A Quantitative Rating System for Pollutant Emission Reduction of Asphalt Mixture

This study presents a comprehensive pollutant reduction rating system for hot mix asphalt (HMA) with three Level I indices and ten Level II indices, covering various aspects in HMA pollutant emissions, energy consumption, and exhausts from construction equipment. The pollutant emission reduction effects are investigated not only in the laboratory for modified asphalt mixtures with variousmixture gradation and binder types but also in the field for several warmmix asphalt (WMA) projects. Furthermore, energy consumption and emission data during pavement construction are obtained from 58 in situ highway projects in 10 provinces of China. Based on the hierarchical clustering method and Bayesian discriminant analysis, individual ranking systems are developed to quantify pollutant emission reduction effects and energy consumption. Subsequently, a comprehensive reduction rating system is established based on the analytic hierarchy process and approximation methods. A case study is demonstrated to implement the proposed system for the assessment of emission reduction effects.


Introduction
The development of low-carbon economy and society is currently a strategic focus around the world [1][2][3].Significant amounts of pollutant gases are emitted during asphalt concrete production and construction under high temperature conditions (more than 160 ∘ C), which not only pollute the environment but also are harmful to the health of highway constructors.Many research efforts based on various methods, such as warm mixing, cold mixing, cold paving technologies, and mixing equipment improvement of asphalt mixture, are conducted to save energy and reduce emissions in road construction.Applying pollutant emission reduction technology has been widely proved to be efficient to reduce energy consumption and decrease pollutant emissions of hot mix asphalt (HMA) mixtures during the process of mixing and construction.For example, the relevant research found that warm mix asphalt technologies can reduce the binder viscosity and enhance the asphalt mixture workability already at lower temperatures (90∼140 ∘ C), which could save 20-70% of the energy consumption when compared to hot mix asphalt, mainly due to the temperature reduction in the warm mix processes [4,5].
Researchers around the world have developed dozens of rating methods to assess pollutant emissions of HMA mixtures, which are primarily based on qualitative principles, while other researchers investigated emission reduction effect and energy consumption quantitatively via quota evaluation.Various green building assessment systems have been established in the past decade, such as BREEAM of UK [6], LEED of USA [7,8], Ecological Architecture Guidelines LNB of Germany [9], International Organization GBTool in which Canada and other nations participated [10], and NABERS in Australia [11,12].These rating systems developed specific standards and scoring systems to establish a healthy and comfortable living environment, save resources, and decrease the effect of pollutants on the environment, which are suitable for site planning, land use, energy conservation, renewable energy utilization, and indoor and outdoor environmental quality.However, these standards are not particularly customized for road construction.Most current quantitative research on energy conservation and emission reduction of HMA mixtures are mainly based on life cycle analysis (LCA).Horvath and Hendrickson [13] developed the EIO-LCA model and quantized the energy consumption and pollutant emission of HMA mixtures.Stripple [14] analyzed the effect of pavement types on the environment using all life cycle methods.On the basis of the life cycle method and expert decision, Mroueh et al. [15] obtained the rating result of environmental influence through quantitative index weight.Papagiannaki and Diakoulaki [16] determined the influencing factors of road transportation in Greece and Denmark using the logarithmic distribution method.PALATE is a calculation model developed at the University of California, which conducts sensitivity analysis on the collected data on energy consumption and pollutant emission [17].Chang'an University proposed an analysis model of energy consumption and pollutant emission using the quota method and developed a quantitative rating method for energy conservation and emission reduction effect of asphalt pavements [18].The majority of researches analyze the effects of energy conservation and emission reduction through the quota method in all life cycles, whereas others examine these effects further by analyzing energy consumption and pollutant emissions reduction.However, no corresponding classification standards regarding the energy consumption and emission reduction effect are established, and evaluations of pollutant emission reduction of asphalt mixture could not be analyzed and assessed comprehensively, which result in the limited application of pollutant emission reduction technology of asphalt mixtures.In addition, current rating methods mainly focus on a single index, and the data volume and the scope of data sources cannot satisfy the rating requirements of pollutant emission reduction of HMA mixtures.Therefore, collecting laboratory and field test data is necessary to achieve an accurate evaluation of the emission reduction and energy conservation effect.
This study develops a rating system of asphalt mixtures from three aspects, namely, pollutant emissions of asphalt mixture, energy consumption, and exhaust from construction machines, to address the aforementioned problems.The classification of indices is presented to promote energy conservation and emission reduction.The rates of pollutant emission reduction are obtained from laboratory testing, and the energy consumption is examined in the field.Individual classification standards for HMA pollutant emission reduction effects and energy consumption are developed based on hierarchical clustering method and Bayesian discriminant analysis, while the comprehensive rating system is established based on the AHP and approximation methods.A case study is demonstrated in this paper for the Haihe Tunnel project in Tianjin to evaluate the emission reduction effects of the WAM-flame retardant mixture.

