Lesson01 WBT

Welcome
to the Highway Materials Engineering Course (HMEC) Module E, Lesson 1:
Introduction. This lesson will introduce you to the newest approach in pavement design:
the AASHTO Pavement Mechanistic Empirical (ME) Design software. The AASHTO
Pavement ME Design software carries forth the design procedures developed in the
Mechanistic Empirical Pavement Design Guide (MEPDG), which was balloted into practice
by AASHTO in 2007. This lesson will provide you with a baseline knowledge in order to
prepare you for the more detailed lessons to be covered in the instructor‐led portion of this
module.
A printer-friendly version of the lesson materials can be downloaded by selecting the

paperclip icon. Only the slides for the this lesson are available.
If you need technical assistance during the training, please select the Help link in
the upper right-hand corner of the screen.
Image description: People sitting in front of computers in a training setting.

The Highway Materials Engineering Course is a comprehensive, 11‐week training event.
Module E: Mechanistic Empirical Pavement Design Guide (MEPDG) is the seventh module
in the HMEC. Although other modules and workshops are presented on MEPDG, Module E
is expected to provide content that is specific to materials engineers.
Participants will be given the information that they need to more proficiently and actively
review agency‐ and consultant‐generated pavement designs to ensure their alignment with
the appropriate materials designs. Participants who do not normally review pavement
design, or only perform a cursory review, will learn how to review pavement design with
more knowledge of what they are looking at, and how the material design selected by the
designer may impact the ultimate pavement service life. This was not done before the
advent of the MEPDG approach.
On the next screen, we’ll review the lesson structure of this module.
Module E consists of six lessons. Lesson 1 is completed in the Web‐based training (WBT)
portion of the module. Lessons 2–6 are completed in the instructor‐led training (ILT)
portion of the module, before the review and final assessment.
By the end of this lesson, you will be able to:
• List the basic factors required in designing a pavement structure;
• Compare mixture design and pavement design related to materials in terms of predicted
pavement performance;
• Distinguish between empirical and mechanistic empirical design methods;
• List the factors that may affect the predicted performance of pavements in Pavement ME
Design;
• Describe the steps of a Pavement ME Design analysis;
• Describe the iterative design process used with Pavement ME Design software;
• Describe the performance parameters analyzed by Pavement ME Design software;
• Explain how pavement prediction models from Pavement ME Design are being
incorporated into the evaluation of asphalt and concrete paving mixtures; and
• Using an example, identify important considerations for pavement design.
During this lesson, knowledge checks are provided to test your understanding of the
material presented.
This lesson will take approximately 180 minutes to complete.

During this lesson, you will be prompted to reference the lesson exercise document. The
documents referenced during this lesson are attached to the lesson in the paperclip icon.
Please take a moment to open and print the document.
Let’s start by reviewing how familiar you are with aspects of the new mechanistic empirical
design of new and rehabilitated pavement structures.
Read each of the five questions presented here and consider the answer. Then, select each
question to review your knowledge of MEPDG.
MEPDG stands for the Mechanistic Empirical Pavement Design Guide.
The MEPDG is a design approach based on mechanistic empirical principles.
Mechanistically, it calculates pavement responses (stresses, strains, and deflections) due to
loading and environmental conditions and uses those responses to compute incremental
damage over time. Empirically, it relates the cumulative damage to observed pavement
distresses.
The empirical model predicts damage, then is calibrated against measured field distress.
Although practitioners may reach a point in the future when this relationship can be
mechanistically predicted through fracture mechanics or some other means, the current
model in the Pavement ME Design software is empirically‐based.
The MEPDG is a design approach. The software based on the MEPDG design approach is
the Pavement ME Design version 2.0.
The functions of the MEPDG are as a pavement design tool (for pavement practitioners to
design pavement layers with a practical approach at reasonable precision levels by
implementing strategic thickness iterations to the process) and a pavement analysis tool
(distress predictions that include rutting, cracking, and rideability or International
Roughness Index (IRI) for flexible pavements and faulting, cracking, punchouts, and IRI for
rigid pavements.
The various inputs are used to represent different means for obtaining the parameters
needed to test the appropriateness of a trial pavement design. Generally the numerically
lower level inputs (Levels 1 and 2) will result in a higher design reliability, which means that
the design you ultimately end up with will have a greater chance in more of the cases to
successfully reach its intended design life.
