INFLUENCE OF TEMPERATURE ON THE FATIGUE LIFE OF

FLEXIBLE PAVEMENTS
Hugo M.R.D. Silva* and Joel R.M. Oliveira
Department of Civil Engineering – University of Minho
Guimarães, Portugal
hugo@civil.uminho.pt
joliveira@civil.uminho.pt
* Corresponding author

Luís G. Picado-Santos
Department of Civil Engineering – University of Coimbra
Coimbra, Portugal
picsan@dec.uc.pt

Abstract

Flexible road pavements are the solution usually implemented in Portugal for the
construction of highways. The properties of the bituminous mixtures used in this type of
pavement are greatly influenced by environmental temperature, mainly due to its effect
on the bitumen viscosity. The main aim of this work is to evaluate the performance of
bituminous mixtures in pavements, at different temperatures, in relation to two of their
fundamental properties, namely stiffness modulus and fatigue cracking resistance.
These properties control the behaviour of recently developed thin surface layers, since
they do not show significant permanent deformation. Some bituminous mixtures, such
as Asphalt Concrete (AC 0/14), Stone Mastic Asphalt (SMA 0/12), Asphalt Rubber (AR
0/10) and Porous Asphalt (PA 0/14), used in surface layers and overlays, were selected
for experimental characterization, The stiffness moduli of those mixtures were
determined at the temperatures of 5, 15 and 25ºC. The AC 0/14 was selected for further
testing in order to assess the fatigue life at the temperatures of 5, 15 and 25ºC. The
influence of temperature on the stiffness modulus and on the fatigue life of the
bituminous mixtures is empirically considered in Shell fatigue life equation normally
used in the pavement design. A comparative analysis between the pavement design
results achieved by using Shell fatigue life and those obtained from laboratory tests
performed on the AC 0/14 at different temperatures was carried out. The main
conclusions of this work are related to the different evolution of stiffness moduli and
fatigue life regarding different temperatures, temperature influence on the total design
thickness of bituminous pavement layers and whether the Shell fatigue life equation is
the most adequate for Portuguese materials and in-service conditions.

INTRODUCTION

The stiffness modulus and the fatigue resistance are two of the most important properties
related to the behaviour of the bituminous materials as they have a large influence in the
flexible pavement design. These properties of bituminous mixtures, primarily those applied
on surface layers and overlays, are greatly influenced by environmental temperature.
Lundström et al. (2003) have already carried out an investigation on stiffness and fatigue
properties of asphalt mixtures at different temperatures. They stated that the classical fatigue
evaluation method showed several inadequacies to assess the influence of temperature. The

1
high scattering among fatigue data and the large amount of tests necessary to cover a given
temperature range leads to a need for several samples and repetitions in order to obtain an
adequate characterisation.

Pavement temperature is an uncontrollable factor. It presents a variability that the design

methods cannot transmit in a real way. Hence it is necessary to use a great safety margin in
order to overcome that variability (Picado Santos, 1994). Thus, the main objective of this
work is to evaluate the influence of the temperature in the performance of bituminous
mixtures and, mainly, in the design of flexible road pavements. Firstly, some mixtures were
produced in the laboratory in order to assess their stiffness moduli at different temperatures.
Next, one mixture was selected so as to determine the fatigue cracking laws at different
temperatures. After that, the laboratory results were used to design flexible pavements at
several design temperatures. Finally, a comparative analysis between the design results of the
Shell fatigue cracking law (Shell, 1978) and those obtained from the laboratory tests at
different temperatures was carried out in order to evaluate the consequences of the lack of
temperature information in the Shell fatigue cracking law (Shell, 1978), especially for
mixtures commonly used in Portuguese pavement surface layers.

LABORATORY STUDY OF SEVERAL BITUMINOUS MIXTURES USED IN

SURFACE LAYERS AT DIFFERENT TEMPERATURES

The main objective of this initial laboratory study was to assess the evolution of the stiffness
moduli of different bituminous mixtures used in surface layers and overlays with temperature.
The following bituminous mixtures were selected for experimental characterisation:
ƒ Asphalt Concrete 0/14 (AC 0/14), being the mixture typically used in road surface layers;
ƒ Stone Mastic Asphalt 0/12 (SMA 0/12), using acrylic fibres to stabilize the mixture;
ƒ Porous Asphalt 0/14 (PA 0/14) stabilized with fibres, in order to study porous mixtures;
ƒ Asphalt Rubber 0/10 (AR 0/10), using 19% of crumb rubber as binder additive.