Selection of Rating Indices
Different pollutant gases, such as asphalt fume, carbon monoxide (CO), carbon dioxide (CO 2 ), nitrogen oxides (NO  ), and sulfur dioxide (SO 2 ), are generated in the mixing and construction of asphalt mixtures.Construction vehicles and equipment consume energy and produce substantial amounts of exhausts.In this study, a rating system will be developed to evaluate pollutant emissions from asphalt mixtures and construction machines.
Currently, there are few accurate and intensive studies on asphalt fumes and conditions.The ultraviolet spectrophotometry method, whose absorption liquid is benzene, is commonly employed.However, this test method is complicated and expensive.Thus, asphalt fumes are excluded in the pollutant emissions of asphalt mixtures according to reasonable principles of operational evaluation.
The pollutant gas emitted from the surface of HMA mixtures is minimal since asphalt mixtures are covered during transportation.Gas detection equipment cannot easily identify pollutant emissions because it is influenced by many factors in the open environment.No accurate test method has been developed to measure the pollutant emission produced during the transportation, paving, and compaction processes.Therefore, pollutant emissions of HMA mixtures during transportation, paving, and compaction are not considered in the rating system.
Converting the amount of CO and CO 2 into equivalent CO  is challenging because gas detector sensors available in the market have different calculation method and accuracy, and the components of CO  are complex.Thus, the rating index is divided into three or four parts (NO  , CO  , and SO 2 or NO  , CO, CO 2 , and SO 2 ) to establish the rating system of pollutant emission reduction of the HMA mixture according to actual conditions of pavement construction.
Energy consumption is mainly produced in four key stages of construction of asphalt mixtures, namely, mixing, transportation, paving, and compaction.Thus, this study develops a rating system of energy consumption based on these four periods.Moreover, the mixing temperature and energy consumption of WMA mixtures are lower comparing to hot mix asphalt (HMA) mixture, whereas not the same trend follows in the mixing process for WMA and HMA mixtures.The differences should be considered in the rating system.
A rating system on reducing pollutant emissions of asphalt mixtures is developed from the three processes in constructing asphalt pavements, namely, transportation, paving, and compaction.
The specific rating system is shown in Table 1.
In addition, rating process used for hot recycling asphalt mixtures should be improved to a higher level.

Data Collection
This study presents a laboratory test equipment to test pollutant emissions and proposes a laboratory method to test pollutant emissions of asphalt mixtures.
The procedure of the laboratory test is shown as follows: the stirring system is sealed using high-temperature-resistant materials with a 3 cm to 4 cm wide aperture on the sealing materials.After stirring the asphalt mixture, the stirring system is elevated to an appropriate height to convey the pollutant gas while a paddle keeps stirring the asphalt  mixture.Pollutant gases are produced through a pipe and tested by HA-856 portable gas detector on the nozzle.After repeated experiments, the optimal experimental information is obtained, such as mixing revolution (17 times), mixing temperature (160 ∘ C), and mixing quality of asphalt mixture (12,000).The design and image of the test setup in evaluating the pollutant emissions are shown in Figure 1.  4.

Reduction
(2) Based on Testing Temperatures and Asphalt Binder Types.
Base asphalt 70# in Shandong and Shaanxi, SBS asphalt in Tianjin, and hard asphalt 20# are selected.The mixing temperatures ranging from 120 ∘ C to 140 ∘ C are chosen to determine the impact of temperatures and asphalt types on the pollutant reduction rate.The reduction rates of CO  and NO  are shown in Table 5.

Energy Consumption and Energy-Saving Effects of Construction Equipment
3.2.1.HMA Mixture.The energy consumption in mixing, transportation, paving, and compaction affects energy conservation and emission reduction effect of the entire process of pavement construction.This study evaluates the energy consumption during the whole construction process.The energy consumption data of highways located in different areas are collected and indicated in Table 6.

