Analysis of fatigue properties of unmodified and polyethylene terephthalate modified asphalt mixtures using response surface methodology

7/26/2019 Analysis of fatigue properties of unmodified and polyethylene terephthalate modified asphalt mixtures using resp

1/11

Analysis of fatigue properties of unmodied and polyethyleneterephthalate modied asphalt mixtures using response surfacemethodology

Mehrtash Soltani a,, Taher Baghaee Moghaddam a,b, Mohamed Rehan Karim a, Hassan Baaj b

a Center for Transportation Research, Department of Civil Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysiab Centre for Pavement and Transportation Technology, Department of Civil and Environmental Engineering, Faculty of Engineering, University of Waterloo,Waterloo N2L 3G1,

Canada

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

Article history:

Received 18 May 2015Received in revised form 10 August 2015Accepted 10 September 2015Available online 12 September 2015

Fatigue is a major distress mode ofexible pavements that generally occurs in the form of irreg-ular (alligator) cracking in the wheel paths. This paper evaluates the effects of applied stress andtemperature on the fatigue l ives of polyethylene terephthalate (PET) modied asphalt mixturesusing response surface methodology (RSM). As it is shown in this study a quadratic model issuccessfullytted to the experimental data. Fatigue lives of mixtures are inuenced by changesin selected parameters. In addition, the effect of temperature variation is more drastic on thefatigue lives than the effects of stress level and modier content.

2015 Elsevier Ltd. All rights reserved.

Keywords:

Asphalt mixtureWaste PETEnvironmental temperature

Response surface methodology

1. Introduction

Road pavement is subjected to external loads including mechanical loading induced by heavy trafc and thermal loading inducedby thermal changes. The applied loads, along with environmental conditions result in pavement deterioration which, in some cases,happens even before its expected service life. Pavement damage is usually occurred in the form of permanent deformation (surfacerutting), fatigue failure and low temperature cracking. Fatigue failure is a common mode of distress of pavement structures whichis caused by successive tensile strain induced by repeated trafc loadings [1]. This form of distress mostly appears as cracking damagewhich initially occurs at the bottom of asphalt layer where the tensile stresses are maximum. Then these cracks spread to the surfaceof asphalt mixture. Previous studies showed the fatigue life of asphalt mixture has correlation with the mode and amount of appliedloads as well as environmental temperature[2,3].

Stone mastic asphalt (SMA) is gap-graded asphalt mixture which has been developed in Germany in 1960s[4]. It has a highpercentage (60 to 80%) of coarse aggregate, greater than 5 mm in size, high binder content (5.5 to 7% by weight), high percentageof mineralller (7 to 11%), and added bers (1%) [5]. Due to inherited structure of SMA, it can provide better permanent deformation(rutting) performance and durability compared to conventional dense-graded mixture[6,7]but it becomes controversial in case offatigue property. However some studies showed that SMA mixture had lower fatigue life [8,9], others concluded that it had better fa-tigue properties compared to the conventional mixture[10,11]. In SMA mixture in order to prevent draindown (due to high asphaltcontent) and improve mixture performance stabilizer additives, bers and polymers are used. In this case, using polymer in asphalt

Engineering Failure Analysis 58 (2015) 238248

Corresponding author.E-mail address:[email protected](M. Soltani).

http://dx.doi.org/10.1016/j.engfailanal.2015.09.0051350-6307/ 2015 Elsevier Ltd. All rights reserved.

Contents lists available atScienceDirect

Engineering Failure Analysis

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / e f a
http://dx.doi.org/10.1016/j.engfailanal.2015.09.005http://dx.doi.org/10.1016/j.engfailanal.2015.09.005http://dx.doi.org/10.1016/j.engfailanal.2015.09.005mailto:[email protected]://dx.doi.org/10.1016/j.engfailanal.2015.09.005http://www.sciencedirect.com/science/journal/http://www.sciencedirect.com/science/journal/http://dx.doi.org/10.1016/j.engfailanal.2015.09.005mailto:[email protected]://dx.doi.org/10.1016/j.engfailanal.2015.09.005http://crossmark.crossref.org/dialog/?doi=10.1016/j.engfailanal.2015.09.005&domain=pdf


2/11

mixture is very common[1215]. Tapkn has utilized polypropylenebers as reinforcementin asphalt mixtureand it wasrealized thatthe mixture fabricated by polypropylene bers had better performance in comparison with control mixture[16].

In many cases, using polymers causes higher construction cost due to high polymer cost. In order to overcome this problem, manystudies have used waste polymers in asphalt mixtures[13,1720].One of the important industrial plastic materials is polyethyleneterephthalate (PET). PET is a semi-crystalline thermoplastic polymer material which has been used in beverage and food industriesfor years. Currently, a large amount of waste PET is being produced worldwide and it is going to cause a serious environmentalchallenge due to its non-biodegradability[21]. Hence, some studies have been previously performed to evaluate the effects of usingpost-consumer PET as secondary materials in road pavement in order to tackle this potential environmental hazard and, moreover,to decrease construction cost imposed by application of polymers in asphalt mixture[2,13,22,23].