The stiffness moduli of those mixtures were determined by using the four-point bending beam
test (EN 12697-26) at the temperatures of 5, 15 and 25 ºC. By comparing the stiffness moduli
of all the studied mixtures it was possible to select the bituminous mixture with higher
thermal susceptibility, which was further evaluated, as presented in the following parts of this
work.

Composition of the Bituminous Mixtures Evaluated in the Laboratory

The mixtures studied in this work comprised granitic aggregates and limestone filler. The Los
Angeles fragmentation index was lower than 15% and the water absorption presented values
between 0.4 and 0.6%. Two types of binders were used. A 35/50 pen bitumen, with
penetration of 34.67 dmm and a softening point of 54.2 ºC was used in the AC 0/14, SMA
0/12 and PA 0/14 mixtures. In order to make the AR 0/10 mixture, it was necessary to
produce a rubber modified binder (RMB) with 35/50 bitumen and 19% of crumb rubber. The
RMB had a penetration of 21.5 dmm, a softening point of 73.5 ºC, a dynamic viscosity at
175 ºC of 1788 cSt, a resilience of 58% and an elastic recovery of 47%. Thus, RMB is more
viscous, especially at higher temperatures, and it is less sensitive to temperature changes than
the unmodified 35/50 pen binder.

The bitumen and the aggregate characteristics are in conformity with the limits of the
Portuguese specifications E80 and APORBET (1998) for the studied mixtures. Two additives

2
were used to produce the bituminous mixtures, namely acrylic fibres “Asfhal+ L60 1.9 dtex –
6mm” supplied by FISIPE and tyre crumb rubber supplied by RECIPAV.

Figure 1 presents the grading curves of the four studied mixtures, obtained through the
combination of granitic aggregates and limestone filler. The design grading curves of AC 0/14
and PA 0/14 mixtures were set in order to fulfill the limits of the present Portuguese
specifications (APORBET, 1998), while the design grading curves of SMA 0/12 and AR 0/10
were set in order to fulfill the limits presented by the producers of the mixtures VIATOP
(2006) and RECIPAV (2001), respectively.

100
AC 0/14
90
SMA 0/12
80
PA 0/14
Percentage passing (%)

70
AR 0/10
60
50
40
30
20
10
0
0.0 0.1 1.0 10.0 100.0
Sieve size (mm)

Figure 1 – Grading curves of the studied mixtures

The AR 0/10 is an open graded mixture using a modified binder with 19% of crumb rubber
additive. Acrylic fibres were also used as additives in order to stabilize the SMA 0/12 (0.3%)
and PA 0/14 (0.2%) mixtures.

Before producing the studied mixtures, a mix design study was carried out using the Marshall
mix design method in order to access their optimum binder content. Thus, it was possible to
define the following optimum binder contents for the studied mixtures: AC 0/14 (5.5%), SMA
0/12 (6.8%), PA 0/14 (5.0%) and AR 0/10 (10.5%), by mass of mixture.

Production and Volumetric Characterisation of the Bituminous Mixtures

After determining the optimum composition of the four mixtures (grading curves of the
aggregates and binder content), the required mixing batches were produced as following
described. First the aggregates and the filler were weighed and placed in an oven at a
temperature of 180 ºC. Meanwhile, the bitumen was heated at a temperature of 160 ºC (or
175ºC for the rubber modified bitumen). After the aggregates and the bitumen had reached the
necessary temperature, the mixture was produced by introducing the aggregates in the mixer,
followed by the fibres and finally the bitumen. A vibratory steel roller compactor (EN 12697-
33) was used to compact each mixture into a rectangular steel mould with dimensions
50×75×8 cm3 at a temperature of 150 ºC. When the compaction was completed, the slab was
left to cool down at room temperature and then removed from the mould. Next each slab was
sawed into 9 beams that were tested in order to evaluate the mixture stiffness modulus. Before
carrying out the stiffness modulus tests, it was necessary to evaluate the dimensions and the

3
apparent density of the specimens, as well as the theoretical maximum density of the
mixtures, in order to calculate their void content values (Table 1).