WMA Mixture.
Thorough surveys of the energy consumption of WMA (the most commonly used WMA types are Evotherm, WMA-Foam, Aspha-min, and synthetic zeolite at a temperature ranging from 110 ∘ C to 130 ∘ C) in the mixing process after employing a pollutant emission reduction technology in different institutions or research units are conducted in this paper.Mixing energy consumption of WMA mixture is shown in Table 7.

Clustering and Bayes Based Methods for Individual
Ranking.Classification that focuses on the reduction rate of pollutant emissions of the HMA mixture should be established to classify the effect of pollutant emission reduction on asphalt mixtures using pollutant emission reduction technology and to determine the relative ranking levels.The commonly used methods include natural breaks classification, iteration method, and hierarchical clustering [20][21][22].
Clustering analytical method is adopted to conduct clustering analysis on pollutant emission reduction and energy consumption of asphalt mixtures.The commonly used clustering methods include hierarchical clustering method, dynamic clustering method, ordinal clustering method, and fuzzy clustering method [23][24][25].The hierarchical clustering method is suitable for classifying samples with a large amount of data, and the samples can be classified without referential patterns [26].Calculation using this method is easy, and the results are definite and accurate.Therefore, this study adopts the hierarchical clustering method to establish the scales of pollutant emission reduction of asphalt mixtures.
The basic principle of hierarchical clustering method is as follows:  samples are assumed, and each sample has  measured indicators.The distance between the samples (or similarity coefficient) and the distance between groups should be defined. samples can be initially regarded as  groups (a group consists of samples).Thus, the distance between groups is equivalent to the distance between samples.Two groups with closest distances are combined to form a new group, and the distance between the new group and other groups will be calculated.The combination will be repeated according to closest distance until all the samples form one group.This combination process can be comprehensively described using the pedigree clustering diagram.
The mathematical principle is as follows:  samples are assumed to be classified into five groups called  1 ,  2 ,  3 ,  4 , and  5 .  represents the number of samples in group   ,  ()  () refers to the center of gravity of group   , and  () () is sample number  in group   .Thus, the sum of the squares of deviations of the pollutant sample in   is expressed as follows: Note.  () () and  () () are m-dimension vectors and   is a value Ward method is used to obtain the sum of the squares of deviations, exhibiting reasonable classification, and is used widely.We employed the Ward method to calculate the ranking level space in this study.
The sum of the squares of deviations is the squared distance between groups when Ward method is adopted because the two groups of the five samples are grouped as one.The distances between the samples utilize Euclidean distance.Thus,  The squared distance between  () and  () , which are the centers of gravity of   and   , respectively, is expressed by  2  : When   and   are combined into   , the recursion formula between   and other groups (  ) is expressed as follows: Analyzing the classification results is necessary to obtain a thorough understanding of the rationality of the classification by hierarchical clustering method.Bayesian discriminant analysis and Fisher method are two commonly used methods to distinguish the rationality of sample classification.Fisher method is suitable for two groups, whereas Bayesian discriminant analysis can be used in multicategory discrimination [27][28][29].This study aims to present five levels of carbon reduction effect of asphalt mixtures.Bayesian discriminant analysis is more suitable for examining the rating results and determining the consequent ranking levels.

Interval Approximation Method for Comprehensive Rating of Pollutant Emission Reduction Effects
4.2.1.Determining Scores for Rating Indices.Mechanical exhaust is a qualitative index in the rating system for evaluating the pollutant emission reduction of asphalt mixtures.Thus, relative scores must be assigned to these indices.Single factor fuzzy evaluation is conducted by expert scoring.The ranking levels and relative score limits are excellent (90∼100), good (80∼90), fair (60∼80), poor (50∼60), and very poor (0∼50).

Interval Estimation and Numbers of Intervals.
Interval estimation refers to the score range estimation by experts.The scores of an unknown index with a certain probability are presented based on the relative score range, which better reflects the actual rating situation than that of a constant value.This study selects the following process in conducting interval estimation about the rating index of the pollutant emission reduction effect of asphalt mixtures.
All indices   of different plans and random variable index   are assumed to obey the normal distribution, whose average parameter and standard division are   and .The bilateral confidence interval is expressed as follows:  Note.The data in this table are collected from the test roads in different engineering project and every piece of data refers to a project.
In the rating system for pollutant emission reduction of asphalt mixtures,  /2 is the assurance rate coefficient (usually expressed by ).This study sets 95% as its assurance rate coefficient; in other words,  = 1.96.Thus, the interval estimations of the actual test data of all rating indices are shown as follows:

Weights for the Rating Indices
Based on the AHP.In the determination process, the analytic hierarchy process (AHP) method is convenient, reasonable, comprehensive, and widely applicable.The AHP is selected to determine the weights of all indices in the rating system.The AHP provides a pairwise comparison of the importance of all factors by the 1-9 scale method to conduct an assessment.The discriminant matrix  of Level I index, discriminant matrix  1 of Level II index under the energy consumption index, discriminant matrix  2 of Level II index under the mechanical exhaust index, and discriminant matrix  3 of Level II index under the pollutant emissions index are presented as follows: Mathematical Problems in Engineering Maximum feature  max of matrices ,  1 , and  2 are 3.0369, 4.1739, and 3.2343, respectively, which are calculated by MATLAB.The maximum features of matrix  3 are three and four.
The average random consistency index  •  is obtained by repeating the feature vector calculation of the random discriminant matrix and averaging the repeated values.The  •  values of different dimensions are shown as follows: = {0, 0, 0.58, 0.90, 1.12, 1.24, 1.32, 1.41, 1.45} .
(1) Weights for Level I Indices.The calculation of the consistency index of Level I indices ( • ) is as follows: Level I index is a third-order judgment matrix, and the value of  •  is 0.58; thus, Therefore, the consistency is reasonable, as previously verified.The integrated feature vector corresponding to  max is   = (0.36, 0.05, 0.59)  , which represents the weights of all Level I indices in the rating system. (

2) Weights for Level II Indices
A The calculation of the consistency index of Level II Indices under energy consumption is as follows: Level II index is the fourth-order judgment matrix, and the value of  •  is 0.90; thus,  ⋅  =  ⋅   ⋅  = 0.0580 0.90 = 0.064 < 0.1.
Therefore, the consistency of the judgment matrix of Level II indices under energy consumption is acceptable.
After the integration, the feature vector corresponding to  max is  1 = (0.70, 0.05, 0.10, 0.15)  , which are the weights of the Level II indices of energy consumption under the rating system for pollutant emission reduction of asphalt mixtures.
B According to the previously presented method, the weight vector of Level II indices of the exhaust emitted by the construction machine is  2 = (0.17, 0.28, 0.55)  .The weight vectors of Level II indices of the pollutant emissions of the asphalt mixture are  3 = (0.40, 0.40, 0.20)  and  3 = (0.29, 0.29, 0.29, 0.13)  .The weights of all the indices are shown in Table 1.

Rating System for Pollutant Emission
Reduction.The rating system includes two ranking level indices.Thus, the approximation method of interval numbers is adopted to approach the Level I index during the rating process.The score set of Level II indices within Level I index are assumed to be Conducting dimensionless treatment of original indices is necessary because dimensions and magnitudes may vary.The treatment formula is expressed as follows: ,  = 0, 1, 2, . . ., ;  = 1, 2, . . ., .
The decision rating principle states that a shorter distance is closer to the ranking level.The level with the minimum distance is the ranking level.The calculation formula of the distance from future rating set   to ranking level   is expressed as follows: In the formula, The ranking level can be determined according to minimum distance   .

Individual Ranking for Pollutant Emission Reductions.
The pollutants included in this study include NO  , CO  , SO 2 , CO, and CO 2 .The reduction rate can be classified into five ranking levels recorded as  = ( 1 ,  2 ,  3 ,  4 ,  5 ) = (excellent, good, fair, poor, and very poor).The classification of pollutant emissions can be conducted using the method mentioned previously.The classification process of the reduction rate of NO  is shown as follows.
Based on the hierarchical clustering method, the field collected data and laboratory test data obtained are classified using SPSS.The classification results are shown in Figure 2.
The clustering results are examined by Bayesian discriminant analysis to verify its reasonability.Regarding the posterior probability of erroneous judgment, 97.4% of the original data are clustered correctly and the classification is highly accurate.The linear discriminant models are established according to the coefficient table:  The discriminant values of all ranking levels can be obtained by adopting formula (19).Critical value  is calculated by using the discriminant function.The critical values of adjacent levels are  12 = 50.13, 23 = 57.23, 34 = 39.52, and  45 = 18.07.The classification standard of the reduction rate of NO  will be obtained by integrating the critical values and considering the testing errors.
The classification standards of the reduction rates of CO  , SO 2 , CO, and CO 2 are determined based on the same method, and the results are shown in Table 8.