Mathematical modeling is useful for real-world application as it is robust in terms of its ability to deal with many constraints andobjectives [24,25]. In addition, using statistical analysis in pavement engineeringis increasing among engineersand designersbecauseit helps to have better perspective about the pavement performance parameters. In this case, factorial design of experiments (DOE)which throughthe useof techniques such as responsesurface methodology (RSM) simultaneously consider severalfactors at differentlevels, and give a suitable model for the relationship between the various factors[2630].

The aim of this study is examining the fatigue property of SMA mixtures at elevated temperatures and stress levels for theunmodied and PET modied mixtures followed by nding interactions between these fundamental factors using RSM basedon central composite design (CCD).

2. Materials and methods

SMA mixtures were fabricated using 80/100 penetration grade asphalt cement. Granite-rich aggregate particles were used for thisinvestigation. Ninepercent ofller was utilized. The aggregate particle size distribution is shown in Fig. 1. As itis shown inthis gure,the SMA mixture contains coarser aggregate particle (68.5% of particles are greater than 4.75 mm) which provides better stone onstone contact. In order to have better understanding of the materials' characteristics several tests were performed on asphalt cementand aggregate particles and the results are listed in Table 1.Ascanbeseenin Table 1, materials'properties aresatisfactorily passed therequirements.

PET akes which have been used for this study were obtained from waste PET bottles. For using PETakes in asphalt mixture, thePET bottles were cut to small parts and crushed using a crushing machine. Thereafter, the crushed PET particles were sieved andparticles smaller than 2.36 mm in size were used for this research.Table 2depicts physical and mechanical properties of PET.

2.1. Mixture fabrication

In order to fabricate SMA mixtures, 1100 g of mixed aggregate andller were heated inside oven at temperature of 160 C for 3 h.Asphalt cement was also heated at 130 C to be suitable for mixing with aggregate particles. Mixing temperature of 160 C was deter-mined using plot of binder viscosity against temperature (viscosity at mixing temperature must be 0.17 0.02 Pa s). PET particleswith different percentages (0%, 0.5% and 1% by weight of aggregate particles) were added directly to the mixture as the method ofdry process. It is worth mentioning that in previous research it was believed that due to the high melting point of PET wet process(adding modier to the asphalt cement) cannot be appropriate because it might hinder the mixing[17]. The loose mixture wascompacted using Marshall compactor and 50 blows of compaction effort were applied on each side of the mixture. It should be men-tioned that all the mixtures were fabricated at their optimum asphalt contents using Marshall mix design method[22,31,32]and theresults are presented inTable 3.

Fig. 1.Aggregate particle size distribution for stone mastic asphalt.

239M. Soltani et al. / Engineering Failure Analysis 58 (2015) 238248


3/11

2.2. Indirect tensile fatigue test

Indirect Tensile Fatigue Test (ITFT) was carried out in the controlled stress mode according to BS EN 12697-24. Universal TestingMachine (UTM) which is a computer controlled system was used for running ITFT. Compressive cyclic load was applied along withdiametrical section of specimen in the form of haversine waveform with 500 ms repetition time and 100 ms pulse width(see Fig. 2). ITFT was conducted at stress levels of 200 kPa, 300 kPa, and 400 kPa which are the stress values mostly used in pavementlaboratories. In addition, temperatures of 10 C, 25 C and 40 C are designated in this study to simulate the pavement temperaturerangesthat fatigue damageusually occurs. Prior to thetest, all thespecimens were conditionedat thecontrolled temperature chamberfor about 2 h to reach the desired test temperature. Fatigue life was dened as the number of load repetitions reached when thespecimen splits[2,3335].

2.3. Method of analysis

One-factor-at-a-time (OFAT) methodology is a conventional approach for optimizing multifactor experiments. OFAT is a change-able single factor method for a specic experiment design while other factors are kept constant. OFAT is unable to generate appropri-ate output because theeffects of interaction amongall factors in thedesign are notexamined truly, and so it is not capable of reachingthe true optimum value[36,37]. Hence, response surface methodology (RSM) has been introduced for parameter optimization in away that number of experiments and interaction among the parameters are reduced to minimal value [3840]. Consequently, DesignExpert 9.0.5.1 was designated for this purpose to generate statistical analysis and experimental designs and to calculate the sorbentadaption conditions.

For this study, a developed quadratic model using RSM was suggested by the software for design and data analysis. In this inves-tigation, the effects of three independent numerical variables including PET modier (A) from zero to 1%, stress levels (B) from200 kPa to 400 kPa and temperatures between 10 C and 40 C, all at three levels, were studied through the central composite design

(CCD). Related literature and preliminary studies were used to choose these variables and the respective regions of interest[2,3,33].

Table 1

Properties of materials.

