Performance Evaluation of Full-Scale Sections of Asphalt

Pavements in the State of Qatar

Husam Sadek, S.M.ASCE 1; Eyad Masad, F.ASCE 2; Okan Sirin, Aff.M.ASCE 3;
Hussain Al-Khalid 4; and Khaled Hassan 5

Abstract: The population and economy in the State of Qatar have been increasing in the past 5 years. Accordingly, traffic loading has also
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

increased rapidly, which affected the performance of existing roads and highways. This high traffic loading merits consideration of the design
and construction of long-lasting pavements that require minimal maintenance. The Transport Research Laboratory (TRL) in collaboration
with the Public Works Authority (PWA) of Qatar constructed a field experiment that consisted of six different pavement sections in order to
investigate the influence of using different materials and asphalt mixture designs on performance. This paper presents a comprehensive study
for the field evaluation of the performance of these trial sections. The evaluation involved the use of the falling weight deflectometer (FWD)
and a vehicle equipped with instruments for measuring permanent deformation and cracking. These field measurements were complemented
with laboratory measurements on field cores: the dynamic modulus, flow number, and semicircular bending tests. The results revealed that the
increase in temperature between winter and summer in Qatar reduced the stiffness of asphalt mixtures by about 80%. The sections in which
polymer-modified bitumen was used had the lowest temperature susceptibility. Moreover, the results showed that the bitumen and aggregate
type significantly affected the stiffness and the trial sections’ resistance to rutting and fracture. DOI: 10.1061/(ASCE)CF.1943-5509
.0000627. © 2014 American Society of Civil Engineers.
Author keywords: Falling weight deflectometer (FWD); Dynamic modulus; Perpetual pavement; Mechanistic-empirical pavement design;
Semicircular bending; Qatar.

Introduction region. Fatani et al. (1992), Al-Abdul Wahhab et al. (1994), and
Elseifi et al. (2012) suggested the use of either a harder bitumen
The construction of highways and road networks in many countries or a polymer-modified bitumen (PMB) to mitigate poor performance
in the Middle East region has experienced remarkable development that is associated with the use of unmodified 60–70 Pen bitumen.
both in size and standards in the last 40 years. However, only a few In the State of Qatar, there have been significant efforts in the
regional research studies have been conducted to evaluate and im- past 5 years on developing better specifications for pavement ma-
prove the performance of asphalt pavements. Several of these stud- terials. These efforts were motivated by the rapid increase of traffic
ies focused on studying the effect of significant increase in traffic loading, and the realization that current specifications do not
loading and hot weather on rutting (Al-Abdul Wahhab et al. 1994, accommodate the international developments in material tests
2001). Bubshait (2001) asserted that the main reason for pavement and improvement of properties. The high traffic loading merits con-
failures in the region is that materials are not carefully selected and sideration of the design and construction of long life or perpetual
mixtures are not designed to suit regional conditions. For example, pavements. Many studies in the United States, Europe, and the Far
unmodified 60–70 Pen bitumen is used widely in the Middle East; East showed the significantly improved performance of perpetual
however, this bitumen is too soft given the high temperatures in the pavements in terms of resistance to surface distresses and deterio-
rations when compared to conventional or determinate life pave-
1
Ph.D. Student, Centre for Engineering Sustainability, School of Engi- ments (Ferne 2006; Merrill et al. 2006; Timm and Newcomb 2006).
neering, Univ. of Liverpool, Brodie Tower, Brownlow St., Liverpool L69 The concept of perpetual pavement design implies prevention of
3GQ, U.K. (corresponding author). E-mail: [email protected] the onset of deterioration in the form of rutting and fatigue cracking
2
Professor, Mechanical Engineering Program, Texas A&M Univ. at in the structural layers, which is a result of increased traffic loading
Qatar, 253 Texas A&M Engineering Building, Education City, P.O. Box and high temperatures. The design of perpetual pavements could be
23874, Doha, Qatar. E-mail: [email protected]
3 achieved by reducing the stress and strain in the pavement either by
Assistant Professor, Dept. of Civil and Architectural Engineering,
College of Engineering, Qatar Univ., P.O. Box 2713, Doha, Qatar. E-mail: increasing the thickness of the pavement layers and/or by using
[email protected] specific materials for each layer that assists in resisting the poten-
4
Centre for Engineering Sustainability, School of Engineering, Univ. of tial distresses in these layers (Merrill et al. 2006; Renteria and
Liverpool, Brodie Tower, Brownlow St., Liverpool L69 3GQ, U.K. E-mail: Hunt 2008).
[email protected] The review of literature revealed that only a few field studies
5
Regional Manager-Middle East, Transport Research Laboratory have been conducted in the countries within the Arabian Peninsula
(TRL), Qatar Foundation, Qatar Science and Technology Park, P.O. Box to evaluate the benefits of using mechanistic-empirical design and
210529, Doha, Qatar. E-mail: [email protected]
the construction of perpetual pavements (Al-Abdul Wahhab et al.
Note. This manuscript was submitted on June 9, 2013; approved on
May 7, 2014; published online on September 8, 2014. Discussion period 1994, 2001; Masad et al. 2011; Sadek et al. 2014). Most of these
open until February 8, 2015; separate discussions must be submitted for studies focused on characterization of local materials, influence of
individual papers. This paper is part of the Journal of Performance of significant increase in traffic loading on performance, and rutting
Constructed Facilities, © ASCE, ISSN 0887-3828/04014123(14)/$25.00. due to high temperature. This paper reports on the findings from a

© ASCE 04014123-1 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 comprehensive study that was conducted to investigate the effect of
material properties and mixture design on performance of asphalt
pavements.

Scope and Objectives of the Study

The Public Works Authority (PWA) of the State of Qatar has

developed several programs and initiated projects to enhance the
specifications and design of pavements. One of these programs
is the “Road Pavement Technology” project with the Transport
Research Laboratory (TRL), which involved reviewing pavement
design and performance besides developing new pavement designs
for the State of Qatar. For this purpose, a field experiment was per-
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