Input Level 1 represents the greatest knowledge about the input parameter for a specific
project but has the highest testing and data collection costs to determine the input value.
Level 1 is used for pavement designs having unusual site features, materials, or traffic
conditions that are outside the inference—space used to develop the correlations and
defaults included for input Levels 2 and 3.
Input Level 2 presents the input value that is calculated from other site specific data or
parameters that are less costly to measure. Input Level 2 also represents measured regional
values that are not project‐specific.
Finally, Level 3 inputs are based on global or regional default values—that is, the median
value from a group of data with similar characteristics. This input level contains the least
detail about the input parameters for the specific project; however, it has the lowest
testing and data collection costs.
In this course we will discuss the various input levels as they relate mostly to materials to
gain some information on how input levels can vary across parameters and how much
influence the selection of different input levels will ultimately have on the predicted
performance of the trial design.
As we just reviewed, there are various types of inputs and among them, there are different
levels of specificity with obtaining or selecting the input values. In the past, pavement
designers were able to generate the inputs and design pavements with very little need to
interact with other units in a transportation organization. This approach did not necessarily
end up with pavements that lasted as long as they should, and in many cases, the
pavements were not as economical as they could be and were oftentimes over‐designed.
These reasons and others like them are some of the motivating factors for how the MEPDG
was developed. From the advent of better technology, such as weigh‐in‐motion sensors
embedded in pavements that measure real‐time traffic loads to the national
implementation of Superpave asphalt mixtures, there was a need for pavement design to
align with modern technology, modern traffic, and modern materials. Pavement designers
need to consider the influence of traffic, climate, materials, and construction properties.
Image description: Schematic of ME Pavement Design process.
Pavement designers have a lot to consider. But where does all the data come from? Select
each data source to learn about different types of data available to assist pavement
designers in building better‐performing and more economical pavements.
Image 1 description: Asphalt pavement layers.
Image 2 description: Pavement design construction asphalt surfacing.
Image 3 description: Trucks in congested traffic.
Image 4 description: Collage of climate conditions.
There are a number of properties that describe the characterization of the typical materials
used in pavement structures. They are generalized into three categories.
Pavement response model material inputs relate to the moduli and Poisson’s ratio used to
characterize layer behavior within the specific model. Bound materials generally display a
linear or nearly linear stress‐strain relationship. Unbound materials display stress‐
dependent properties. Granular materials generally are “stress hardening” and show an
increase in modulus with an increase in stress. Fine‐grained soils generally are “stress
softening” and display a modulus decrease with increased stress.
On the other hand, material‐related pavement distress criteria are linked to some measure
of material strength (shear strength, compressive strength, modulus of rupture) or to some
manifestation of the actual distress effect (repeated load permanent deformation, fatigue
failure of Portland cement concrete (PCC) materials).
For all materials used in a pavement structure, the following parameters must be
described: Unit weight; Poisson’s ratio; and other thermal properties such as thermal
conductivity, heat capacity, and surface absorptivity. Additional details on each property
will be presented in the instructor‐led training lessons of this module.
Image description: Asphalt pavement layers.
The unbound layers in a pavement include any granular layers which support the paved surface.
In the Pavement ME Design software, the granular base, subbase, and subgrade layers are defined by
primarily their strength (through classification of the aggregate or soil type and the resilient modulus—a
measure of stiffness under repeated loading) and their properties in the presence of water (through defining
the saturation properties and the effect of moisture on density of these layers).
Classification properties measure the gradation and Atterberg limits from bulk sample recovered
from the subsurface investigation. Volumetric properties of hot mix asphalt (HMA) mixtures refer to
air voids, asphalt content, and aggregate gradation. Whenever this volumetric data is unavailable
from construction records, selected cores recovered from the project are used to measure these
properties. Volumetric Properties of unbound layers and subgrade measure the moisture content
and dry density of undisturbed samples recovered during the subsurface investigation.
Moisture‐density and soil‐water characteristic relationships are used to characterize the unbound
layers and subgrade. The moisture‐density relationship focuses on the target density and moisture
content of the unbound layers and subgrade. The soil‐water characteristic relationship focuses on a
specialized set of moisture retention parameters that are fundamental to predicting the moisture
content and soil stiffness of the subgrade and unbound pavement layers.
