PA - D Lec 10

Pavement Analysis and Design
TE-503 A/TE-503
Lecture-10
18-11-2019
Dr. Zia-ur-Rehman
DTEM
Flexible Pavement Design
AASHTO Method
The design procedure recommended by the American
Association of State Highway and Transportation Officials
(AASHTO) is based on the results of the extensive AASHO
Road Test conducted in Ottawa, Illinois, in the late 1950s
and early 1960s.
The AASHO Committee on Design first published an
interim design guide in 1961. It was revised in 1972 and
1981. In 1984-85, the Subcommittee on Pavement Design
and a team of consultants revised and expanded the guide
under National Cooperative Highway Research Program
(NCHRP) Project 20-7/24; they issued the guide in 1986.
The guide was revised in 1993 with practically no change in
the design method.
2

AASHTO Method
The empirical performance equations obtained from the
AASHO Road Test are still being used as the basic models
in the current guide, but were modified and extended to
make them applicable to other regions in the nation. It
should be kept in mind that the original equations were
developed under a given climatic setting with a specific set
of pavement materials and subgrade soils.
The climate at the test site is temperate with an average
annual precipitation of about 34 in. The average depth of
frost penetration is about 28 in. The subgrade soils consists
of A-6 and A-7-6 that are poorly drained, with CBR values
ranging from 2 to 4.
3

AASHTO Method-Design Variables
Time Constraints
To achieve the best use of available funds,
the AASHTO design guide encourages the
use of a longer analysis period for high-
volume facilities, including at least one
rehabilitation period. Thus, the analysis
period should be equal to or greater than
the performance period.

AASHTO Method-Design Variables-Performance Period
The performance period refers to the time that an initial
pavement structure will last before it needs rehabilitation or
the performance time between rehabilitation operations. It
is equivalent to the time elapsed as a new, reconstructed or
rehabilitated structure deteriorates from its initial
serviceability to its terminal serviceability.
The designer must select the performance period within the
minimum and maximum allowable bounds that are
established by agency experience and policy. The selection
of performance period can be affected by such factors as the
functional classification of the pavement, the type and level
of maintenance applied, the funds available for initial
construction, life cycle costs and other engineering
5
considerations.
AASHTO Method-Design Variables-Analysis Period
The analysis period is the period of time that any design
strategy must cover. It may be identical to the selected
performance period. However, realistic performance
limitations may necessitate the consideration of staged
construction or planned rehabilitation for the desired
analysis period.
In the past, pavements were typically designed and analyzed
for a 20-year performance period. It is now recommended
that consideration be given to longer analysis periods
because they can be better suited for the evaluation of
alternative long-term strategies based on life cycle costs.
Table 11.13 contains general guidelines for the length of the
analysis period.
6

AASHTO Method-Design Variables-Analysis Period

AASHTO Method-Design Variables-Traffic
The design procedures are based on cumulative expected
18-kip (80-kN) equivalent single-axle load (ESAL).
If a pavement is designed for the analysis period without

any rehabilitation or resurfacing, all that is required is the
total ESAL over the analysis period.
However, if stage construction is considered and

rehabilitation or resurfacing is anticipated, a graph or
equation of cumulative ESAL versus time is needed so that
the ESAL traffic during any given stages can be obtained.

AASHTO Method-Design Variables-Reliability
Reliability is a means of incorporating some degree of
certainty into the design process to ensure that the various
design alternatives will last the analysis period. The level of
reliability to be used for design should increase as the
volume of traffic, difficulty of diverting traffic and public
expectation of availability increase.
Table 11.14 presents recommended levels of reliability for
various functional classifications.
Application of the reliability concept requires the selection
of a standard deviation that is representative of local
conditions. It is suggested that standard deviations of 0.49
be used for flexible pavements and 0.39 for rigid pavements.
These correspond to variances of 0.2401 and 0.1521.
9

The AASHTO
Definition of
reliability is: “The
reliability of the
pavement
design-performance
process is the
probability that a
pavement section
designed using the
process will perform
satisfactorily over
the traffic and
environmental
conditions for the
design period
10