formed to investigate the influence of materials and mixture design

on performance. This experiment involved the construction of six
different pavement sections in Qatar. The sections were constructed
in 2010 on a route used by heavy truck traffic. The conditions, per-
formance, and mechanical properties of these sections were moni-
tored after 3 years of service by conducting the following tests and
measurements:
• Falling weight deflectometer (FWD) tests during the spring and
summer seasons in order to evaluate the stiffnesses of pavement
layers;
• Automatic road analyzer vehicle in order to measure pavement
rutting;
• Dynamic modulus (jE! j) test on field cores to assess the stiffness
of asphalt mixtures;
• Flow number (FN) test on field cores to assess the resistance to
rutting; and
• Semicircular bending (SCB) test in order to evaluate the fracture
Fig. 1. Location of trial sections in Qatar (image © 2014 Google; map
resistance of the mixtures.
data © 2014 Google)
In addition to the preceding tests, mechanistic-empirical analy-
sis was performed to evaluate the performance of these pavement
sections after 3 years and also after 20 years.
design asphalt content at which the refusal density is 3% in the
voids in mineral aggregate (VMA). This requirement is supposed
Description of Field Trial Sections to ensure that deformation will not occur to reduce percentage air
voids to lower than 3% in the field. This design method includes
Six different asphalt pavement sections, about 150 m lengths each, other criteria to achieve durable mixture and field compaction.
were constructed in 2010 as a part of an access road to a sand
processing plant in the south of Qatar. The location shown in Fig. 1
was selected due to its high traffic loading on the trial sections.
Fig. 2 shows all trial sections were designed and constructed to
have thick asphalt layers. As discussed before, the sections were
designed in order to evaluate the influence of type of bitumen, type
of aggregate, and type of mixture design on performance.
Fig. 3 illustrates the layers and materials used in each section.
The aggregate used in the surface course for all trial sections was
Gabbro imported from United Arab Emirates. This is an igneous
rock that has been used in road construction and concrete structures
for long time in this region. The difference among the surface
course layers is in the mixture design and bitumen type. The asphalt
base course used in these sections differed in bitumen type, aggre-
gate type, and mixtures design. The same granular subbase with
limestone aggregate was used for all sections. The estimated design
modulus for the subbase is 450 MPa (TRL Report C 2010; TRL
Report D 2010). The subgrade is weathered limestone with a design
modulus of 200 MPa (TRL Report C 2010; TRL Report D 2010;
Sadek et al. 2012).
Section 4 is the control section that was designed based on the
standards of Qatar Construction Specifications (QCS), which is
essentially the Marshall method. The percentage refusal density
Fig. 2. (a) Typical cross section of trial sections; (b) core extracted
(PRD) design method [BS EN 12697/32:2003 (2003); TRL 2002]
from the trial site
was used in all other sections. The PRD test is used to determine the

© ASCE 04014123-2 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. Layers and materials’ properties of all trial sections

Section 5 included the use of Shell Thiopave, which is a sulphur in Fig. 4(a). This vehicle is instrumented to measure rutting. The
extended asphalt, while Section 6 included PMB with an SBS FWD test, shown in Fig. 4(b), was also conducted to evaluate pave-
modifier. Pen 60–70 base bitumen was used in developing the ment structural condition by acquiring the deflections of each layer.
modified asphalts. Limestone, which is a local aggregate in Qatar, These deflections were used to back-calculate the moduli of the
was used in the base course in Section 3 to compare it with the various layers [asphalt concrete (AC), granular subbase, and sub-
performance of sections in which Gabbro was used. One day after grade] using Elmod6 software (Dynatest Elmod6 version 6.1.44).
paving, three pairs of cores were taken from each trial section and The FWD test was conducted in February (spring season) and
tested for compositional analysis as shown in Table 1. August (summer season) in 2012 to monitor the temperature sus-
Since the opening of the trial road in August 2010, about 1,800 ceptibility of the trial sections at low and high temperatures.
trucks pass daily in each direction; about 50% of the trucks are fully Fig. 5 shows ten cores with each core having a diameter of
loaded with sand (45 t). Based on the traffic data, three axle con- 150 mm and a height of 320 mm that were extracted from each
figurations of trucks are passing the trial road 6 days a week: five- trial section.
axle; four-axle; and three-axle trucks. The total axle load equivalent For laboratory testing, a total of 12 field cores, 2 replicate sam-
factors for these trucks were calculated to be 11.33, 10.5, and 4.5, ples from each section, were tested to determine the jE! j and the FN
respectively. These values are considered to be very high, which led [AASHTO designation: TP 79-11 (AASHTO 2011)] using the as-
to the high 20-year ESALs of 115.4 million. Table 2 summarizes phalt mixture performance tester (AMPT) as shown in Fig. 6(a).
the traffic loading calculation. Up to the point when the field mea- Another six cores, one from each section, were subjected to the
surements were performed in summer 2013, traffic has reached monotonic SCB test using the facility at the University of Liverpool
approximately 17 million ESALs. in United Kingdom to assess the fracture resistance of each section
as shown in Fig. 6(b).

Field Testing and Results FWD Test Results

In order to monitor the performance of the trial sections, several The FWD test was conducted twice on the trial road. The first test
field and laboratory tests were conducted on these sections. The was conducted in the spring season (February 2012), the average
field work started in January 2012 with collecting pavement air temperature was about 23°C while the average surface temper-
condition data by the automatic road analyzer vehicle as shown ature was about 25°C. The second test was conducted in the

Table 1. Compositional Analysis Summary for All Trial Sections

Mix Bitumen content Stability Flow Stiffness VIM VMA VFB Filler=
Section design (%) (kN) (mm) (kN=mm2 ) (%) (%) (%) bitumen ratio
Surface course
1 PRD 3.9 14.8 2.6 6.2 4.0 14.2 71.7 1.49
2 PRD 3.8 14.7 2.6 5.7 4.9 14.3 69.1 1.35
3A
3B QCS 3.8 13.4 2.7 5.2 5.3 14.8 64.4 1.11
4
5 PRD 3.8 14.4 2.5 5.8 4.7 14.1 68.5 1.42
6
Base course
1 PRD 3.6 15.2 3.1 4.9 4.1 13.3 69.0 1.75
2 PRD 3.4 14.1 2.6 5.5 4.5 13.2 65.8 1.71
3A PRD 4.4 11.5 2.9 4.0 4.2 14.2 71.2 1.23
3B
4 QCS 3.5 14.1 2.6 5.4 4.8 13.7 64.9 1.31
5 QCS 3.9 NA NA NA NA NA NA 1.13
6 PRD 3.5 15.2 2.6 5.9 4.2 13.2 67.9 1.48

© ASCE 04014123-3 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 Table 2. Traffic Loading Calculation The moduli for the AC layers of the various sections in the
Fully One year 20-year summer time were very close to each other (≈2,000 MPa) in con-
Axle Trucks=day Trucks= loaded ESALs ESALs trast to the spring season. Fig. 7 shows that the stiffness of the as-
configuration (both lanes) year truck factor (million) (million) phalt mixtures decreased by around 80% between the spring and
Five-axle 600 0.19 × 106 11.33 2.2 44
summer seasons.
Four-axle 1,000 0.31 × 106 10.5 3.3 66 Analysis of variance (ANOVA) was conducted to test the stat-
Three-axle 200 0.06 × 106 4.5 0.27 5.4 istical differences among the moduli for each of the layers. The
Total 1,800 0.56 × 106 — 5.77 115.4 analysis was done using statistical significance level of 5% (using
α ¼ 5%). In this analysis, the p-value is the probability of acquir-
ing a test statistic at least as extreme as the one that was actually
observed, assuming that the null hypothesis is true. One often
summer season (August 2012); the average air temperature was rejects the null hypothesis when the p-value is less than the pre-
46°C while the average surface temperature was about 63°C. Fig. 7 determined significance level α indicating that the observed result
shows the moduli values for the various layers back-calculated would be highly unlikely under the null hypothesis. Table 3 shows
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

using Elmod6. the summary and results of ANOVA.