Saturated hydraulic conductivity is required to determine the transient moisture profiles in
compacted unbound materials and to compute their drainage characteristics.
The properties listed can be measured in the laboratory, in the field, or by using tools such as the Arizona
State University Soil Unit Map tool.
There are a number of inputs that relate to layers constructed HMA. This would include the
asphalt concrete surface layers as well as a bituminous‐treated base layer.
The inputs required by the Pavement ME Design software include the HMA dynamic
modulus (a measure of stiffness with load frequency and temperature), the HMA creep
compliance, asphalt concrete indirect tensile strength (at low temperatures), mixture
volumetric properties, and the HMA coefficient of thermal expansion. These are only some
of the inputs that are most sensitive for asphalt pavements, but are not the only inputs that
are or could be used in the Pavement ME Design software.
For these properties, there are three levels of inputs. Level 1 inputs are project‐
specific and are measured directly from laboratory or field testing. Level 2 inputs
are estimated from other material properties using predictive equations. Level 3
inputs are based on global or regional default values.
Therefore, the properties listed can be measured in the laboratory or by using the defaults
provided in the software.
There are a number of inputs that relate to layers constructed with concrete materials. This
would include the concrete surface slabs as well as a cement‐treated base layer.
The inputs required by the Pavement ME Design software include the Portland cement
concrete elastic modulus (a measure of stiffness), the flexural strength (or modulus of
rupture), coefficient of thermal expansion, and in the case of continuously‐reinforced
concrete pavements (CRCP), the indirect tensile strength of the concrete. These inputs
were found through a number of research studies to be some of the most sensitive inputs
for PCC; however, they are not the only inputs that a designer can use in Pavement ME
Design. For example, the concrete’s compressive strength is a Level 3 strength input and
the coefficient of thermal expansion (CTE) is a thermal input to concrete pavement design.
The properties listed can be measured in the laboratory or by using the defaults provided in
the software.
Image description: Concrete pavement layer.
The properties of the trial pavement structure are critical to starting a design analysis in
Pavement ME Design software. A user must first select the type of, number of, and
thickness of each layer. One can also define the level of bond between the paved layers
(denoted by an estimate of friction between the paved surfaces and also with the
underlying base layer), which is a consideration of conditions during the pavement’s
construction.
The endurance limit is an input for flexible pavements that is defined as the tensile strain or
stress below which no load‐related fatigue damage occurs.
In the case of designing a pavement rehabilitation, the existing pavement condition must
also be input. The structural condition for concrete pavements is described by input values
for the percent of slabs cracked or for the number of punch outs. The structural condition
for asphalt pavements is described by input values for amount of surface area that is
cracked.
The final input for general pavement structure that considers construction aspects is the
milling depth for flexible pavement overlays.
Image description: Pavement design construction asphalt surfacing.
There are a number of inputs that relate to traffic that can be relatively straightforward,
such as the traffic volumes, heavy vehicle percentages, lane and directional truck
distributions, operational (actual speed passengers are expected to travel) speed, and the
number of lanes. However, other traffic inputs can be much more complex, such as the
truck classification percentages, type and amount of traffic growth, axle load spectra, and
the traffic volume adjustment factors.
These are just a summary of the traffic inputs but many more exist in the module for traffic.
Data for the traffic inputs can be measured by weigh‐in‐motion stations in the field or can
come from projection data from transportation planners.
In this module, we will introduce the traffic inputs but will not cover them in extensive
detail.
Image description: Trucks in congested traffic.
The climate inputs are obtained from a climate file (weather station data). The Enhanced
Integrated Climatic Model (EICM) uses this climate data to model the temperature
distribution throughout the pavement layers. The EICM is a tool used to model
temperature and moisture within each pavement layer and the foundation soil. This
climatic model considers hourly ambient climatic data in the form of temperatures,
precipitation, wind speed, cloud cover, and relative humidity from weather stations across
the US for estimating pavement layer temperatures and moisture conditions.
The depth to groundwater table must be input by the user. It can be defined as an annual
value or can be input seasonally (meaning, one input value per each of the four seasons).