When stage construction is considered, the reliability of
each stage must be compounded to achieve the overall
reliability; that is,
in which n is the number of stages being considered.
For example, if two stages are contemplated and the desired

level of overall reliability is 95%, the reliability of each stage
must be (0.95)1/2 or 97.5%
11

AASHTO Method-Design Variables-Environmental Effects
The AASHO design equations were based on the results of
traffic tests over a two-year period. The long-term effects of
temperature and moisture on the reduction of serviceability
were not included. If problems of swell clay and frost heave
are significant in a given region and have not been properly
corrected, the loss of serviceability over the analysis period
should be estimated and added to that due to cumulative
traffic loads.
Figure shows the serviceability loss versus time curves for a
specific location. The environmental loss is a summation of
losses from both swelling and frost heave. The chart may be
used to estimate the serviceability loss at any intermediate
period, for example, a loss of 0.73 at the end of 13 years.
12

13

Of course, if only swelling or frost heave is considered, there
will be only one curve on the graph. The shape of these
curves indicates that the serviceability loss due to
environment increases at a decreasing rate. This may favour
the use of stage construction because most of the loss will
occur during the first stage and can be corrected with little
additional loss in later stages.
The serviceability loss due to roadbed swelling depends on
the swell rate constant, the potential vertical rise, and the
swell probability; that due to frost heave depends on the
frost heave rate, the maximum potential serviceability loss,
and the frost heave probability.
14

AASHTO Method-Design Variables-Serviceability
Initial and terminal serviceability indexes must be established to
compute the change in serviceability, ΔPSI, to be used in the
design equations. The initial serviceability index is a function of
pavement type and construction quality.
Typical values from the AASHO Road Test were 4.2 for flexible
pavements and 4.5 for rigid pavements.
The terminal serviceability index is the lowest index that will be
tolerated before rehabilitation, resurfacing and reconstruction
become necessary.
An index of 2.5 or higher is suggested for design of major
highways and 2.0 for highways with lower traffic. For relatively
minor highways where economics dictate a minimum initial capital
outlay, it is suggested that this be accomplished by reducing the
design period or total traffic volume, rather than by designing a
terminal serviceability index less than 2.0. 15

AASHTO Method-Design Variables-Design Equation
16

AASHTO Method-Design Variables-Design Nomograph
17

AASHTO Method-Design Variables-Design Nomograph-Numerical problem
Given W18 = 5 x 106, R = 95%, So = 0.35, MR = 5000 psi and
ΔPSI =1.9, determine SN using nomograph.
Solution: As shown by the arrows on nomograph, starting
from R = 95%, a series of lines are drawn through So=0.35,
W18=5x106, MR=5000 psi and ΔPSI=1.9 and finally intersect
SN at 5.0. So SN=5.0.
18

AASHTO Method-Design Variables-Effective Roadbed Soil Resilient Modulus
The effective roadbed soil resilient modulus MR is an
equivalent modulus that would result in the same damage if
seasonal modulus values were actually used. The equation
for evaluating the relative damage to flexible pavements is
as under:
19

AASHTO Method-Design Variables-Computation of Effective Roadbed Soil Resilient Modulus
20

AASHTO Method-Design Variables-Structural Number
Structural number is a function of layer thicknesses, layer
coefficients and drainage coefficients and can be computed
from:
Layer Coefficient
The layer coefficient ai , is a measure of the relative ability
of a unit thickness of a given material to function as a
structural component of the pavement. Layer coefficients
can be determined from test roads or satellite sections, as
was done in the AASHO Road Test, or from correlations
with material properties, as shown in Figures.
21

AASHTO Method-Design Variables-Structural Number-Layer coefficients
22

23

24

25

26

27

AASHTO Method-Design Variables-Structural Number-Layer Coefficient
It is recommended that the layer coefficient be based on the
resilient modulus, which is a more fundamental material
property. The procedure for determining the resilient
modulus of a particular material varies with its type. Except
for the higher stiffness materials, such as HMA and
stabilized bases, that may be tested by the repeated load
indirect tensile test (ASTM D-4123), all materials should be
tested by the resilient modulus test methods (AASHTO
T274).
In following the AASHTO design guide, the notation MR, as
used herein, refers only to roadbed soils, whereas E1 , E2 and
E3 apply to the HMA, base and subbase, respectively.
28