As would be expected, the results illustrate that the moduli of Based on the results in Table 3, the p-value of the HMA layer is
the AC layer and subbase layer are higher at the lower temperature. less than 0.05. So, the null hypothesis is rejected and 95% confident
The difference in moduli between the two seasons was the highest that the mean modulus of this layer is not statistically equal among
for the asphalt layer because of the temperature dependency of the the trial sections. For the subbase and subgrade layers, the p-value
asphalt behavior. The difference between the moduli of the two is more than 0.05 and this means that 95% confident that the mean
seasons for the subgrade results was relatively small. modulus for each of the layers is not statistically significant among
In general, it can be stated that the AC layer, with PMB, in the trial sections.
Sections 5 and 6 had the lowest temperature susceptibility, while
Section 2, with 60–70 Pen bitumen, had the highest temperature
susceptibility. This result showed that modified bitumen reduced Rutting Results from Automatic Road Analyzer Vehicle
the temperature’s impact on the stiffness of the asphalt pavement The wire model algorithm was employed for measuring rut
layers. depth using the automatic road analyzer vehicle. The wire model

Fig. 4. (a) Automatic road analyser vehicle collecting data from site; (b) FWD test conducted on the site

Fig. 5. Location and layout of cores

© ASCE 04014123-4 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 6. (a) AMPT for dynamic modulus and flow number tests; (b) set up for monotonic SCB test

Fig. 7. Comparison between moduli of each layer in low and high air temperatures

algorithm within the vehicle system connects the high points on the
pavements transverse profile and establishes the rut depth under
Table 3. ANOVA Summary and Results for Dynamic Modulus Results of
Each Layer
these points as shown in Fig. 8. The vehicle has 1,028 data points
across the 4-m transverse profile; the data was filtered down to 40
Section Count Sum Average Variance p-value data points across that profile. From that the wire model will con-
HMA layer nect the high points across a 3-m stretch centered along the profile;
1 10 93,686 9,368 5,672,269 6.71 × 10−8 the model will then take the highest distance from the wire to the
2 8 90,309 11,288 6,263,932 pavement for either 1.5-m side or record that has the left/right
3A 6 64,705 10,784 5,821,235 wheel path rut depth.
3B 8 72,691 9,086 3,529,195
The values of rut depth were taken as an average of 10 values
4 8 71,147 8,893 4,488,666
5 25 170,850 6,834 2,210,381
(1 value=m). Fig. 9 shows an example of rutting data collected
6 13 90,004 6,923 1,103,870 by the automatic road analyzer vehicle for trial Section 1. Table 4
Subbase layer shows the average of maximum rut depth for both directions was
1 10 9,795 979 102,693 0.079 obtained to compare the trial sections against rutting. In general,
2 8 8,458 1,057 147,928 it was observed that all sections of the road had little rutting.
3A 6 7,954 1,325 295,126 The effect of using different bitumen but same aggregate and
3B 8 9,461 1,182 64,248 mix design was one of the major targets of this study. Thus, trial
4 8 9,032 1,129 1,11,667 Sections 1, 2, and 6 were compared. Table 4 shows that Section 2
5 25 22,736 909 68,243
6 13 13,945 1,072 67,584
has the highest rutting depth value. This section consists of a sur-
Subgrade layer face course and a base of percentage refusal density design (PRD)
1 10 2,289 228 5,234 0.082 with 60–70 Pen bitumen and the aggregate used was Gabbro. This
2 8 1,803 225 9,335 concludes that Section 6 with PMB bitumen performed slightly
3A 6 1,611 268 12,541 better than Sections 1 and 2.
3B 8 1,642 205 529 Then, a comparison between pavement Sections 2 and 3A was
4 8 1,856 232 7,132 conducted to assess the influence of using the local aggregate, lime-
5 25 4,762 190 3,380 stone, in resisting rutting. Both sections consist of a surface course
6 13 2,266 174 2,186
and a base of PRD design with 60–70 Pen bitumen but with different

© ASCE 04014123-5 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 Laboratory Testing and Results

The dynamic modulus (jE! j), flow number (FN), and SCB tests
were used for evaluating the asphalt mixtures in terms of rutting
susceptibility, fatigue, and fracture resistance in laboratory and field
performance. The jE! j and FN are key tests within the simple per-
formance test (SPT) suite. They were developed under NCHRP
Project 9-19 and applied in the Superpave mix design procedure
(Witczak et al. 2002; Bonaquist et al. 2003; Zhu et al. 2011).
Fig. 8. Rut depth measurement using the wire model algorithm
Recently, Mohammad et al. (2011) conducted FN, dynamic modu-
lus, and SCB tests on HMA mixtures containing high-reclaimed
asphalt pavement (RAP) content with crumb rubber additives. The
results showed that the aforementioned tests are useful to character-
ize the mixtures and discriminate between them.
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

The SCB test, first proposed by Chong and Kuruppu (1984), has
been widely used by researchers to evaluate the fracture resistance
of asphalt mixtures by loading several specimens with different
notch depths monotonically until failure. Molenaar et al. (2002),
Elseifi et al. (2012), and Kim et al. (2012) described this test as
a rapid and simple test to be performed on easy to prepare spec-
imens. These advantages made this test preferable than other tests
such as bending beam test and favored by many researchers. From
the results of this test, the maximum tensile stress, resilient modu-
lus, fracture toughness (or stress intensity factor, K), and fracture
energy of HMA mixtures can be determined (Molenaar et al. 2002;
Ozer et al. 2009; Othman 2011). In a study by Arabani and
Ferdowsi (2009), the ability of the SCB test to characterize the
Fig. 9. Rutting depth values measured by the road analyzer vehicle for tensile strength and fracture toughness of mixtures was compared
trial Section 1 to a set of common static and dynamic tests. The results showed the
SCB test to be reliable and having good correlation with fracture
parameters obtained from other fracture tests.