Property Unit Used specication Value Requirements

Asphalt

Penetration at 25 C 0.1 mm ASTM D 5 87.5 Softening point C ASTM D 36 46.6 Flash point C ASTM D 92 300 Fire point C ASTM D 92 320

Specic gravity (g/cm3) ASTM D 70 1.03

Coarse aggregate

L.A. abrasion % ASTM C 131 19.45 b30Flakiness index % BS 812 Part 105.1 2.72 b20Elongation index % BS 812 Part 105.2 11.26 b20Aggregate crushing value % BS 812 part 3 19.10 b30Bulk specic gravity (g/cm3) ASTM C 127 2.60 Absorption % ASTM C 127 0.72 b2

Fine aggregate

Bulk specic gravity (g/cm3) ASTM C 128 2.63 Absorption % ASTM C 128 0.4 b2Soundness loss % ASTM C 88 4.1 b15

Table 2

Physical and mechanical properties of PET.

Property Unit Method Value

Water absorption % ASTMD570 0.11Tensile strength psi ASTM D638 11,500Tensile modulus psi ASTM D638 4 105

Elongation at break % ASTM D638 70Flexural strength psi ASTM D790 15,000Flexural modulus psi ASTM D790 4 105

Approx glass transition temperature C 75Approx melting temperature C 250

Specic gravity g/cm3 ASTM D792 1.35

240 M. Soltani et al. / Engineering Failure Analysis 58 (2015) 238248


4/11

Table 4 shows the levels and range of the actual values of independent numerical variables. By usingEq. (1) all dened numericalvariables transformed to the coded form.

xi XiX0

X 1

where xidescribes the coded value of the ith independent factor which is dimensionless. Actual value is dened as Xi, X0is thecenter point actual value and X refers to the step change of the ith variable.

A total of 34 experiments in random order were performed, together with ve replications at the center points to provide accurateassessment of errors (Table 4). The fatigue life was dened as the response to develop design of experiment modeling.

(Eq. (2)) was developed to calculate the dependent variables[41,42]:

Y0X

n

i1ixi

Xn

i1iix

2iX

n

i1

Xn

j1ijxixj 2

InEq. (2), Y is the calculated response,0is constant value. Independent variables in coded form are described as x i, and xj. Thecoefcients ofi andii are the linear and quadratic terms.ij is the interaction term coefcient, is therandom error, and thestudiednumber of factors is dened as n.

In addition, in order to assesstheappropriatenessof proposed model, analysis of variance (ANOVA) was performed. The coefcientsof determination (R2 and R2adj) express the wellness of the t to the suggested model. These values can be determined using thefollowing equations[43]:

R2

1 SSresidual

SSmodelSSresidual3

R2adj1

SSresidual.

D FresidualSSmodelSSresidual

.D FmodelD Fresidual

: 4

Table 3

Summary of mix design.

PET(%) BSGa VMAb (%) VFAc (%) OACd (%)

0 (unmodied) 2.294 18.12 77.92 6.770.5 2.296 17.34 76.90 6.361 2.283 17.55 77.20 6.51

a Bulk specic gravity of compacted mixture.b

Void in mineral aggregate.c Void lled with asphalt.d Optimum asphalt content.

Fig. 2.Indirect Tensile Test loading set-up.



5/11

In this equation,SSis the sum of squares andDFis degrees of freedom.Eq. (5),Eq. (6)and an F-test in the program were used to check the model's adequate precision ratio (AP) to determine the

statistical importance of the model[44]:

Adequate PrecisionmaxY min Y ffiffiffiffiffiffiffiffiffiffi

V Y p 5

V

Y 1n

Xn

i1V

Y p2

n 6

where Yis thepredicted response,Prepresents thenumberof model parameters, residual mean squareis describedas 2,andnis the number of experiments.

After the F-test had been performed, the insignicant terms were found and eliminated from the model. Thereafter, the nalizedmodel was introduced based on the signicant variables.

Table 4

Experimental design layout and experimental results of the responses.

Run Factor 1: PET (%) Factor 2: Stress level (kPa) Factor 3: Temperature (C) Fatigue cycles

1 0 200 10 196,7202 1 400 40 15413 0.5 300 10 184,5214 0 200 40 43415 1 300 25 7752

6 0.5 300 25 60727 0.5 200 25 51,4218 0.5 400 25 12439 0 300 25 301910 0.5 300 25 606311 1 400 10 167,28112 0 300 25 303313 1 200 10 385,86614 0 400 10 91,29115 0 200 40 432916 0.5 300 25 606117 0.5 300 25 605918 1 400 40 154419 0.5 300 25 608820 0 400 40 86621 1 200 10 379,731

22 1 300 25 773923 0.5 300 40 313324 0.5 300 40 314125 0 400 40 87826 0.5 300 10 179,49127 1 400 10 167,31228 0.5 400 25 123929 1 200 40 782930 0 400 10 91,30231 0.5 200 25 53,32032 1 200 40 793133 0.5 300 25 607734 0 200 10 188,821

Fig. 3.Fracture patterns (left: ideal fracture, right: single cleft fracture).