Asphalt-Concrete Surface Course
Figure is a chart relating the layer coefficient of a dense-
graded HMA to its resilient modulus at 70°F (21°C).
Caution should be used in selecting layer coefficients with
modulus values greater than 450,000 psi (3.1 GPa), because
the use of these larger moduli is accompanied by increased
susceptiblity to thermal and fatigue cracking. The layer
coefficient a1 for the dense-graded HMA used in the
AASHO Road Tests is 0.44, which corresponds to a resilient
modulus of 450,000 psi (3.1 GPa).
29

Asphalt-Concrete Surface Course
30

Untreated and Stabilized Base Courses
Figures (Already shown) show the charts that can be used to
estimate the layer coefficient a2 for untreated, bituminous-
treated, and cement-treated base courses.
In lieu of Figures, the following equation can also be used to
estimate a2 for an untreated base course from its resilient
modulus E2:
a2 = 0.249(log E2 )-0.977
The layer coefficient a2 for the granular base material used

in the AASHO Road Test is 0.14, which corresponds to a
base resilient modulus of 30,000 psi (207 GPa).
31

The resilient modulus of untreated granular materials
depends on the stress state θ, as under:
E2 = K1 θ K2
Typical values of K1 for base materials range from 3000 to

8000; those of K2 range from 0.5 to 0.7.
Values of K1 and K2 for each specific base material should be

determined using AASHTO Method T274. In the absence of
this information, the values shown in Table 11.16 can be
used.
32

33

The resilient modulus of the base course is a
function not only of K1 and K2 , but also of the
stress state θ. Values for the stress state within the
base course vary with the roadbed soil resilient
modulus and with the thickness of the surface
layer. Typical values of θ are shown in Table
11.17. Given K1, K2 and θ, E2 can be determined
from the above equation.
34

35

Granular Subbase Course
Figure (Already shown) provides the chart that may be used
to estimate layer coefficient a3 of granular subbase courses.
The relationship between a3 and E3 can be expressed as:
a3 = 0.227(log E3 )-0.839
The layer coefficient a3 for the granular subbase in the

AASHO Road Test is 0.11, which corresponds to a resilient
modulus of 15,000 psi (104 MPa).
As with granular base courses, values of K1 and K2 for
granular subbase courses can be determined from the
resilient modulus test (AASHTO T274) or estimated from
36
Table 11.18.
37

Values of K1, K2 θ and E3 for the subbase in the AASHO
Road Test are shown in Table 11.19.
38

AASHTO Method-Design Variables-Structural Number-Drainage Coefficient
Depending on the quality of drainage and the availability of
moisture, drainage coefficients m2 and m3 should be applied to
granular bases and subbases to modify the layer coefficients.
At the AASHTO Road Test site, these drainage coefficients are
all equal to 1.
Table 11.20 shows the recommended drainage coefficients for

untreated base and subbase materials in flexible pavements.
The quality of drainage is measured by the length of time for
water to be removed from bases and subbases and depends
primarily on their permeability. The percentage of time during
which the pavement structure is exposed to moisture levels
approaching saturation depends on the average yearly rainfall
and the prevailing drainage conditions.
39

AASHTO Method-Design Variables-Structural Number-Drainage Coefficient
40

AASHTO Method-Design Variables-Structural Number-Selection of Layer Thickness
Once the design structural number SN for an initial

pavement structure is determined, it is necessary to select a
set of thicknesses so that the provided SN, as computed by
above equation, will be greater than the required SN.
Note that above equation does not have a single unique
solution. Many combinations of layer thicknesses are
acceptable, so their cost effectiveness along with the
construction and maintenance constraints must be
considered to avoid the possibility of producing an
impractical design.
41

From a cost-effective viewpoint, if the ratio of costs for
HMA and granular base is less than the corresponding ratio
of layer coefficients times the drainage coefficient, then the
optimum economical design is to use a minimum base
thickness by increasing the HMA thickness.
Minimum Thickness
It is generally impractical and uneconomical to use layers of
material that are less than some minimum thickness.
Furthermore, traffic considerations may dictate the use of a
certain minimum thickness for stability. Table 11.21 shows
the minimum thicknesses of asphalt surface and aggregate
base. Because such minimums depend somewhat on local
practices and conditions, they may be changed if needed.
42