aggregate type for the base. From the results shown in Table 4, it
can be concluded that the difference between both sections in rut Dynamic Modulus Test Results
depth was negligible. The jE! j test is a nondestructive test that is used to measure the
In addition, a comparison between trial Sections 2 and 4 with dynamic modulus and phase angle for each trial section. A total
Gabbro aggregate and between 3A and 3B with limestone were of 12 specimens collected from the base layer of each section were
conducted to assess the effect of mix design. It is shown that trial cored and prepared to standard size of 100 mm diameter and
Section 2 has higher rut depth value than Section 4. On the other 150 mm in height. A repeated load, with zero confinement, was
hand, the rut depth is marginally higher for Section 3B than applied at 4.4, 21.1, 37.8, and 54°C with loading frequencies of
Section 3A. The results show that the mix design didn’t affect 25, 10, 5, 1, 0.5, and 0.1 Hz [AASHTO designation: TP 79-11
significantly the rut depth results. (AASHTO 2011)]. In this study, 21.1°C was taken as the reference
Finally, rut depth values for trial Sections 5 and 6 were com- temperature. Master curves were developed from the test data using
pared to evaluate the effect of having different base layer. Both sec- a sigmoidal fitting function proposed by Pellinen et al. (2002). The
tions consist of a surface course of PRD with PMB bitumen but sigmoidal function can be described as follows:
different base layer. From the table, it was shown that Section 5
has a higher rut depth value than Section 6. These results showed α
LogðjE! jÞ ¼ δþ ð1Þ
that using Shell Thiopave in the base did not affect the performance 1 þ eβþγðlog ξÞ
of the pavement section significantly.
In general, it was observed in the field that the full-scale trial where δ = minimum modulus value; α, β, and γ = regression co-
road had little rutting. However, all trial sections performed very efficients for the fitting function; and ξ = reduced frequency. Table 5
similar against rutting after 2 years of service. and Fig. 10 shows the master curves and shift parameters.
Generally, the dynamic modulus values for all mixtures are sig-
nificantly high compared to results reported in other studies in lit-
Table 4. Rut Depth Results for All Trial Sections
erature (Goh et al. 2011; Kim et al. 2009; Zhu et al. 2011;
Average of maximum rut Bonaquist 2010). The range of jE! j values are similar to the values
Section depth for both lanes (mm) from a previous study by Masad et al. (2011) conducted on field
1 2.73 cores prepared using modified PG 76-22 bitumen and extracted
2 3.06 from a major freeway in Qatar. Fig. 10 shows that, at the high re-
3A 2.97 duced time and temperature end, Sections 6 and 3B exhibit highest
3B 3.10 jE! j values, which implies these two sections are the most resistant
4 2.55 to rutting. The range of jE! j values is an indication that mixture
5 2.97 designs used in Qatar can be too stiff, which raises a concern about
6 2.53
the durability and fatigue resistance of these mixtures.

© ASCE 04014123-6 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 Table 5. Master Curve and Shift Parameters From all preceding comparisons, it was noticed that the type
Section δ α β γ A b c of bitumen and aggregate mainly affected the dynamic modulus of
HMA mixes in the trial sections.
1 2.867 3.864 −1.634 0.490 0.0009 −0.166 3.062
2 2.708 3.933 −1.807 0.438 0.001 −0.168 3.086
3B 3.168 3.266 −1.663 0.414 0.0009 −0.164 3.028 FN Test Results
4 0.109 6.581 −2.324 0.363 0.0011 −0.177 3.208
5 3.613 3.006 −1.594 0.552 0.0008 −0.157 2.939 The specimens were subjected to FN test using the AMPT follow-
6 3.277 3.319 −1.534 0.422 0.0009 −0.164 3.044 ing the methods described in NCHRP 465 and 513 reports. The FN
test, which is also called as repeated load test in some studies, was
performed at a temperature of 54.4°C with zero confinement under
137 kPa deviatoric stress up to 10,000 cycles or until achieving
50,000 cumulative permanent microstrain. In this test, the higher
the FN value, the better is the trial section resistance to permanent
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

deformation. Fig. 12 illustrates the FN test results for the six HMA
mixtures considered in this study.
All FN values are high when compared to those published in
several previous studies conducted in the United States under
similar conditions (Mohammad et al. 2006; Wang et al. 2011;
Apeagyei et al. 2011; Kanitpong et al. 2012). This high FN value
is an indication that these mixtures are expected to have high re-
sistance to rutting. This is a result of the mixture designs used
which utilize fully crushed aggregate, relatively low bitumen con-
tent when compared with mixtures in the United States, and are
developed based on specifications that emphasize the high resis-
tance to permanent deformation given the high temperature ex-
Fig. 10. Dynamic modulus master curves for all trial sections pected in Qatar and the region.
To assess the sensitivity of using different bitumen on the resis-
tance of mixtures against rutting, FN results of Sections 1, 2, and 6
and then Sections 4 and 5 were compared. Fig. 12 presented that
In order to study the effect of using different bitumen types
Section 1, with 40–50 Pen bitumen, and Section 2, with 60–70 Pen
on mixture stiffness, master curves of trial Sections 1, 2, and 6
bitumen, had very close FN values while Section 6, with PMB, had
and then Sections 4 and 5 were compared as shown in
slightly lower FN value. On the other hand, Fig. 12 shows that
Figs. 11(a and b), respectively. Both figures show that all compared
Section 5 with Thiopave bitumen had a lower FN value than the
sections had high dynamic modulus values at low reduced time control Section 4 and this could be attributed to aging characteri-
(high frequency) and temperature. In contrast, the stiffness of zation or the volumetric of the mix. In general, all of them per-
Section 1, with 40–50 Pen bitumen, in Fig. 11(a) became lower formed well against rutting but the results proved that the
than the other Sections 2 and 6 at high-reduced time and temper- bitumen type and grade are important factors in rut resistance of
ature. This inferred that modified bitumen make AC layers stiffer at HMA mixes. The FN was found to increase with increasing the
high temperature than those of unmodified bitumen, which im- bitumen grade while the use of modified bitumen did not improve
proves the permanent deformation resistance. In addition, PMB the performance of the mixture (Bonaquist 2010).
flattened the master curve of Section 6 and reduced the effect of Then, FN results of Section 2, with Gabbro, and Section 3B,
temperature and frequency on the stiffness. with limestone, were compared to evaluate the influence of using
Fig. 11(b) showed no difference at low reduced time (high fre- different aggregate types. Section 2 had significantly higher FN
quency) and temperature between using 60–70 Pen and Thiopave cycles than that of Section 3B. This result demonstrated that using
bitumen when the mix design is QCS and the aggregate used is the local aggregate (limestone) decreased the ability of the mixtures
Gabbro. The difference is considerable at high-reduced time and to resist rutting. This can be attributed to the relatively low angu-
temperature where Section 5 with Thiopave is stiffer than Section larity of limestone, which is one of the main factors affecting the
4 with 60–70 Pen bitumen. reduction of the FN value and shows low rut resistance.
After that, jE! j master curves of Section 2, with Gabbro, and Also, the effect of using different mix designs was assessed
Section 3B, with limestone, were compared to assess the effect of by comparing FN results of Sections 2 (Marshall/PRD) and 4
using different aggregate types as shown in Fig. 11(c). Section 3B (Marshall/QCS). Both sections performed the same and showed
had higher dynamic modulus than that of Section 2 at high- similar resistance against rutting. This observation confirms that the
reduced times and temperatures. This result demonstrated that us- mix design is not a main factor in rut resistance of HMA mixes.
ing the local aggregate (limestone) increased the stiffness of the Conferring to the preceding comparisons, it was clearly ob-
mixture in this study, which is affecting positively on its perfor- served that the type of the bitumen and aggregate angularity mainly
mance against rutting. affects resistance of permanent deformation for asphalt mixtures.
In Fig. 11(d), a comparison between master curves of Sections 2
(Marshall/PRD) and 4 (Marshall/QCS) is shown. The effect of
using different mix designs was evaluated here. Both sections Monotonic SCB Test Results
performed the same at low-to-intermediate reduced times while The configuration of the SCB test consists of three-point monoton-
Section 2 has a slightly higher dynamic modulus value at very high ically increasing compressive loading that induces tension in the
reduced time or temperature. It can be stated that Marshall/PRD bottom part of a SCB specimen. The set up of the test consists of
and Marshall/QCS design mixes did not have much effect on two supporting rollers at the bottom edge and a loading roller at
the stiffness of both sections. the midpoint of the semicircular arch (Liu 2011; Mull et al. 2002).