6/11

3. Results and discussion

As it was mentioned by Allen at al., three different modes of failure have been observed for the ITFT. In the rst mode which hasbeen observed in most cases the specimen fractured completely; however, in the second mode specimen did not fail due to the frac-ture and no visible crack was observed. In such case, the failure was attributed to accumulation of permanent vertical deformation.Additionally, the third mode of failure was dened by indentation of the loading strip into the specimen [34]. In this study, ITFTwas carried out in the controlled stress mode. The ITFTtest was conducted on the mixtures at elevated temperatures and stress levels,and thefracture patterns are shown in Fig. 3.Asitcanbeseeninthis gure two types of fracture are observed known as ideal fractureand single cleft fracture[35].Table 4represents the layout for experimental design and the fatigue lives responses. According to thistable the fatigue lives vary between 866 and 385,866 cycles. Having these values, RSM was utilized to nd interactions between theoutputs and variables which are independent. Eventually, after a regression analysis had been appliedto all responsesdescribed in thedesign matrix, a tted quadratic polynomial equation was produced. The highest order polynomials in which the additional termswere signicant and the models were not aliased were suggested by software. The numerical parameters (A, B and C) were used togenerate the predictive model according toEq. (7):

Final Log10 fatigue equation3:8 0:15A0:37B0:92C 0:57C2: 7

Checking the adequacy of the model is an important part of the data analysis, as the model functions would give improper re-sponses in case the t is not adequate[39,45]. Hence, in this study, in order to assess the signicance and adequacy of the model,ANOVA analysis was performed and the results are reported inTable 5. In addition, this table shows the quadratic models forcoded factors, and represents other statistical parameters in logarithmic scale for the fatigue life. In this table, p-values which areless than 0.0001 imply that the model and parameter are signicant (model and term with p-value b0.05 indicate the model andthe term are signicant for 95% condence intervals) for assessing the value of responses[46].

As the results show, PET (A), stress level (B), temperature (C), C2 are signicant terms with p-values less than 0.05. However, A2,B2, AB, AC and BC were insignicant (p-value N0.100). Therefore, in order to improve the model and optimize the results, the insig-nicant term can be removed from the model[47].

In order to check the tness of model regression coefcients, R2 and R2adjwere calculated. Values of 0.9579 and 0.9422 were ob-tained for R2 andR2adj, respectively. This shows that 94.22% of the total variation in thefatigue life response could be explained by thequadratic model. The high R2 and adjusted R2 values indicate that there is a good agreement between predicted and actual values[40,41,48]. Ratio of signal-to-noise is measured by adequate precision to compare the variety of the estimated amounts at the designpoints withthe average prediction error. Adequate model discrimination wasfound in this study when the adequate precision ratio of25.936 was calculatedwhich is much higher than the value of 4 [49].Thelackoft (LOF) F-test was also used to evaluate the adequacyof the model. LOF depicts the variation of the data around the tted model, and the amount of LOF would be signicant if the modeldoes nott thedata well. It is worth noting that despite theLOF being signicant, a reasonable agreement between the predicted andadjusted R2 were found for all responses and it can be concluded that the models suggested for all responses can be used to navigateinto design space to nd an optimum condition[50,51].

3.1. Statistical analysis

In order to have better understating about model satisfactoriness, diagnostic plots such as the predicted versus actual values areworthwhile.Fig. 4shows the actual versus predicted values of parameters for fatigue modeling. As shown in thisgure there is anadequate agreement between the actual data which were obtained through experiment and the predicted ones. This agreement

Table 5

ANOVA analysis for fatigue life.

Source Sum of squares df Mean Square F value p-valueProb NF

Model performance

Model 22.74 9 2.53 60.73 b0.0001 SignicantA PET 0.44 1 0.44 10.48 0.0035 SignicantB Stress level 2.76 1 2.76 66.32 b0.0001 SignicantC Temperature 16.76 1 16.76 402.84 b0.0001 SignicantAB 5.472E004 1 5.472E004 0.013 0.9096 InsignicantAC 7.234E004 1 7.234E004 0.017 0.8962 InsignicantBC 0.13 1 0.13 3.12 0.0899 InsignicantA2 0.082 1 0.082 1.96 0.1739 InsignicantB2 0.051 1 0.051 1.24 0.2772 InsignicantC2 1.74 1 1.74 41.83 b0.0001 SignicantResidual 1.00 24 0.042Lack oft 1.00 5 0.20 9017.24 b0.0001 SignicantPure error 4.205E004 19 2.213E005Cor total 23.73 33

Adequate precision (AP) 25.936



7/11

can also be understood by AP value (AP N 4) for the fatigue responses (see Table 5). This veries that thepredicted model can be usedto navigate the design space dened by the CCD.

3.2. One factor analysis

One factor analysis is changing one factor at a timemethod. That is to say, in this method a single factor varies while all otherfactors are kept constant for a particular set of experiments. This process exists for optimizing other variables which would be timeconsuming. In this method, trial and error commonly exist for the optimization of variables, and, moreover, there is always a lackof reaching a true optimum amount which is obtained by seeing the interaction among different variables[50,52]. Furthermore, inone factor analysis when the software evaluates one parameter, other parameters are kept constant at their middle ranges. For in-stance, in case of PET content evaluation, temperature and stress level are kept constant at 25 C and 300 kPa respectively.

