, Table 1 (theCSOL option is treat...</p></li></ul>


View more >