© ASCE 04014123-7 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 11. Dynamic modulus master curves for (a) Sections 1, 2, and 6; (b) Sections 4 and 5; (c) Sections 2 and 3B; (d) Sections 2 and 4

The spacing between the two supports is 0.8 times the diameter as shows each slice was symmetrically cut into two semicircular
shown in Fig. 13(a). samples with a notch of 12.5 mm (0.5 in.).
For the purpose of evaluating the fracture resistance of the trial Two semicircular specimens from each core were then subjected
sections, one core from each section was tested in the laboratory of to the monotonic SCB test and the applied load was strain-controlled
the University of Liverpool. These six cores consisted of a surface at 5 mm=min and at temperature of 10°C. The other half of semi-
course (≈70 mm) and a base course (≈250 mm). The base course circular specimens was tested at 10 mm=min at the same temper-
of each core was cut symmetrically from the middle of the speci- ature to investigate the effect of loading rate on the performance of
men into two circular slices, each 50 mm in height. Fig. 13(b) the mixtures. The applied load, the horizontal and vertical displace-
ment in addition to the measured stress and strain were monitored
and recorded during the test. This test data was used to calculate the
fracture toughness, fracture energy, and maximum tensile stress at
the bottom of specimens of each section using the equations shown
in Table 6.
Fig. 14 illustrates the stress-strain curve for each trial section
obtained from the tests at 5 mm=min. It was observed from the test
that materials were brittle and failed abruptly with little deforma-
tion or plastic flow. Also, Fig. 14 shows no well-defined peaks or
slow fracture tail, common with softer materials and slower test
speeds. Fig. 15 illustrates the summary of the results of both SCB
tests for each trial section in this study.
Generally, the fracture toughness, fracture energy and maximum
tensile stress values of all mixtures are high and reflected good per-
formance when compared with published values for other mixtures.
However, Section 1, Marshall/PRD with 40–50 Pen bitumen and
Gabbro, had the highest average fracture parameters among the trial
sections. This high resistance to propagation to fracture is primarily
Fig. 12. Flow number test results for each trial section
caused by the stiffness of these mixtures.

© ASCE 04014123-8 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 13. (a) Typical set up for SCB test; (b) notched SCB test specimen

It is documented that mixtures with high stiffness have good of fracture toughness, fracture energy, and tensile stress as shown
resistance to fatigue cracking and fracture when they are used in in Fig. 15 as well. The results of Section 3B in the test of lower
thick sections, while they have low resistance to fatigue when used loading rate (5 mm=min) were ignored because the specimen was
in thin pavement sections (Masad et al. 2008; Tayebali et al. 1994). highly segregated. Generally, the result demonstrated that both
These mixtures are all used in thick sections in this experiment so sections had very fracture resistance but Section 2, with Gabbro,
they are expected to have very good resistance to bottom-up fatigue outperformed with higher fracture energy than Section 3B, with
cracking because the stress at the bottom of the layer would be very limestone, as shown in Fig. 15(b).
small. However, given their high stiffness, these mixtures would be Then, the influence of mix design on the fracture resistance was
expected to have problems with fatigue cracking if used in thin sec- evaluated by comparing the use of Marshall/PRD in Section 2 and
tions or if the bond between the surface course and base course is Marshall/QCS in the control Section 4. From results shown in
compromised. In addition, the very high stiffness of mixtures with Fig. 15, it can be stated that using the conventional mix design
low asphalt content might contribute to problems in durability and Marshall/QCS had a little bit more toughness, energy, and tensile
top-down cracking. stress than PRD mix design. The small difference between both
To evaluate the significance of bitumen on the resistance to frac- results emphasized that mix design is not a main factor affecting
ture, Sections 1, 2, and 6 were compared. Fracture toughness and the fracture resistance compared to bitumen or aggregate type.
fracture energy of field samples are affected by bitumen type as
concluded by Li et al. (2006). In Fig. 15(a), the sections were
compared based on the fracture toughness. The results showed that Mechanistic-Empirical Analysis for Performance
Section 1 with unmodified 40–50 Pen bitumen had the highest frac- Evaluation
ture toughness in both loading rates. This is the case for fracture
energy and maximum tensile stress. The researchers at Texas A&M University at Qatar developed
After that, Sections 4 and 5 with QCS mix design and Gabbro a pavement analysis tool for the analysis of asphalt pavement
were compared to evaluate the significance of using Shell Thiopave performance. This analysis tool uses the same fatigue and rutting
bitumen on the resistance of fracture. In Figs. 15(a and c), the re- models in the Mechanistic-Empirical Pavement Design Guide
sults showed that both sections had very similar fracture toughness (M-E PDG) software, and it is programmed in Excel.
and maximum tensile stress values at both loading rates. However, Fig. 16 shows the inputs of the developed pavement analysis
the case for fracture energy is different as shown in Fig. 15(b). Sec- tool were the structure and properties of the pavement layers;
tion 5, with Thiopave bitumen, dissipated more energy to propagate the dynamic modulus values obtained from jE! j laboratory tests,
the crack than Section 4 with unmodified 60–70 Pen bitumen. Gen-
erally, it can be stated that Thiopave bitumen improved the perfor-
mance of the mixture in fracture resistance.
The performance of Sections 2 and 3B were assessed to acquire
insight into the effect of aggregate type by comparing their results

Table 6. Equations Used to Calculate Fracture Toughness, Fracture

Energy, and Tensile Stress
Parameter Equation used Reference
! P " pffiffiffiffiffiffi
Fracture toughness (K) K ¼ 2rt Y 1 πa Lim et al. (1993)
Fracture energy (Gf ) Gf ¼ AWLig RILEM TC
50-FMC (1985)
Maximum tensile σ ¼ 3.564 PD:t
ult
Shu et al. (2010)
stress at the bottom
of the specimen (σ)
Note: ΔP = applied load (N); a = notch depth (mm); ALig = ligament area
(mm2 ); D = specimen diameter (mm); Pult = ultimate applied load (N); r = Fig. 14. Stress-strain curves obtained from SCB test for all mixtures at
specimen radius (mm); t = specimen thickness (mm); W = work of fracture
5 mm=min
(kN · mm); Y 1 = normalized stress intensity factor in mode I.