Figs. 5, 6 and 7 show the one factor analysis of PET percentage, stress level and temperature on logarithmic scale of fatigue life re-

spectively. The logarithmic scale of fatigue life is shown for better underrating of difference between values. Fig. 5indicates that byincreasingthe PET the fatiguelife is also increased. A possible reason for this result might be the mechanical properties of PET particlesin themix. In fact, because themeltingpoint of PET is high (over 250 C) and is higher than the mixture'sfabrication temperature, thePET particles do not melt during mixing. The solid PET particles can make mixture more elastic and cause higher fatigue life underloading application. For another factor (Fig. 6) it can be observed that by increasing stress level thefatigue life is decreased. Same pat-tern is found for temperature when by increasing the temperature the fatigue life is decreased (Fig. 7).Fig. 7also depicts that

Fig. 4.Design-expert plot; predicted vs. actual values plot for fatigue life (logarithmic scale).

Fig. 5.Effect of PET percentage on the fatigue life (logarithmic scale).



8/11

increasing the temperature has negative effect on the fatigue life and that at higher temperatures (over 30 C) the fatigue life isshifting to a constantvalue. This represents the importance of ambient temperature on the fatiguelife of asphalt mixture. Thendingsof this study are based on controlled-stress test mode which are in support of previous studies [8,5356] that found the fatigue life ofasphalt mixtures increased at lower temperatures.

3.3. Effects of temperature, stress level and PET variables on the fatigue life using response surfaces

Three-dimensional response surface plots of the predictive quadratic model for the effect of stress level and temperature onlogarithmic scale is presented inFig. 8. The response surfaces were generated based onEq. (7).

Fig. 8indicates at higher temperature and stress level the fatigue life is decreased. The variation of temperature for all stress levelseems to be signicant. In physical denition, when the ambient temperature increases, the asphalt binder becomes less stiff which

may weaken the fatigue resistance of asphalt mixtures and resultsin lower fatigue life. On the other hand thevariation of stress levelsat higher temperatures is less effective on the fatigue life compared to lower temperature. That is to say, the changes in fatigue livesare more tangible at lower stress levels and temperatures.

Fig. 9 indicates the effect of temperature and PET percentage on the SMA mixtures. Overall, increasing temperature had a negativeeffect on the fatigue life. However, the effect of adding PET for improving the fatigue life is highlighted. Changes in fatigue life cannotbe attributed to the mixture air void content because all the mixtures were fabricated at their optimum asphalt contents with 4% air

Fig. 6.Effect of different stress levels on the fatigue life (logarithmic scale).

Fig. 7.Effect of different temperature on the fatigue life (logarithmic scale).



9/11

voids. In addition, improvement of fatigue life cannot be due to the higher asphalt content in the mixture because as it is shown inTable 3all the modied mixtures have lower asphalt contents than the unmodied mixture.

The correlation between stress level and PET content on the fatigue life of SMA mixture is shown in Fig. 10. Higher fatigue life isfound for the modied asphalt mixture associated with lower stress levels. By reducing the amount of PET in asphalt mixture thefatigue life is decreased at all stress levels. In contrast, by increasing the stress level asphalt mixture experienced lower fatigue lifeat all PET content. It can also be concluded that both PET increment and decrease in the stress level have roughly the same effecton the fatigue life of asphalt mixture.

4. Conclusions

This paper aimed to evaluate the effect of applied load and temperature on the fatigue lives of unmodied and PET modiedasphalt mixture. Statistical analysis was used in this investigation to nd the interaction between selected variables. A goodagreement was found between predicted and actual values which indicated second-order response surface models provide a suitablemodel to predict the fatigue life values within the range of dened factors. Based on the results achieved in this study the followingconclusions can be derived:

(1) The results showed that the changes in the fatigue lives are more tangible at lower stress levels and temperatures.(2) Both PET increment and decrease in the stress level have roughly the same effect on the fatigue life of asphalt mixture.(3) The effect of temperature on the fatigue lives is more drastic compared to stress level and PET content.

Fig. 8.Effects of stress level and temperature on the fatigue life (logarithmic scale), 0.5% PET.

Fig. 9.Effects of PET percentage and temperature on the fatigue life (logarithmic scale), 300 kPa stress level.



10/11

Acknowledgements

The authors would like to thank to the University of Malaya Research Fund (Project No. RP010A-13SUS) for providing theopportunity to make this research project.