The depth to groundwater table can be measured and there are a number of tools to find
it, including the FHWA long‐term pavement performance (LTPP) database, the ASU Soil Unit
Map application, the USGS or other equivalent maps, or by using the default value in the
Pavement ME Design software.
Drainage is accounted for in a general sense from the EICM by the inclusion of an

infiltration and drainage model. More information on how the pavement materials are
adjusted by the EICM will be provided in the instructor‐led portion of this module.
Image description: Collage of climate conditions.
In addition to providing data to assist pavement designers in building better‐performing
and more economical pavements, materials characterization, traffic, and pavement
structural features are also required for defining the pavement management properties in
the design of rehabilitation for existing pavements.
Pavement existing structural data can be obtained through the evaluation of existing
pavements, such as by the use of falling weight deflectometer, light weight deflectometer,
or other means of assessing the remaining structural value of the original pavement.
Image 1 description: Asphalt pavement layers.
Image 2 description: Pavement design construction tar surfacing.
Image 3 description: Tail lights from congested traffic.
Select all that apply. Which factors are required in designing a pavement structure?
a) Layer types and thickness;
b) Project location;
c) Coefficient of thermal expansion;
d) Reliability;
e) Axle load spectra; or
f) Depth to groundwater table.
The correct answers are a) Layer types and thickness, b) Project location, c) Coefficient of
thermal expansion, e) Axle load spectra, and f) Depth to groundwater table.
Let’s answer another review question about the function of the MEPDG. Does the MEPDG
have a process for mixture design?
Select the answer button the learn more.
Image description: A series of five images displaying C.M., water, air, fine aggregate, and
course aggregate.
course aggregate.
The MEPDG approach was not developed in a way that it could effectively be used to
design mixes. However, its trial‐and‐error process can be used to help indicate whether an
existing pavement design meets the performance criteria within a certain level of reliability.
The issue of mixture durability and its influence on performance are not modeled in the
Pavement ME Design software at this time. The current MEPDG approach assumes that any
issues specific to mixtures are dealt with externally in the mixture design or materials
specification process.
course aggregate.
Let’s pause for another knowledge check.
True or false? The MEPDG has the capability to design asphalt mixtures.
a) True; or
b) False.
The correct answer is b) False.
Let’s briefly review the history of design methodologies.
Select each year on the timeline to learn more.
Image description: Photo of a construction site along a stretch of road.
Early on with the advent of paved surface highways and the emergence of automobiles, the
typical pavement design did not consider the traffic or environmental loads, nor any
consistent attention to soil conditions. The common knowledge at the time of soil
mechanics was applied and the basic premises for pavement design (to protect the
subgrade) used from the time of the Roman Empire onward was followed.
During World War II, the equivalent single wheel load (ESWL) was developed by the armed
forces for airfield pavements. This mechanistic–based procedure was founded on the idea
that the pavement’s thickness is governed by a single‐wheel load. The design approach was
based on fixed traffic and was used for heavy load and low‐volume pavements.
The Maryland Road Test (1950) was conducted on a 1.1‐mile section of existing US 301.
The principal objective of this road test was to determine the relative effects of four
different axle loadings using two vehicle types on a specific concrete pavement design. The
loads employed were 18,000 lbs. and 22,400 lbs. on single axles, and 32,000 lbs. and
44,000 lbs. on tandem axles.
The WASHO Road Test (1953) was conducted on 1 mile of specially built flexible pavement
in Malad, Idaho. Four loops were constructed under carefully controlled conditions; each
had a variety of surface and base thicknesses. Loop loadings were similar to those of the
Maryland Road Test. Elaborate measurements were taken to determine pavement
behavior.
Among the most significant findings of the WASHO test were:
1. The asphalt pavement contributes substantially to the strength of the roadway structure;
and
2. By carrying the pavement structure for the full roadway width, including shoulders,
substantial reductions in thickness can be realized.
Finally, in 1961 the AASHO Road Test was constructed and formed the basis of the AASHO
design that was followed until the advent of the MEPDG approach.
The AASHO Road Test was the last of the major road tests in the United States, conducted
about 80 miles southwest of Chicago. This site was chosen because the soil within the area
is uniform and is representative of that found in large areas of the country.
The basic principles established and validated by the road test still serve as the basis for a
large number of performance‐based design procedures being used in the US today.
We’ll discuss this road test in more detail on the following screens.