© ASCE 04014123-9 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 15. Comparison between all trial sections for both loading rate: (a) fracture toughness; (b) fracture energy; (c) maximum tensile stress

the air temperature, the responses of the pavement layers of the trial εpðACÞi ¼ εvðACÞi ðβ 1 k1 10−β 2 3.35412 T β3 1.5606 N β 4 0.4491 Þ ð5Þ
sections from a linear elastic software (WinJULEA) obtained from
FWD results, and finally the traffic loading data after 3 years of
service. Then, it predicts the fatigue damage in terms of alligator where εvðACÞi = vertical compressive strain at middepth of AC
(bottom-up) cracking and it can also measure the expected rutting sublayer i obtained from WinJULEA analysis; β 1;2;3;4 = regional
depth in each layer and in every season separately.
In order to predict the alligator cracking and rutting depth,
the following models from M-E PDG software were used:
100
Fatigue ðAlligatorÞ Cracking ¼ ð2Þ
1 þ ec2 ð−2þlog FDÞ
0

where FD = the fatigue damage that is calculated from actual num-

ber of traffic loads within a specific time divided by the number of
allowable repetitions to failure for bottom-up cracking (N f ); where
N f = determined using the tensile strain at the bottom of AC thick
layer acquired from WinJULEA analysis with its dynamic modulus
jE! j value obtained from the laboratory test; c20 = a calibration fac-
tor, which depends on the thickness of the AC layer and is calcu-
lated as follows:
c20 ¼ −2.40874 − 39.748 ð1 þ hAC Þ−2.856 ð3Þ
Then, permanent deformation (rutting depth) is calculated as
follows:
X
m

Permanent Deformation ðRuttingÞ ¼ hi εpðACÞi ð4Þ

1

where hi = thickness of AC sublayer i; and εpðACÞi = vertical plastic

Fig. 16. Developed pavement analysis tool flow chart
strain of AC sublayer i which is calculated by Eq. (5).

© ASCE 04014123-10 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 Fig. 17. Pavement analysis tool results for trial Sections 1, 2, and 6: (a) alligator cracking; (b) rutting depth
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

calibration factors which are assumed to be 0.7 based on pre- was used. As the previous comparison, the results of alligator
liminary work conducted by researchers in Texas A&M at Qatar; cracking in Fig. 18(a) for these two sections were significantly
k1 = coefficient calculated from the thickness of the AC layer; T = low as observed in the field. In general, the results showed that
pavement surface temperature that is calculated from the recorded using Shell Thiopave bitumen didn’t affect the alligator cracking
air temperatures for three seasons in Qatar using SHRP models percentage a lot compared to unmodified bitumen especially when
available in Mohseni et al. (2005); and N = actual number of traffic the temperature is high. In Fig. 18(b), again, the comparison speci-
loads (ESALs). fied that total rutting depths attained from the road analyzer vehicle
The performance models implemented in this tool were used to is greatly lower than those calculated using the analysis tool for
conduct comparative analysis of the performance of the different traffic loading of 2 years only. From the figures, the total alligator
pavement sections. These models need to be calibrated in the future cracking and the total rutting depth are higher for Section 5 with
based on comprehensive field measurements. Thiopave than those for trial Section 4.

Effect of Bitumen
Effect of Aggregate Type
The trial Sections 1, 2, and 6 were compared to check the influence
of bitumen type on the performance of the mixes against alligator Comparing Sections 2 and 3B assessed the impact of using lime-
cracking and rutting in different seasons. The results of alligator stone as an alternative of Gabbro. The performance of the trial sec-
cracking in Fig. 17(a) for these three trial sections were signifi- tions against fatigue cracking (alligator) and rutting in different
cantly low and this is complying with the field observations. This seasons was evaluated. Fig. 19(a) showed that Section 2, with
was expected because of the use of perpetual pavement concept in Gabbro, and Section 3B, with limestone, had very low cracking
the trial sections (Sadek et al. 2014). In Fig. 17(b), the comparison percentage (less than 0.5%). The results of the road analyzer
indicated that total rutting depths obtained from the automatic road and the pavement analysis tool were not matching; however, it
analyzer vehicle is lower than those calculated using the analysis showed that there was no significant difference in the rutting resis-
tool for traffic loading of 2 years only. The rutting model in the tance if limestone was used as alternative of Gabbro as shown in
analysis tool needs to be further calibrated to better match field Fig. 19(b).
results in the future. In general, the type/grade of bitumen did
not affect the performance significantly after 2 years of loading.
Effect of Mix Design
Generally, the performance of Section 2, with 60–70 Pen bitumen,
had the lowest cracking percentage and rutting depth. The effect of using altered mix designs was studied by comparing
Then, Sections 4 and 5 were also compared to evaluate the im- the performance of Sections 2 and 4 in resisting alligator cracking
pact of bitumen type on the performance of the mixes against in addition to rutting. Fig. 20(a) concluded that the alligator crack-
fatigue and rutting in different seasons when the QCS mix design ing values for both sections are very small and negligible but

Fig. 18. Pavement analysis tool results for trial Sections 4 and 5: (a) alligator cracking; (b) rutting depth

© ASCE 04014123-11 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 Fig. 19. Pavement analysis tool results for trial Sections 2 and 3B: (a) alligator cracking; (b) rutting depth
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 20. Pavement analysis tool results for trial Sections 2 and 4: (a) alligator cracking; (b) rutting depth

Fig. 21. Pavement analysis tool predicted results for 20 years: (a) alligator cracking; (b) rutting depth

Section 2, PRD mix design, showed better resistance than QCS mix
design in Section 4. Similar to the previous subsections, Fig. 20(b)
showed that the rutting results of the road analyzer did
not match with the results of the analysis tool but there was no
significant difference in performance between PRD and QCS
mix designs.
In addition, Fig. 21 shows that the developed pavement analysis
tool can be used to predict the performance after 10 or 20 years.
According to all preceding comparisons, the results of alligator
cracking are complying with field observations and thus could be
predicted by the developed pavement analysis tool while rutting
model needs further calibration to match the field results. Accord-
ingly and based on the fact that Qatar has different conditions than
the United States, the value of the first regional calibration factor,
β 1 , was reduced to be 0.3 and Fig. 22 shows that this can give closer
Fig. 22. Rutting depth of all trial sections after reducing β 1
results to the field results.