References

[1] H. Di Benedetto, C. De La Roche, H. Baaj, A. Pronk, R. Lundstrm, Fatigue of bituminous mixtures, Mater. Struct. 37 (2003) 202216.[2] T. Baghaee Moghaddam, M.R.Karim, T. Syammaun, Dynamic properties of stone mastic asphalt mixtures containing waste plastic bottles,Constr. Build. Mater. 34

(2012) 236242.[3] G.G.Al-Khateeb, K.A.Ghuzlan, The combined effect of loading frequency, temperature,and stress level on the fatiguelife of asphalt paving mixtures using the IDT

test configuration, Int. J. Fatigue 59 (2014) 254261.[4] H. Baaj, D. Perraton, H. Di Benedetto, M. Paradis, Contribution l'etude de la relation entre le module complexe et la rsistance la fatigue et l'ornirage d'un

enrob SMA, Proceedings of the 48th Annual Conference of the CTAA, Halifax NS, November 1719, 2003.[5] H. Baaj, M. Paradis, Use of post-fabrication asphalt shingles in stone matrix asphalt mix (SMA-10): laboratory characterization and field experiment on autoroute

20 (Quebec), Proceedings of the 53rd Annual Conference of the Canadian Technical Asphalt Association, Saskatoon SK, November 1619, 2008.[6] R.B. Schmiedlin, Stone matrix asphalt: the Wisconsin experience, Trans. Res. Rec. 1616 (1988) 3441.[7] I.M. Asi, Laboratory comparison study for the use of stone matrix asphalt in hot weather climates, J. Constr. Build Mater. 20 (2006) 982989.[8] F.M. Nejad, E. Aflaki, M.A. Mohammadi, Fatigue behavior of SMA and HMA mixtures, J. Constr. Build Mater. 24 (2010) 11581165.[9] T.B. Moghaddam, M.R. Karim, A. Mahrez, A review on fatigue and rutting performance of asphalt mixes, J. Sci. Res. Essays 6 (4) (2011) 670682.

[10] D. Perraton, H. Baaj, H. Di Benedetto, M. Paradis, Evaluation de la rsistance la fatigue des enrobs bitumineux fonde sur l'evolution de l'endommagement dumatriau en cours de l'essai : aspects fondamentaux et application l'enrob SMA, Can. J. Civ. Eng. 30 (2003) 902913.

[11] D. Perraton, H. Baaj, A. Carter, Comparison of some pavement design methods from a fatigue point of view. Effect of fatigue properties of asphalt materials, RoadMater/ Pavement 11 (2010) 833861.

[12] A.I. Al-Hadidy, T. Yi-qiu, Mechanistic approach for polypropylene-modified flexible pavements, J. Mater. Des. 30 (2009) 11331140.[13] E. Ahmadinia, M. Zargar, M.R. Karim, M. Abdelaziz,P. Shafigh, Using waste plastic bottles as additive for stone mastic asphalt, J. Mater. Des. 32 (2011) 48444849.[14] C. Brovelli, M. Crispino, J. Pais, P. Pereira, Assessment of fatigue resistance of additivated asphalt concrete incorporating fibers and polymers, J. Mater. Civ. Eng. 2

(6) (2014) 554558.[15] C. Brovelli, M. Crispino, J. Pais, P. Pereira, Using polymers to improve the rutting resistance of asphalt concrete, J. Constr. Build Mater. 77 (2015) 117123.[16] S. Tapkin, The effect of polypropylene fibers on asphalt performance, Build. Environ. 43 (2008) 10651071.[17] D. Casey, C. McNally, A. Gibney, M.D. Gilchrist, Development of a recycled polymer modified binder for use in stone mastic asphalt, J. Resour. Conserv. Recycl. 52

(2008) 11671174.

[18] S. Hnslolu, E. Aar, Use of waste high density polyethylene as bitumen modifier in asphalt concrete mix, J. Mater. Lett. 58 (2004) 267

271.[19] P.T.L. Liseane, Fontes, G. Trichs, J.C. Pais, A. Paulo, A. Pereira, Evaluating permanent deformation in asphalt rubber mixtures, J. Constr. Build Mater. 24 (2010)11931200.

[20] M. Yu, G. Wu, Jinchuan Zhou and said easa, proposed compaction procedure for dry process crumb rubber modified asphalt mixtures using air void content andexpansion ratio, J. Test. Eval. 42 (2014) 111.

[21] M.N. Siddiqui, Conversion of hazardous plastic wastes into useful chemical products, J. Hazard. Mater. 167 (2009) 728735.[22] T. Baghaee Moghaddam, M. Soltani, M.R. Karim, Evaluation of permanent deformation characteristics of unmodified and polyethylene terephthalate modified

asphalt mixtures using dynamic creep test, J. Mater. Des. 53 (2014) 317324.[23] T. Baghaee Moghaddam, M. Soltani, M.R. Karim, Experimental characterization of rutting performance of polyethylene terephthalate modified asphalt mixtures

under static and dynamic loads, Constr. Build. Mater. 65 (2014) 487494.[24] S. Kikuchi, N. Kronprasert, S. Easa, Aggregate blending using fuzzy optimization, J. Constr. Eng. Manage. ASCE 138 (2012) 14111420.[25] M. Soltani, T.B.Moghaddam, M.R. Karim, S. Shamshirband, C. Sudheer, Stiffness performance of polyethyleneterephthalate modified asphalt mixtures estimation

using support vector machine-firefly algorithm, Measurement 63 (2015) 232239.[26] A. Khodaii, H.F. Haghshenas, H. Kazemi Tehrani, M. Khedmati, Application of response surface methodology to evaluate stone matrix asphalt stripping potential,

KSCE J. Civ. Eng. 17 (2013) 117121.[27] A. Kavussi, M. Qorbani, A. Khodaii, H.F. Haghshenas, Moisture susceptibility of warm mix asphalt: a statistical analysis of the laboratory testing results, Constr.