43

General Procedure
44

General Procedure
The procedure for thickness design is usually started from
the top, as shown in Figure and described as follows:
1. Using E2 as MR, determine from nomograph the
structural number SN1 required to protect the base and
compute the thickness of layer 1 from
2. Using E3 as MR, determine from nomograph the

structural number SN2 required to protect the subbase
and compute the thickness of layer 2 from
45

General Procedure
3. Based on the roadbed soil resilient modulus MR,
determine from nomograph the total structural number SN3
required and compute the thickness of layer 3 from
46

Numerical problem
Figure is a pavement system with the resilient moduli, layer
coefficients, and drainage coefficients as shown. If predicted
ESAL = 18.6x106, R = 95%, So = 0.35 and ΔPSI = 2.1, select
thicknesses D1, D2 and D3.
47

Numerical problem
48

AASHTO Method-Stage Construction
If the maximum performance period is less than the
analysis period, any initial structure selected will require an
overlay to last out the analysis period.
The thickest recommended initial structure is that
corresponding to the maximum performance period.
Thinner initial structures, selected for the purpose of life
cycle cost analyses, will result in shorter performance
periods and require thicker overlays to last out the same
analysis period.
The design of the initial structure for stage construction
works the same as that for new construction, except that the
reliability must be compounded over all stages.
49

AASHTO Method-Stage Construction
If the loss of serviceability is caused by traffic
loads alone, the length of the performance period,
which is related to W18, for a given serviceability
loss can be determined from nomograph.
However, if the serviceability loss is caused by
both traffic loads and the environmental effects of
roadbed swelling and frost heave, the
performance period for a given terminal
serviceability can be determined only by an
iterative process, as illustrated by the following
example. 50

AASHTO Method-Stage Construction-Numerical problem
51

AASHTO Method-Stage Construction-Numerical problem
52

Shoulder Design-Prediction of Traffic
The factors affecting shoulder design are similar to those affecting
mainline pavement design. The major difference is the amount of
traffic. Traffic volume on shoulders is lower than on mainline
pavements and more difficult to predict.
Three types of traffic may be considered in shoulder design:

encroaching traffic, parking traffic and regular traffic. Regular
traffic is considered only if the use of shoulder as an additional
lane for peak hour or detoured traffic is anticipated. If there is no
regular traffic, the sum of encroaching and parking traffic is used
to design the inner edge of shoulder adjacent to the mainline
pavement, and parking traffic is used to design the outer edge of
shoulder. If a uniform thickness is used for shoulders, only the
inner edge need be considered.
53

Shoulder Design-Prediction of Traffic
However, this may not be true for rigid pavement shoulders,
because the stresses and deflections at the outer edge are
much greater than those at the inner edge and may cause
more damage even though the traffic volume is smaller.
Encroaching Traffic
When there is a paved shoulder and no lateral obstruction
within the shoulder area, trucks using the outer traffic lane
tend to encroach on the shoulder as much as 12 in. and
sometimes even more. In California (1972), shoulder
sections are designed for 1% of the mainline traffic in the
adjacent lane with a minimum traffic index of 5, which
corresponds to approximately 104 equivalent 18-kip single-
axle load applications.
54

Shoulder Design-Prediction of Traffic-Encroaching Traffic
Results of a study in Georgia (Emery, 1975) showed that the
use of 1% is low for some traffic flow conditions. In the
absence of additional data, Barksdale and Hicks (1979)
recommended that, for free flow traffic conditions in rural
areas of the south, the inner edge of the shoulder should be
designed for at least 2 to 2.5% of the truck traffic on the
outer lane. Because local conditions vary significantly, the
best way to determine the percentage of encroaching traffic
is to make an actual survey on a segment of highway with
paved shoulders, which has the same traffic, geometric, and
topographic conditions as the design case in question.
55