© ASCE 04014123-12 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 Conclusions Al-Abdul Wahhab, H. I., and Balghunaim, F. A. (1994). “Asphalt pavement
temperature related to arid Saudi environment.” J. Mater. Civ. Eng.,
With the aim of assessing the long-term performance of the trial 10.1061/(ASCE)0899-1561(1994)6:1(1), 1–14.
sections, some field and laboratory tests were conducted. The Apeagyei, A. K., Diefenderfer, B. K., and Diefenderfer, S. D. (2011). “Rut-
following are the major conclusions based on the results of ting resistance of asphalt concrete mixtures that contain recycled asphalt
this study: pavement.” Transportation Research Record 2208, Transportation
• Field and laboratory tests indicated that all trial sections con- Research Board, Washington, DC, 9–16.
tinue to perform very well 3 years after construction; Arabani, M., and Ferdowsi, B. (2009). “Evaluating the semi-circular bend-
ing test for HMA mixtures.” IJE. Trans. A, 22(1), 47–58.
• According to the results of all tests, only the bitumen and ag-
Bonaquist, R. (2010). “Wisconsin mixture characterization using the
gregate type significantly affected the stiffness and the trial
asphalt mixture performance tester (AMPT) on historical aggregate
sections’ resistance to rutting and fracture;
structures.” WHRP 09-03, Wisconsin Highway Research Program.
• All mixes were generally very stiff and had low rutting depths Bonaquist, R. F., Christensen, D. W., and Stump, W. (2003). “Simple per-
but Section 6, with PMB bitumen and Gabbro, was the stiffest formance tester for superpave mix design: First-article development and
under condition of high temperature. In contrast, Section 1, with
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

evaluation.” NCHRP Rep. No. 513, National Cooperative Highway

40–50 Pen bitumen and Gabbro in the base layer, was the Research Program, Washington, DC.
least stiff; BS EN 12697/32:2003 (2003). “Bituminous mixtures. Test methods for
• Increase in temperature in Qatar reduced the stiffness of AC hot mix asphalt. Laboratory compaction of bituminous mixtures by
layers by about 80%. In addition, at high temperature, all sec- vibratory compactor.” BSI, London, U.K.
tions had almost the same modulus value of 2,000 MPa mea- Bubshait, A. A. (2001). “Quality of pavement construction in Saudi
sured using FWD; Arabia.” Pract. Period. Struct. Des. Constr., 10.1061/(ASCE)1084-
• Sections 5 and 6, with PMB, had the lowest in temperature 0680(2001)6:3(129), 129–136.
susceptibility. However, Section 2, with unmodified 60–70 Pen Chong, K. P., and Kuruppu, M. D. (1984). “New specimen for fracture
bitumen, was the highest in temperature susceptibility; and toughness determination for rock and other materials.” Int. J. Fract.,
• The developed pavement analysis tool with mechanistic- 26(2), R59–R62.
empirical functions was very useful to predict the performance Dynatest Elmod6 Version 6.1.44. [Computer science]. Dynatest
of the trial sections against alligator cracking and permanent International.
Elseifi, M. A., Mohammad, L. N., Ying, H., and Cooper, S., III (2012).
deformation. The functions will be calibrated against more
“Modeling and evaluation of the cracking resistance of asphalt mixtures
comprehensive field data.
using the semi-circular bending test at intermediate temperatures.” Road
It is obvious from this study that rutting, fracture, and alligator
Mater. Pavement Des., 13(S1), 124–139.
cracking are not major distresses for the perpetual pavements in Fatani, M, Al-Abdul Wahhab, H. I., Balgunaim, F. A., Bubshait, A. A., and
Qatar. More laboratory tests will be conducted in the near future Al-Dubaiab, I. (1992). “National research project: Evaluation of perma-
to evaluate the fatigue resistance and durability of these sections. nent deformation of asphalt concrete pavement in Saudi Arabia.” Final
Results will help make recommendations for updating the materials Rep., King Abdulaziz City for Science and Technology, Riyadh, Saudi
and design of asphalt pavements, and determining more of the input Arabia.
parameters for mechanistic-empirical analysis of asphalt pavements Ferne, B. (2006). “Long-life pavements—A European study by ELLPAG.”
in the State of Qatar. Int. J. Pavement Eng., 7(2), 91–100.
Goh, S. W., You, Z., Williams, C., and Li, X. (2011). “Preliminary dynamic
modulus criteria of HMA for field rutting of asphalt pavements:
Michigan’s experience.” J. Transp. Eng., 10.1061/(ASCE)TE.1943-
Acknowledgments
5436.0000191, 37–45.
The authors would like to acknowledge the financial support pro- Kanitpong, K., Charoentham, N., and Likitlersuang, S. (2012). “Investiga-
tion on the effects of gradation and aggregate type to moisture damage
vided by Qatar National Research Fund (QNRF) through the
of warm mix asphalt modified with Sasobit.” Int. J. Pavement Eng.,
National Priority Research Program (NPRP) Project 08-310-2-110. 13(5), 451–458.
This project was awarded jointly to Texas A&M at Qatar University Kim, Y., Lee, H., and Heitzman, M. (2009). “Dynamic modulus and
and Qatar University. The researchers at Texas A&M at Qatar also repeated load tests of cold in-place recycling mixtures using foamed
acknowledge the support through NPRP 4-789-2-293 Project that asphalt.” J. Mater. Civ. Eng., 10.1061/(ASCE)0899-1561(2009)21:
was used to develop the pavement analysis tool presented in this 6(279), 279–285.
paper. In addition, the authors would like to extend their gratitude Kim, Y.-R., Zhang, J., and Ban, H. (2012). “Moisture damage characteri-
to the Fugro Peninsular Company for its support. Last but by no zation of warm-mix asphalt mixtures based on laboratory-field evalu-
ation.” Constr. Build. Mater., 31, 204–211.
means least, the authors are indebted to Eng. Mohammed Sadeq for
Li, X., Zofka, A., Li, X., Marasteanu, M., and Clyne, T. R. (2006). “Inves-
his efforts to develop the pavement analysis tool. The technical in- tigation of the low-temperature fracture properties of three MnROAD
put of Dr. Dallas Little during this study is highly appreciated. asphalt mixtures.” Final Rep., MN/RC-2006-15, Minnesota Local Road
Research Board, St. Paul, MN.
Lim, I. L., Johnson, I. W., and Choi, S. K. (1993). “Stress intensity factors
References for semi-circular specimens under three-point bending.” Eng. Fract.
Mech., 44(3), 363–382.
AASHTO. (2011). “Standard method of test for determining the dy- Liu, J. (2011). “Fatigue life evaluation of asphalt rubber mixtures using
namic modulus and flow number for hot mix asphalt (HMA) using the semi-circular bending test.” Adv. Mater. Res., 255–260, 3444–3449.
asphalt mixture performance tester (AMPT).” TP 79-11, Washington, Masad, E., Castelo Branco, V., Little, D., and Lytton, R. (2008). “A unified
DC. method for the analysis of controlled-strain and controlled-stress fatigue
Al-Abdul Wahhab, H. I., Asi, I. M., and Ramadhan, R. H. (2001). “Mod- testing.” Int. J. Pavement Eng., 9(4), 233–246.
eling resilient modulus and temperature correction for Saudi roads.” Masad, E., Kassem, E., and Little, D. (2011). “Characterization of asphalt
J. Mater. Civ. Eng., 10.1061/(ASCE)0899-1561(2001)13:4(298), pavement materials in the State of Qatar: A case study.” Int. J. Road
298–305. Mater. Pavement Des., 12(4), 739–765.