Table 1 – Statistical analysis of the void content values for the studied mixtures (%)
Mixture Average value Standard deviation Median Maximum Minimum
AC 0/14 3.11 0.37 3.00 3.80 2.81
SMA 0/12 3.43 0.18 3.35 3.75 3.35
PA 0/14 18.80 0.34 18.85 19.29 18.48
AR 0/10 9.94 0.19 10.02 10.02 9.59

As expected, the results show two dense mixtures (AC 0/14 and SMA 0/12), with a reduced
void content (3.11 and 3.43 %), and two open graded mixtures (PA 0/14 and AR 0/10), with
higher void contents (15.80 and 9.94 %, respectively). These void content differences will
certainly result in a dissimilar performance of the mixtures during the stiffness modulus
testing. The void content of all the studied mixtures was slightly below the expected values (4
% for the dense mixtures, 25% for the PA 0/14 mixture and 12 % for the AR 0/10 mixture),
probably due to some elongated flat-shaped aggregates used in the mixture. These problems
may affect the performance of the mixtures, namely the permanent deformation resistance.

Stiffness Modulus of the Studied Mixtures at Different Temperatures

The structural performance of the pavements is directly related to the mechanical behaviour of
the bituminous mixtures, which is characterised through the stiffness modulus, phase angle,
fatigue and reflective cracking resistance, permanent deformation and water susceptibility,
among others. The stiffness modulus of a bituminous mixture mainly depends on the
temperature and loading time, which are related to weather conditions and traffic speed.

The stiffness modulus of the studied mixtures was obtained by using the four-point bending
beam test, with a repetitive sinusoidal loading configuration, in four specimens with a
dimension of 5.1×6.3×38.0 cm3, as stated in the standard EN 12697-26. Each loading cycle
applied a maximum tensile strain of 100×10-6 in the intermediate sector and at the bottom of
the beam. The tests were carried out following a growing order of frequencies, applying 0.1,
0.2, 0.5, 1, 2, 5, and 10 Hz at different test temperatures of 5, 15 and 25 ºC. The variation of
the stiffness modulus and phase angle of the studied mixtures with temperature, at the
reference frequency of 10 Hz, is presented in Figure 2.

Figure 2 shows that AC 0/14 is the mixture with the highest stiffness modulus at all test
temperatures, while PA 0/14 and AR 0/10 mixtures present the lowest moduli. The phase
angle value specifies the type of behaviour of the bituminous mixtures, which is essentially
elastic for lower phase angles and viscous for higher values. The PA 0/14 presents the highest
phase angles at all test temperatures. The lowest phase angle is obtained by the AC 0/14 at 5
and 15 ºC, whereas at 25 ºC it is the AR 0/10 which shows the lowest phase angle. Stiffness
properties of bituminous mixtures are mainly influenced by their void content, and dense
mixtures are those with higher stiffness moduli and, usually, lower phase angles.

4
 12000 35
AC 0/14

Stiffness Modulus (MPa)

10000 SMA 0/12 30
PA 0/14 25

Phase angle (º)

8000 AR 0/10
20
6000
15
AC 0/14
4000
10 SMA 0/12
2000 PA 0/14
5
AR 0/10
0 0
5 10 15 20 25 5 10 15 20 25
Temperature (ºC) Temperature (ºC)

Figure 2 – Influence of the temperature in the stiffness modulus and phase

angle at 10 Hz for the studied bituminous mixtures

In spite of having the highest stiffness modulus, AC 0/14 is also the mixture which shows the
highest thermal susceptibility, since it presents a great variation of the stiffness modulus and
phase angle, especially at elevated temperatures (near 25 ºC). In contrast, the AR 0/10 mixture
has the smallest temperature susceptibility due to the use of a crumb rubber additive.