Energy Conservation.
This study classifies the data of the energy-saving effect in the mixing process of the WMA mixture into five ranking levels by hierarchical clustering method based on Ward method, recorded as  = ( 1 ,  2 ,  3 ,  4 ,  5 ) = (excellent, good, fair, poor, and very poor).Linear discriminant models are established according to the coefficient table:  According to formula (20), the discriminant values of all the ranking levels can be calculated using the discriminant function expression of the energy-saving effect classification.The critical values of adjacent levels are  12 = 34.26, 23 = 25.57, 34 = 36.95,and  45 = 61.39.The classification standard of reduction rate of NO  will be obtained by integrating the critical values and considering testing errors, as indicated as follows: excellent [61, 100), good [37, 61), fair [34, 37), poor [26,34), and very poor (0, 25).

Energy Consumption
(1) Mixing Process.The traditional energy sources that are used in the mixing process of asphalt mixtures are coal and heavy oil.New energy sources include coal gasification and natural gas.The energy consumption in the mixing process can be divided into five ranking levels, namely, excellent, good, fair, poor, and very poor, according to energy consumption and energy types.The classification methods are shown as follows: excellent (theoretical consumption of natural gas), good (theoretical consumption of coal gasification), fair and poor (theoretical consumption of heavy oil), and very poor (raw coal).
The classification results of ranking levels in the mixing process are shown as follows: when the upper limit of natural gas consumption is calculated, the mixing temperature will select a higher temperature, which is 175 ∘ C, and aggregate heating temperature is 190 ∘ C. In the process of calculating theoretical minimum natural gas consumption, the specific heat capacity of aggregate is selected as the lower limit with the mixing temperature and aggregate heating temperature of 160 ∘ C and 190 ∘ C, respectively.Fuel is heavy oil, and the asphalt-aggregate ratio is 5%.The measuring parameters are shown in Table 9.The calculated amount of heavy oil consumption in the mixing process of modified asphalt mixture is 7.86 kg/t according to previously presented parameters.The conversion coefficient of heavy oil to standard coal is 1.4286, whereas the conversion coefficient of natural gas to coal is 1.2143.The converted consumption of heavy oil into natural gas by using the conversion coefficient of standard coal is 9.25 m 3 /t.This value is the maximum theoretical natural gas consumption, which is used to calculate the minimum natural gas consumption in the mixing process of asphalt mixtures.The minimum consumption is 3.26 m 3 /t.The excellent level ranges from 3.26 m 3 /t to 9.25 m 3 /t of natural gas consumption in the asphalt mixing process.The final results of other classes calculated by the same method are shown in Table 10.
(2) Transportation, Paving, and Compaction Processes.Energy consumption is categorized into five ranking levels in transporting asphalt mixture, which are excellent, good, fair, poor, and very poor.Hierarchical cluster analysis and Bayesian discriminant analysis are used to establish discriminant models based on the field collected data of unit energy consumption during transportation.The analyses are shown as follows:  The discriminant values of all ranking levels can be calculated using formula (21).Critical value  will be calculated using the discriminant function.The critical values of the adjacent levels are  12 = 0.045,  23 = 0.035,  34 = 0.031, and  45 = 0.015.Thus, the classification standards of energy consumption in transporting of the asphalt mixture are obtained.
The classification standards of energy consumption in the paving and compaction process are determined by using the same calculating method, as shown in Table 11.

Exhaust Pollutants.
The regulations on the limits for exhaust pollutants from diesel engines of nonroad mobile machinery are indicated in the Chinese national Standard Limits and Measurement Methods for Exhaust Pollutants from Diesel Engines of Non-Road Mobile Machinery (I, II) (GB 20891-2007).This standard is appropriate for diesel engines of nonroad mobile machinery with less than 560 kW net power under unsteady speed.Five ranking levels ranging from excellent to very poor are used to classify the rating index for mechanical exhaust of asphalt pavement derived from transportation, paving, and compaction process.
Five ranking levels ranging from excellent to very poor are used to classify the rating index for mechanical exhaust of asphalt pavement derived from transportation, paving, and compaction process.Regarding the ranking levels of mechanical exhaust, if four items reach the standard, the ranking level is excellent.The ranking level is good, if three items reach the standard.The ranking level is fair or poor with three or two items reaching the standard.If there is no item reaching the standard, the ranking level is very poor.