For the study performed in Georgia, trucks were selected at
random and followed by observers for 10 miles. Those
trucks not completing the full 10-mile trip were dropped
from the analysis. The longitudinal distance for each
encroachment is estimated from the prevailing speed and
the time during which the truck encroaches on the shoulder.
The percent encroaching traffic, which is the ratio of load
applications on the shoulder to those on the adjacent lane,
can be computed by:
56

in which Pe is the percent encroaching traffic,

Ne is the total number of encroachments per day,
Le is the average length of each encroachment,
No is the number of load applications per day on the outside lane, and
Lo is the length of observed distance, such as 10 miles used in the Georgia
study.
Note that Ne Le is the total length of encroachment for all trucks and Ne Le /
No is the length of encroachment per truck, so Pe can also be defined as the
length of encroachment per truck within an observed distance Lo.
Field observations indicate that Pe usually varies from 1 to 8% of the

traffic volume on the adjacent lane. The percentage of parking traffic
should be added to Pe , because any truck must encroach to park on the
shoulder. 57

Shoulder Design-Prediction of Traffic-Encroaching Traffic-Numerical problem
Given the number of trucks on the outside lane per day No
= 2239, the number of shoulder encroachments in the 10
mile stretch Ne = 7389, and the average distance of each
encroachment Le=384ft, determine the percent
encroachment.
Pe = [7389 x 384 / 2239 x 52800] x 100 = 2.4%
58

Shoulder Design-Prediction of Traffic- Parking Traffic
Parking traffic is the number of load applications for trucks
that park on the shoulder for emergencies or other
purposes. This information can also be estimated for the
design section by using traffic counts on an existing
pavement with similar traffic and design characteristics.
Parking traffic usually varies greatly along a given route,
depending on geometric and interchange conditions.
Because most trucks park near interchange ramps, it may
be necessary to identify separate design sections for areas
where parking is likely and for areas where minimum
parking is expected. The parking survey should last at least
one day and cover the early morning hours, during which
more parking usually takes place.
59

Shoulder Design-Prediction of Traffic- Parking Traffic
Similar to percent encroaching traffic, the percent parking traffic
can be computed by:
in which Pp is the percent parking traffic, which is the ratio

between parking traffic and the traffic on the outside lane:
Np is the number of parked trucks per day;
Lp is the average distance the trucks drive on the shoulder during a
typical stop, which can be determined from a field survey; No is the
number of trucks traveling on the outside lane per day; and
Lo is the length of segment for the parking survey.
Field observations indicate that Pp may range from 0.0005 to 0.02
percent. Note that the percent parking traffic is much smaller than
the percent encroaching traffic and can usually be neglected. 60
Shoulder Design-Prediction of Traffic- Parking Traffic-Numerical problem
The average number of trucks that might park on a 2-mile
stretch of shoulder during one day could range from 1 to 25.
If the number of trucks on the outside lane is 2951 per day
and a truck drives an average of 200 ft during each parking
maneuver, determine the range of percent parking traffic.
For Np = 1
Pp = [1 x 200 / 2951 x 2 x 5280] x 100 = 0.00064%
For Np = 25
Pp = 25 x 0.00064 = 0.016%
Range of Pp is 0.00064 to 0.016%. 61

Shoulder Design-Prediction of Traffic- Regular Traffic
If it is anticipated that the shoulder will be used by regular
traffic at any stage of its design life, this additional traffic
should be added to the encroaching and parking traffic to
form the total shoulder design traffic. The ultimate design is
to consider the shoulder as an extra lane with the same
traffic and cross section as that of the mainline outer lane.
62

PA - D Lec 10

Uploaded by

Copyright:

Available Formats

PA - D Lec 10

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PA - D Lec 10

Uploaded by

Copyright:

Available Formats

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

If a pavement is designed for the analysis period without

However, if stage construction is considered and

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

in which n is the number of stages being considered.

For example, if two stages are contemplated and the desired

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

The layer coefficient a2 for the granular base material used

Pavement Analysis and Design

Typical values of K1 for base materials range from 3000 to

Values of K1 and K2 for each specific base material should be

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

The layer coefficient a3 for the granular subbase in the

Pavement Analysis and Design

Pavement Analysis and Design

Table 11.20 shows the recommended drainage coefficients for

Pavement Analysis and Design

Pavement Analysis and Design

Once the design structural number SN for an initial

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

2. Using E3 as MR, determine from nomograph the

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design

Pavement Analysis and Design