When the stiffness modulus of bituminous mixtures is determined at different temperatures, it

is possible to present the results by using other representations. Thus, Figure 3 presents the
master curves (reference temperature of 15 ºC) and the complex modulus curves in the Cole
and Cole plane of the studied mixtures. The master curve for a certain reference temperature
is obtained by translating the “modulus-frequency” curves at several temperatures into the
same curve at a reference temperature. The translation factor log(αT) is usually defined by
Equation 1, known as WLF formula (Williams et al., 1955), where C1 and C2 are standard
constants of the mixture, T is the test temperature and TR is the reference temperature.

- C1× ( T - TR )
log (α T ) = (1)
C2 + ( T - TR )

1.0E+05 10000
Stiffness modulus (MPa).

1.0E+04
E2 (MPa)

1000

1.0E+03 AC 0/14
AC 0/14
SMA 0/12
SMA 0/12
PA 0/14
PA 0/14
AR 0/10
AR 0/10
1.0E+02 100
0.001 0.01 0.1 1 10 100 1000 100 1000 10000 100000
Equivalent frequency (Hz) E1 (MPa)

Figure 3 - Master curves (reference temperature of 15 ºC) and complex

modulus curves in the Cole and Cole plane for the studied bituminous mixtures

The master curves and the complex modulus curves in the Cole and Cole plane confirm that
the AC 0/14 is the mixture with highest stiffness properties at different temperatures and
frequencies, but it is also more susceptible to temperature than the other mixtures. The master
curve of AR 0/10 clearly shows that the rubber modified binder greatly reduces the
temperature and frequency susceptibility, especially at higher temperatures. Thus, AC 0/14
was selected for fatigue cracking characterization since it is the mixture with the highest
thermal susceptibility and the most commonly used mixture in surface layers in Portugal.

5
STIFFNESS MODULUS AND FATIGUE CRACKING PROPERTIES OF ASPHALT
CONCRETE 0/14 AT DIFFERENT TEMPERATURES

Stiffness Modulus of AC 0/14 at Different Temperatures

The variation of the stiffness modulus and phase angle of the AC 0/14 mixture with the test frequency,
at the 3 test temperatures (5, 15 and 25 ºC), also known as isothermal curve, is presented in Figure 4.

12000 50
T=5ºC T=15ºC T=25ºC
10000
Stiffness modulus (MPa)

40

Phase angle (º)

8000
30
6000
20
4000

2000 10
T=5ºC T=15ºC T=25ºC
0 0
0.1 1 10 0.1 1 10
Frequency (Hz) Frequency (Hz)

Figure 4 - Isothermal curves of stiffness modulus and phase angle for the
AC0/14 mixture

The evolution of the stiffness modulus and phase angle with the frequency is relatively similar
at 5 and 15 ºC, being slightly different at 25 ºC, when the AC 0/14 starts changing its
behaviour from elastic to viscous. Moreover, a similar variation of the stiffness modulus and
phase angle is observed either when temperature varies 10 ºC or when the frequency changes
from 10 to 1 Hz. The corresponding change in traffic speed is less common to occur than a
temperature variation of 10 ºC, meaning that temperature is usually the main parameter that
controls the stiffness behaviour of bituminous mixtures in road pavements.

Resistance to Fatigue Cracking of AC 0/14 at Different Temperatures

The fatigue cracking resistance of bituminous mixtures is their ability to resist repetitive
bending loads before failure. Fatigue cracking is one of the main mechanisms of pavement
distress and it manifests itself through the appearance of cracks on the surface due to traffic
loading. When fatigue tests are carried out in strain control, the fatigue cracking resistance of
the bituminous mixture is characterised by the strain applied during the test, according to
Equation 2 (where N is the fatigue cracking resistance of the mixture, ε0 is the tensile strain
and a, b are coefficients obtained experimentally).
b
⎛ 1 ⎞
N = a × ⎜⎜ ⎟⎟ (2)
⎝ε0 ⎠

In this work, the fatigue cracking resistance of the AC 0/14 mixture was obtained through
four-point bending tests (two supports and two loading points), with a repetitive sinusoidal
loading configuration without resting periods. These tests were carried out at three different
temperatures (5, 15 and 25 ºC) and at a loading frequency of 10 Hz in fifteen specimens per
temperature (with dimensions of 5.1×6.3×38.0 cm3). Tests were carried out at five different
tensile strain levels (100×10-6, 150×10-6, 250×10-6, 300×10-6 and 350×10-6), with three
specimens tested for each strain level.