5.4.1.
Rating by Single Index.The average energy consumption and reduction effect of pollutant emissions in all processes can be obtained based on the test results of the WMA-flame retardant pavement of Haihe Tunnel in Tianjin Province during the construction process.The data shown in Table 12 are obtained using interval estimation.
Five expert scores for the qualitative indices of mechanical exhausts and the relevant interval estimations are shown in Table 13.
The approximation method of interval numbers is adopted to elevate all single indices, and the calculation process of the energy-saving effect is introduced as an example.The interval estimation is [56,79], which is between [61, 100) and [37, 61) representing excellent and good.Thus, the distance between the index and excellent level is derived as follows: The distance between the index and good level is derived as follows: The calculation and comparison result indicate that the energy-saving index is excellent.The rating result of energysaving effect is excellent and those of ranking Level II indices are good.The rating result of mechanical exhaust in construction process is excellent.The rating result of reduction rate of CO  is excellent and those of NO  and SO 2 are good.

Comprehensive Rating
The distance between all levels and rating indices set of pollutant emissions can be calculated as follows:  1 = 0.080,  2 = 0.016,  3 = 0.036,  4 = 0.063, and  5 = 0.120.According to the rating principles, the ranking level by minimum   can be determined.The calculation results indicate that the minimum value is  2 , which indicates that the distance between Level II and the rating indices set is the minimum.Therefore the ranking level is good.The remaining indices are evaluated by using the same method and the results of energy consumption, quantitative mechanical exhaust, and pollutant emissions of asphalt mixture are excellent, excellent, and good, respectively.
(2) Comprehensive Ranking.Decision matrix  of pollutant emission reduction and energy-saving effect on the Haihe Tunnel project in Tianjin Province is expressed as follows: ( All the single indices are normalized, and normalized matrix  can be obtained according to the calculation result.
Decision rating matrix  will be obtained by multiplying weight  and normalized matrix : ( The distance between all levels and pollutant emission reduction and energy consumption effect of Haihe Tunnel in Tianjin Province can be calculated as follows:  1 = ∞,  2 = 0.024,  3 = 0.056,  4 = 0.067, and  5 = 0.096.According to the rating principles, the ranking level can be determined based on minimum   .The calculation results indicate that the minimum value is  2 , which determines the distance between Level II and pollutant emission reduction effect in Haihe Tunnel project in Tianjin Province is minimum.The rating of the comprehensive effect of pollutant emission reduction is classified as "good."

Conclusions
This study develops a comprehensive rating system for pollutant emission reduction and energy-saving effect of asphalt pavement construction regarding energy conservation, quantitative mechanical emission discharges, and reduction rate of pollutant emissions of asphalt mixtures with the following key contributions.
(i) Reduction rates of pollutant emissions are measured through laboratory testing for three types of pollutant reduction modifiers.The energy consumption data of all periods during the construction of 58 highways from 10 provinces are collected.
(ii) Based on the hierarchical clustering method and Bayesian discriminant analysis, this paper has established a rigorous ranking system to quantify the reduction effects of pollutant emissions of asphalt mixtures, mechanical energy consumption, and mechanical exhausts using the five ranking levels from excellent to very poor.
(iii) This study presents a rating framework for the comprehensive pollutant emission reduction effect of asphalt mixtures based on the AHP and approximation methods.The pollutant emission reduction effects of WMA-flame retardant mixture used in the Haihe Tunnel project in Tianjin Province are evaluated in accordance with the proposed rating system to demonstrate its feasibility for real-world applications.
This paper provides reference for evaluating the pollutant emission and energy conservation effect of asphalt pavement construction.Future work includes the update of system data and the development of relevant software interfaces for production application.

Figure 1 :
Figure 1: Design of the testing setup.

Table 1 :
Rating system of pollutant emission reduction of asphalt mixture.

Table 2 :
Reduction rate of NO and CO  from different modifiers and asphalt types.

Table 3 :
Reduction rate of NO  and CO  influenced by different gradations.

Table 4 :
Reduction rate of pollutant emissions of WMA.
Note.The data in this table are collected from the test roads in different engineering project and every piece of data refers to a project.

Table 5 :
Reduction rate of NO  and CO  influenced by temperature and asphalt types.

Table 6 :
Energy consumption of different process of asphalt paving construction.
Note.Every date in this table refers to the energy consumption in one project in relative province of China.

Table 7 :
Data of energy consumption in mixing process.

Table 8 :
Score limits of ranking levels of reduction rate.

Table 9 :
Calculation parameters of energy consumption in mixing process.

Table 13 :
Values of rating indices effect.
These three indicators of Level I indices are benefit indicators.After normalization, normalized matrix  could Mathematical Problems in Engineering 13 be obtained.Decision rating matrix  could be determined by multiplying normalized matrix  by weight .