Build. Mater. 52 (2014) 511517.

[28] A. Khodaii, H.F. Haghshenas, T.H. Kazemi, Effect of grading and lime content on HMA stripping using statistical methodology, Constr. Build. Mater. 34 (2012)131135.

Fig. 10.Effects of PET percentage and stress level on the fatigue life (logarithmic scale), 25 C.

http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0005http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0005http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0005http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0010http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0010http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0010http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0010http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0015http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0015http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0015http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0015http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0020http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0020http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0020http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0020http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0025http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0025http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0025http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0025http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0030http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0030http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0030http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0035http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0035http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0035http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0040http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0040http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0040http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0045http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0045http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0045http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0050http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0050http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0050http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0050http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0055http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0055http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0055http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0055http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0060http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0060http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0060http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0065http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0065http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0065http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0070http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0070http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0070http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0070http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0075http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0075http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0075http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0080http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0080http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0080http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0085http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0085http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0085http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0085http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0095http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0095http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0095http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0095http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0100http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0100http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0100http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0100http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0105http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0105http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0105http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0110http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0110http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0110http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0110http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0115http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0115http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0115http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0115http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0120http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0120http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0120http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0125http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0125http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0125http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0125http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0130http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0130http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0130http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0130http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0135http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0135http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0135http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0135http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0140http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0140http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0140http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0140http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0140http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0140http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0135http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0135http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0130http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0130http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0125http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0125http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0120http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0115http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0115http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0110http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0110http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0105http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0100http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0100http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0095http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0095http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0090http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0085http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0085http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0080http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0075http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0070http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0070http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0065http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0060http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0055http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0055http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0050http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0050http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0045http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0040http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0035http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0030http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0025http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0025http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0020http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0020http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0015http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0015http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0010http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0010http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0005


11/11

[29] T. Baghaee Moghaddam, M. Soltani, M.R. Karim, Stiffness modulus of polyethylene terephthalate modified asphalt mixture: a statistical analysis of the laboratorytesting results, J. Mater. Des. 68 (2015) 8896.

[30] T. Baghaee Moghaddam, M. Soltani, M.R. Karim, H. Baaj, Optimization of asphalt and modifier contents for polyethylene terephthalate modified asphalt mixturesusing response surface methodology, Measurement 74 (2015) 159169.

[31] A. Khodaii, A. Mehrara, Evaluation of permanent deformation of unmodifiedand SBSmodified asphaltmixtures using dynamic creep test, Constr. Build. Mater. 23(2009) 25862592.

[32] S.F. Kalyoncuoglu, M. Tigdemir, A model for dynamic creep evaluation of SBS modified HMA mixtures, Constr. Build. Mater. 25 (2011) 859866.[33] J. Yan, F. Ni, M. Yang, J. Li, An experimental study on fatigue properties of emulsion and foam cold recycled mixes, Constr. Build. Mater. 24 (2010) 21512156.[34] B. Allen, I. Artamendi,P. Phillips, Influenceof temperature andagingon laboratoryfatigue performance of asphaltmixtures,Advanced Testingand Characterization

of Bituminous Materials, Two Volume Set, Taylor & Francis group, London, 2009, ISBN 978-0-415-55854-9.

[35] A.M. Hartman, M.D. Gilchrist, G. Walsh, Effect of mixture compaction on indirect tensile stiffness and fatigue, J. Transp. Eng. 127 (2001) 370378.[36] N.L. Frigon, D. Mathews, Practical Guide to Experimental Design, John Wiley and Sons, New York, 1997.[37] D.C. Montogomery, Design and Analysis of Experiments, John Wiley and Sons, New York, 2005.[38] A.I. Khuri, J.A. Cornell, Response Surfaces, Design and Analyses, 2nd ed. Marcel Dekker Inc., New York, 1996.[39] R.H. Myer, D.C. Montogomery,Response Surface Methodology. Process and Product Optimization Using Designed Experiment. 2nd ed, John Wiley and Sons, New

York, 2002.[40] R. Azargohar, A.K. Dalai, Production of activated carbon from Luscar char: experimental and modeling studies, Microporous Mesoporous Mater. 85 (2005)

219225.[41] M.Y. Can, Y. Kaya, O.F. Algur,Response surface optimization of the removalof nickel fromaqueous solution by conebiomass ofPinus sylvestris, Bioresour. Technol.