6
The initial stiffness modulus of each tested specimen and the number of cycles that
correspond to a 50% reduction of its initial value (failure criterion) was registered, thus
obtaining the fatigue life of the specimen. Some correction factors (known as shift factors)
should be applied in order to develop a meaningful relationship between the results of
laboratory tests and field performance. A shift factor of 20 (Brown et al., 1985) was
considered in order to take into account the influence of rest periods. Rao Tangella et al.
(1990) also adopted a value of 20 as a shift factor, considering high loading rate sinusoidal
bending tests carried out without rest periods. An additional factor of 1.1 (Brown et al., 1985)
has also been used to simulate the lateral distribution of loads, while the fatigue life associated
with the crack propagation phase has been considered by the use of controlled-strain bending
tests (Shell, 1978). Based on these studies, the ultimate value used in the present work was 22.

The fatigue cracking laws of AC 0/14 mixture at different temperatures, presented in Figure 5,
were drawn and determined based on all fatigue tests results, according to Equation 2, after
applying the shift factor of 22 to estimate the field fatigue performance. Figure 5 also presents
the Shell fatigue cracking law (Shell, 1978) for the AC 0/14 mixture (Vb=11.7% and
SMix=5450 MPa) usually used in pavement design, in order to have a reference indicator.

1000
-0.2215
εt = 7498.4×N
2
r = 0.9932 -0.1418
εt = 2079.2×N
Tensile strain (E-6)

2
r = 0.9893

100
-0.1753
εt = 2734.8×N
2
AC 0/14 - T=5ºC r = 0.9431 εt = 3387.4×N
-0.2

AC 0/14 - T=15ºC 2
r =1
AC 0/14 - T=25ºC
AC 0/14 - Shell law
10
1E+05 1E+06 1E+07 1E+08 1E+09
Number of cycles

Figure 5 – Laboratory fatigue cracking laws at different temperatures and Shell

fatigue law obtained for the AC 0/14 mixture

It was observed that, at high strain levels (above 200×10-6), the AC 0/14 mixture presents the
highest fatigue cracking resistance at a temperature of 25 ºC, since the mixture is more
flexible at elevated temperatures, and thus it supports high strain levels that would easily
fracture the mixture if the behaviour were more rigid (at lower temperatures). In contrast, at
lower strain levels the AC 0/14 mixture presents the highest fatigue cracking resistance at a
temperature of 5 ºC, probably because the strain level is very small and it is applied in the
elastic domain of a mixture with a higher stiffness modulus, thus delaying the beginning of
the fatigue cracking phenomenon. The AC 0/14 mixture shows the lowest fatigue cracking
resistance at 15 ºC. Fatigue cracking resistance obtained in the laboratory was superior to that
calculated with the Shell fatigue life equation, given that the latter was obtained with
probabilistic values that are usually more conservative.

The variation of fatigue cracking resistance with the strain level, given by the slope of the
fatigue laws, reduces with the increase of the temperature, since the flexible behaviour at

7
higher temperatures is less sensitive to strain level changes. This variation is similar in the
laboratory and in the Shell fatigue cracking laws at an approximate temperature of 20 ºC.

The parameters of the fatigue laws presented in Table 2 were calculated with basis on the
graphs in Figure 5. Table 2 also illustrates two essential variables for the evaluation of the
fatigue cracking resistance, namely:
ƒ N100 – fatigue cracking resistance for a tensile strain of 100×10-6 (usual value in
pavements);
ƒ EXT6 – tensile strain which causes a fatigue cracking resistance of 1×106 cycles (this is the
main result required by the European Standard EN 12697-24).