The AASHO Road Test was constructed in Ottawa, Illinois between 1958 and 1960, and
included both new flexible and new rigid pavement test pavements. This test road provided
the opportunity to study the field performance of pavements, which could be used to
create empirical models for pavement design. The major focus of the study was the effects
of axle loading types on pavement structures.
Image description: Photo of the AASHTO Road Test.
Now we’ll review the schematic of the AASHO test loops.
This diagram presents the overall view of the frontage roads and test track loops.
This diagram shows the flexible and rigid pavement strips on the top and bottom tangents
of the track.
The performance data that was collected included mechanical responses (such as
deflections and strains) and roughness measurements. The results from the AASHO Road
Test were also used to develop the present serviceability index (PSI), which established a
correlation between user assessment (ride experience) and performance indicators
(measurable characteristics).
Image description: Series of diagrams of (1) flexible pavement strips on the top tangent of
the track; and (2) the rigid pavement strips on the bottom tangent of the track.
There were a number of achievements that derived from the AASHO Road Test, but above
all it resulted in a simplified pavement design based in field observations (an empirical
process), which is why it was in practice for more than 40 years.
The performance equations for different pavement damage types were established. These
helped in two ways: 1. To establish the measure of whether a pavement is meeting its
intended purpose (through the serviceability concept), and 2. To calculate whether the
pavement is meeting its intended structural capacity (through the structural number
concept).
How is serviceability defined? What is the structural number (SN)? Select the Answer
button to learn more.
Image description: Photo of the AASHO Road Test track.
Image description: Photo of the AASHO Road Test track.
The serviceability can be defined as the initial user rating of the pavement minus the
terminal user rating of the pavement, in which:
• PSI0 = Initial serviceability after construction;
• PSIT = Terminal serviceability after end of pavement life (damage); and
• ΔPSI = Change in serviceability during the life of the pavement.
Select the next arrow for the definition of the structural number.
The structural number (SN) is used to determine what stiffness and thickness is needed to
protect the subgrade for a FLEXIBLE PAVEMENT design. It is also used to keep each
individual layer from being overstressed. The nomographs were developed to get the SN
needed to provide adequate protection for Flexible pavements.
Image description: Nomographs.
It is important to recognize that at the time (1950s) when the AASHTO test procedure was
developed, this road test marked a significant improvement in the way pavements were
built and it was a remarkable effort at that point in time. Four major ideas and
advancement came from the AASHO Road Test:
1. The serviceability concept (user evaluation of pavement condition was considered as
part of the design process);
2. The ESAL process (traffic stream was homogenized into a factor that represented
damage to the pavement);
3. The structural number (flexible pavement structure was reduced to a single number);
and
4. The effective modulus of subgrade reaction (rigid pavement foundation considered as a
composite value of all underlying granular layers, including shallow bedrock). The empirical
design for rigid pavements was a three‐way relationship amongst slab thickness, traffic and
loss of serviceability.
These enabled the design procedure to be simplified into a pavement design nomograph
for flexible pavements structural number, and a 2‐part nomograph for rigid pavement
thickness. Pavement design became standardized throughout the US and much of the
world.
What did the axle load study, conducted through the AASHO Road Test, accomplish? Select
the answer button to learn more.
The axle load study conducted through the AASHO Road Test provided a way to relate
traffic damage to pavement performance through the equivalent single axle load (ESALs)
concept.
The traffic design is based on the total number of passes of the standard axle load and how
that translates to damage in the pavement.
In addition, load equivalency factors (LEFs) were established to define the damage per pass
to pavement by the axle in question relative to the damage per pass of a standard 18,000‐
lb. axle load.
Image description: Heavy duty tank transporter used in the AASHO Road Test.
Considering the advances made in understanding material characterization, the effect of climate on
pavement materials and pavement response, and in monitoring of pavement performance, it is now time for
a new design procedure. The old AASHTO design procedure has perhaps outlived its capacity.