97 (2006) 17611765.[42] Z. Aksu, F. Gnen, Binary biosorption of phenol and chromium (VI) onto immobilized activated sludge in a packed bed: prediction of kinetic parameters and

breakthrough curves, Sep. Purif. Technol. 49 (2006) 205216.[43] B.K. Krbahti, M.A. Rauf, Response surface methodology (RSM) analysis of photoinduceddecoloration of toludine blue, Chem. Eng. J. 136 (2008) 2530.[44] B.K. Krbahti, M.A. Rauf, Determination of optimum operating conditions of carmine decoloration by UV/H2O2 using response surface methodology, J. Hazard.

Mater. 161 (2009) 281286.[45] B.K. Krbahti, M.A. Rauf, Application of response surface analysis to the photolytic degradation of basic red 2 dye, Chem. Eng. J. 138 (2008) 166171.[46] M. Zabeti, W.M.A.W. Daud, M.K. Aroua, Optimization of the activity of CaO/Al2O3 catalyst for biodiesel production using response surface methodology, Appl.

Catal. A Gen. 366 (2009) 154159.[47] V. Hosseinpour, M. Kazemeini, A. Mohammadrezaee, Optimisation of ru-promoted Ir catalysed methanol carbonylation utilising response surface methodology,

Appl. Catal. A Gen. 394 (2011) 166175.[48] U.K. Garg, M.P. Kaur, V.K. Garg, D. Sud, Removal of nickel (II) from aqueous solution by adsorption on agricultural waste biomass using a response surface

methodological approach, Bioresour. Technol. 99 (2008) 13251331.[49] T. lmez, The optimization of Cr (VI)reduction and removal by electrocoagulation using response surfacemethodology, J. Hazard. Mater. 162 (2009) 13711378.[50] S.M.S. Wan, W.M.A. Daud, A. Houshmand, A. Arami-Niya, The application of response surface methodology to optimize the amination of activated carbon for the

preparation of carbon dioxide adsorbents, Fuel 94 (2012) 465472.[51] J. Snchez-Romeu, J.M. Pas-Chanfrau, Y. Pestana-Vila, I. Lpez-Larraburo, Y. Masso-Rodrguez, M. Linares-Domnguez, et al., Statistical optimization of immuno-

affinity purification of hepatitis B surface antigen using response surface methodology, Biochem. Eng. J. 38 (2008) 18.[52] R.L. Mason, R.F. Gunst, J.L. Hess, Statistical Design and Analysis of Experiments, Eighth Applications to Engineering and Science, 2nd ed. Wiley, New York, 2003.[53] J.A. Epps, C.L. Monismith, Fatigue of asphalt concrete mixtures-summary of existing information. Fatigue of compacted bituminous aggregate mixtures, ASTM

STP508, Am. Soc. Test. Mater. (1971) 1945.[54] P.S. Pell, I.F. Taylor, Asphaltic road materials in fatigue, Proceedings of the association of the asphalt pavement technologists (AAPT), vol. 38 1969, pp. 577593

(Los Angeles, California, USA).[55] G.G.Al-Khateeb, K.A.Ghuzlan, The combined effect of loading frequency, temperature,and stress level on the fatiguelife of asphalt paving mixtures using the IDT

test configuration, Int. J. Fatigue 59 (2014) 254261.

[56] S. Bhattacharjee, R.B. Mallick, Effect of temperature on fatigue performance of hot mix asphalt tested under model mobile loadsimulator, Int.J. Pavement Eng. 13(2012) 166180.

http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0145http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0145http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0145http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0145http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0150http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0150http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0150http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0150http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0155http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0155http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0155http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0155http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0160http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0160http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0160http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0165http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0165http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0165http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0170http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0170http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0175http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0175http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0175http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0180http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0185http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0190http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0195http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0195http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0200http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0200http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0200http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0200http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0210http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0210http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0210http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0210http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0215http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0215http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0215http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0220http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0220http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0220http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0220http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0225http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0225http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0225http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0230http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0230http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0230http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0230http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0235http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0235http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0235http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0235http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0240http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0240http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0240http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0240http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0245http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0245http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0245http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0250http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0250http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0250http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0250http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0255http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0255http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0255http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0255http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0260http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0265http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0265http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0265http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0265http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0270http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0270http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0270http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0270http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0275http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0275http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0275http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0275http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0280http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0280http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0280http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0280http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0280http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0280http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0275http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0275http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0270http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0270http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0265http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0265http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0260http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0255http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0255http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0250http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0250http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0245http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0240http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0240http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0235http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0235http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0230http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0230http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0225http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0220http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0220http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0215http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0210http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0210http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0205http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0200http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0200http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0195http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0195http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0190http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0185http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0180http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0175http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0170http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0170http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0165http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0160http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0155http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0155http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0150http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0150http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0145http://refhub.elsevier.com/S1350-6307(15)30093-5/rf0145

Documents

Analysis of fatigue properties of unmodified and polyethylene terephthalate modified asphalt mixtures using response surface methodology