Table 2 – Fatigue cracking law parameters used to calculate N100 and EXT6 values
Temperature a b R2 N100 EXT6
23
5 ºC 2.5028×10 7.05219 0.9893 1.968×109 293.15
15 ºC 4.0361×1019 5.70451 0.9431 1.574×108 242.73
17
25 ºC 3.1204×10 4.51467 0.9932 2.917×108 351.54
17 7
Shell – 20 ºC 4.4600×10 5.00000 1.0000 4.460×10 213.73

The AC 0/14 mixture presented its highest N100 at 5 ºC (12.5 or 6.7 times superior than its
fatigue resistance at 15 ºC and 25 ºC, respectively, and 44.1 times superior than the Shell
fatigue life calculation). Moreover, the AC 0/14 mixture supports higher tensile strain in each
loading cycle at a temperature of 25 ºC, in order to achieve a fatigue resistance of 1×106
cycles (20% or 45% higher than the strain level determined at 5 ºC and 15 ºC, respectively,
and 64% higher than the Shell fatigue life calculation).

The evolution of the parameter a in relation to temperature is exponential, while the evolution
of the parameter b is linear. Thus, the influence of the temperature in the fatigue cracking
resistance of the studied AC 0/14 mixture can be expressed by Equation 3.
-0.1290×T+7.6638
−7.2083E − 01×T ⎛1⎞
N = 4.2082E+24 × e ×⎜ ⎟ (3)
⎝ ε0 ⎠

It may be concluded that temperature significantly influences stiffness and fatigue cracking
performance of the bituminous mixtures. It is an important factor, as it controls the pavement
design, which should be further evaluated.

INFLUENCE OF TEMPERATURE ON THE DESIGN OF A BITUMINOUS ROAD

PAVEMENT LAYER

A comparative analysis between the pavement design results achieved by using the Shell
fatigue life equation and those obtained from laboratory tests carried out in the AC 0/14 at
different temperatures was carried out in this part of the work.

The pavement design was carried out by using design charts whereby the thickness of each
layer is obtained as a function of the traffic design. The procedure used to determine the
design charts was based on the approach used by Oliveira et al. (2007). In order to design the
pavement at different temperatures, the stiffness moduli previously obtained at 10 Hz, as well
as the fatigue cracking laws obtained in laboratory and the Shell fatigue life equation, were

8
used. Three pavement foundation classes were considered (50, 100 and 150 MPa), on top of
which a granular sub-base (20 cm) was applied, based on MACOPAV proposal (JAE, 1995).

The bituminous layer (assumed as being the AC 0/14 mixture) thickness, presented in the
design charts, was determined in order to prevent the pavement failure by fatigue cracking
(maximum strain at the bottom of the bituminous layer) and by permanent deformation
(maximum strain at the top of the foundation). The results obtained were based on the
response of each pavement structure to the application of a standard axle load of 80 kN. The
calculations were carried out using a multilayer linear elastic analysis with BISAR 3.0 (Shell,
1998). Several iterations were made for each pavement structure, consecutively increasing the
thickness of the bituminous layer, and determining the expected fatigue life of the pavement
according to the obtained strains. The results of this study are presented in Figure 6, on a
comparative basis, in terms of the total design thickness of the bituminous layer (AC 0/14)
needed at different design temperatures and for different foundation classes.

300
T =5ºC; Found=50MPa

T =5ºC; Found=100MPa
250
T =5ºC; Found=150MPa
AC thickness (mm)

200 T =15ºC; Found=50MPa

T =15ºC; Found=100MPa
150 T =15ºC; Found=150MPa

T =25ºC; Found=50MPa
100
T =25ºC; Found=100MPa

T =25ºC; Found=150MPa
50
Shell=20ºC; Found=50MPa

0 Shell=20ºC; Found=100MPa
1E+05 1E+06 1E+07 1E+08 1E+09 Shell=20ºC; Found=150MPa
ESALs

Figure 6 – AC 0/14 bituminous layers design chart at different temperatures for

pavements with 3 classes of foundation and 20 cm of granular sub-base

It was possible to observe that temperature may have a massive influence in the design of
flexible road pavements, mainly for higher traffic design values. In the example of the AC
0/14 design presented in Figure 6, temperature has a greater influence than the foundation
class. The reduction of temperature increases the fatigue resistance of the bituminous layer,
thus reducing the AC 0/14 thickness needed to bear the traffic loads.