© ASCE 04014123-13 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.

 Mechanistic-Empirical Pavement Design Guide (M-E PDG) Version Sadek, H., Masad, E., Sirin, O., Al-Khalid, H., and Little, D. (2012). “The
6.1.44. [Computer software]. Applied Research Associates— implementation of mechanistic-empirical pavement design method
Transportation, Arizona State Univ., Phoenix, AZ. to evaluate asphalt pavement design in Qatar.” 5th Eurasphalt &
Merrill, D., Dommelen, A. V., and Gáspár, L. (2006). “A review of practical Eurobitume Congress, EAPA, Brussels, Belgium.
experience throughout Europe on deterioration in fully-flexible and Sadek, H., Masad, E., Sirin, O., Al-Khalid, H., Sadeq, M., and Little, D.
semi-rigid long-life pavements.” Int. J. Pavement Eng., 7(2), 101–109. (2014). “Implementation of mechanistic-empirical pavement analysis in
Mohammad, L. N., Cooper, S. B., Jr., and Elseifi, M. A. (2011). “Charac- the State of Qatar.” Int. J. Pavement Eng., 15(6), 495–511.
terization of HMA mixtures containing high reclaimed asphalt pave- Shu, X., Huang, B., and Vukosavljevic, D. (2010). “Evaluation of cracking
ment content with crumb rubber additives.” J. Mater. Civ. Eng., resistance of recycled asphalt mixture using semi-circular bending test.”
10.1061/(ASCE)MT.1943-5533.0000359, 1560–1568. Paving Materials and Pavement Analysis—Proc., 2010 GeoShanghai
Mohammad, L. N., Wu, Z., Obulareddy, S., Cooper, S., and Abadie, C. Int. Conf., China, 58–65.
(2006). “Permanent deformation analysis of hot-mix asphalt mixtures Tayebali, A., Deacon, J. A., Coplantz, J., Harvey, J. T., and Monismith, C. L.
with simple performance tests and 2002 mechanistic-emperical pave- (1994). “Mixture and mode-of-loading effects on fatigue response of
ment design software.” Transportation Research Record 1970, Trans- asphalt-aggregate mixtures.” Asphalt Paving Technology: Proc., Asso-
portation Research Board, Washington, DC, 133–142. ciation of Asphalt Paving Technologists Technical Sessions, Vol. 63,
Downloaded from ascelibrary.org by Qatar University Library on 09/25/15. Copyright ASCE. For personal use only; all rights reserved.

Mohseni, A., Carpenter, S., and D’Angelo, J. (2005). “Development of 118–151.

SUPERPAVE high-temperature performance grade (PG) based on rut- Timm, D. H., and Newcomb, D. E. (2006). “Perpetual pavement design for
ting damage.” J. Assoc. Asphalt Paving Technol., 74, 197–254. flexible pavements in the US.” Int. J. Pavement Eng., 7(2), 111–119.
Molenaar, A. A. A., Scarpas, A., Liu, X., and Erkens, S. M. J. G. (2002).
TRL. (2002). “A guide to the design of hot mix asphalt in tropical and sub-
“Semi-circular bending test: Simple but useful?” J. Assoc. Asphalt
tropical countries.” Overseas Road Note 19, Overseas Centre, Transport
Paving Technol., 71, 794–815.
Research Laboratory, Crowthorne, Berkshire, U.K.
Mull, M. A., Stuart, K., and Yehia, A. (2002). “Fracture resistance char-
TRL Report C. (2010). “Phase C: Performance of existing pavement in
acterization of chemically modified crumb rubber asphalt pavement.”
Qatar.” TRL Client Project Rep. 282, Transport Research Laboratory,
J. Mater. Sci., 37(3), 557–566.
Berkshire, U.K.
Othman, A. M. (2011). “Evaluation of hydrated lime effect on the perfor-
mance of rubber-modified HMA mixtures.” J. Elastomers Plast., 43(3), TRL Report D. (2010). “Phase D: Design of site trials.” TRL Client Project
221–237. Rep. 282, Transport Research Laboratory, Berkshire, U.K.
Ozer, H., Aurangzeb, Q., Roberts, G. L., Al-Qadi, I. L., Carpenter, S. H., Wang, H., Al-Qadi, I. L., Faheem, A. F., Bahia, H. U., Yang, S., and
and Trepanier, J. (2009). “Evaluation of RAP impact on hot-mix asphalt Reinke, G. H. (2011). “Effect of mineral filler characteristics on asphalt
design and performance.” J. Assoc. Asphalt Paving Technol., 78, mastic and mixture rutting potential.” Transportation Research Record
317–351. 2208, Transportation Research Board, Washington, DC, 33–39.
Pellinen, T. K., Witczak, M. W., and Bonaquist, R. F. (2002). “Asphalt mix WinJULEA. [Computer software]. Engineering Research & Development
master curve construction using sigmoidal fitting function with non- Center Software.
liner least squares optimization.” 15th ASCE Engineering Mechanics Witczak, M. W., Kaloush, K., Pellinen, T., El-Basyouny, M., and Von
Conf., Columbia Univ., New York. Quintus, H. (2002). “Simple performance test for superpave mix
Renteria, R. A., and Hunt, E. (2008). “Super pavements.” Roads Bridges, design.” NCHRP Rep. No. 465, National Cooperative Highway Re-
46(6), 47–51. search Program, Washington, DC.
RILEM Technical Committee 50-FMC. (1985). “Determination of fracture Zhu, H., Sun, L., Yang, J., Chen, Z., and Gu, W. (2011). “Developing
energy of mortar and concrete by means of three-point bend tests on master curves and predicting dynamic modulus of polymer-modified
notched beams.” Materials and Structures, Vol. 106, RILEM Publica- asphalt mixtures.” J. Mater. Civ. Eng., 10.1061/(ASCE)MT.1943-
tions SARL, 285–290. 5533.0000145, 131–137.

© ASCE 04014123-14 J. Perform. Constr. Facil.

J. Perform. Constr. Facil. 2015.29.