The design of a pavement using Shell fatigue life equation exemplified in this work is very
conservative in some cases (comparison with the design thickness at 5 ºC), leading to the
duplication of the bituminous layer thickness design. However, since Portuguese design
temperatures are usually higher (Picado Santos, 1994), the Shell fatigue equation can be used
with reasonable results, in spite of being conservative to some extent.

CONCLUSIONS

The main conclusions that can be drawn from this study are the following:
ƒ Amongst the four bituminous mixture studied, the traditional AC 0/14 is the one with
highest stiffness modulus and with greatest thermal susceptibility. In contrast, the AR 0/10
showed the smallest temperature susceptibility due to the use of crumb rubber as binder
additive;

9
ƒ Temperature is usually the main parameter controlling the stiffness behaviour of
bituminous mixtures in road pavements if compared with traffic speed;
ƒ At high strain levels (above 200×10-6) the AC 0/14 mixture presents the highest fatigue
cracking resistance at a temperature of 25 ºC. In contrast, at lower strain levels the AC 0/14
mixture presents the highest fatigue cracking resistance at a temperature of 5 ºC;
ƒ The increased fatigue resistance of the AC 0/14 for lower temperatures results in a
reduction in the thickness needed to support the traffic design;
ƒ The Shell fatigue life equation can be used with reasonably good results in pavement
design for temperatures near 20 ºC, in spite of being slightly conservative.

REFERENCES

APORBET (1998). Misturas Betuminosas. Contribuição para a Normalização do Fabrico e

da Aplicação. Lisboa.
Brown, S. F., Brunton, J. M., and Stock, A. F. (1985). The Analytical Design of Bituminous
Pavements. Proc. Institution of Civil Engineers, Part 2, Vol.79, 1-31.
JAE (1995). MACOPAV – Manual de Concepção de Pavimentos para a Rede Rodoviária
Nacional, Junta Autónoma de Estradas, Lisboa.
Lundström, R., Di Benedetto, H. and Isacsson U. (2004). Influence of Asphalt Mixture
Stiffness on Fatigue Failure. Journal of Materials in Civil Engineering,
November/December, pp. 516-525.
Oliveira, J.R.M., Zoorob, S.E., Thom, N.H. and Pereira, P.A.A. (2007). A Simple Approach
to the Design of Pavements Incorporating Grouted Macadams. 4th International
Conference on Bituminous Mixtures and Pavements, Greece.
Picado-Santos, L. (1994) Consideração da Temperatura no Dimensionamento de Pavimentos
Rodoviários Flexíveis. Dissertação submetida para obtenção do grau de Doutor em
Ciências de Engenharia. Vol. 1 e 2. Universidade de Coimbra. Coimbra.
Rao Tangella, S.C.S., Craus, J., Deacon, J.A. and Monismith, C.L. (1990). Summary Report
on Fatigue Response of Asphalt Mixtures, Prepared for Strategic Highway Research
Program, Project A-003-A, Institute of Transportation Studies, University of California,
Berkeley.
RECIPAV (2001). Ficha Técnica – 3/04. MBA-BMB. RECIPAV. Cartaxo, Portugal. 14 May
2007. <http://www.recipav.pt /imagens/fichas_tecnica_bmb_3.pdf>
Shell International Oil Products BV, (1998). BISAR 3.0 – BItumen Stress Analysis in Roads.
User Manual. The Hague.
Shell International Petroleum Company Ltd., (1978). Shell Pavement Design Manual.
London.
VIATOP (2006). 2001 issue ZTVAsphalt-StB 01. JRS - J. Rettenmaier & Söhne
GMBH+CO.KG. Rosemberg, Alemanha. 11 May 2007. <http://www.sma-viatop.com/
SMAviatop_engl/regelwerke/ausgabe2001.shtml?navid=25>
Williams, M.L., Landel, R.F. and Ferry J.D. (1955). The Temperature Dependence of
Relaxation Mechanisms in Amorphous Polymers and Other Glass-Forming Liquids.
Journal of the American Chemical Society, Vol. 77, pp. 3701-3707.

10