There are a number of pressing reasons why the design procedure based on the results of the AASHO Road
Test began to be less effective for today’s pavements and these limitations include the following
considerations:
• The models generated from the road test data were based on only one climate and in situ soil type;
• The testing itself was run over the course of two years, which means that pavement materials perhaps did
not develop the aging properties or other changes in properties with time that a real in‐service pavement
would exhibit;
• There was a maximum of 2‐million axle loads applied during the testing, which is only a fraction of what
many high‐type pavements experience in modern times;
• The truck suspension systems, tires, and construction equipment were representative of the late 1950s and
much has evolved in these areas over the past 60 years;
• Paving materials and mixture designs are reflective of a time when polymer modification, reclaimed asphalt
or concrete pavement, widened slabs, tied shoulders, and certain types of base stabilization did not exist; and
• The sites constructed in the road test were for new roadways, whereas the vast majority of roadway
construction is currently for rehabilitation consisting of overlays or partial‐replacement of an existing
pavement.
These and other limitations illustrate the need for moving pavement design procedure to something more
reflective of the technological environment that designers are working in today.
Image 1 description: Listing of the load configurations on each of the test track’s heavy vehicles’ single and
tandem axles, given for each test loop.
Image 2 description: Listing of the asphalt and concrete test pavement substructures showing the thickness
of each of the test loop typical sections.
Another serious drawback of the AASHO based procedure is the limitation of the
serviceability concept. Under the AASHO design process, serviceability was defined by two
points: initial and terminal.
Pavement managers have been trying to define the performance between these curves for
some time. The most common curve shape is represented by curve A. But, there have been
many a pavement which performed in a way such as curve B, and even a few pavements
that have shown performance like that of curve C.
So, which is the correct shape? Do we have to wait until after we build pavement to
determine how they perform? What does the AASHO data‐based procedures tell us about
points between initial and terminal?
Image description: Graphic of AASHO design process.
For the design of Flexible Pavements, the Structural number (SN) is the tool which was
developed through the AASHO procedure used to determine the pavement structure
appropriate for traffic and serviceability. But, this number means very little outside of the
road test. The SN was used to approximate the structural bearing capacity of an individual
flexible pavement layer, based on its layer properties (layer coefficient) and thickness. The
SN was used along with the layer and moisture coefficients to calculate the required
thickness of each individual pavement layer above the subgrade. The SN of flexible
pavements is not a fundamental engineering property and cannot be determined in the
laboratory, nor can it be established for new materials that were not used at the time of
the construction of the road test.
In Concrete Pavement design, the basic performance relationship is the log(W) = f (D, ΔPSI)
where W = cumulative axle loads, D = slab thickness, and ΔPSI = loss of serviceability. The
serviceability was based on the observed performance at the AASHO road test and doesn’t
reflect serviceability of pavements in very different climate locations, different materials
performance, or traffic. Therefore, designers are limited in how much they can optimize for
thinner pavements and build more cost effectively.
With the lack of a load carrying capacity to relate to a pavement, a crucial element in
designing pavements today, one asks themselves the question of whether this is the best
approach upon which to base our present‐day pavement design process?
Select each type of pavement to learn more.
Other limitations exist for flexible pavements. The original AASHO design procedure was
not based on fundamental properties of asphalt mixtures and an estimate of its stiffness or
layer coefficient were not properties that could be directly measured in a laboratory. The
models are based on the asphalt mixes of 60 years ago and do not include the new
materials that relate to Superpave binders or other types of asphalt mixtures (with RAP,
warm mix asphalt, etc.).
Image description: Illustration displaying surface course, base course, sub‐base course, and
subgrade.
In terms of rigid pavements, there are also a few limitations. The modulus of subgrade
reaction (k) is used to estimate the “support” of the PCC slab by the layers below. Usually,
an “effective” k (keff) is calculated, which reflects base, subbase, and subgrade contributions
as well as the loss of support that occurs over time due to erosion and stripping of the
base, subbase, and subgrade.
Typically, large changes in keff have only a modest impact on PCC slab thickness. However,
the modulus of subgrade reaction is not a true engineering property for the foundation soil
because it depends not only upon the soil stiffness, but also upon the slab (or footing) size
and stiffness. In addition, the stabilized layers in rigid pavements were represented only by
bumping up the stiffness proportionately to the layer thickness of the base. There was no
laboratory testing to really determine what the relationship between a stabilized base and
the layer thickness of both the concrete layer and the base layer should be.
Image description: Illustration displaying surface course, base course, sub‐base course, and
subgrade.
The developers of the AASHO method realized the mathematical difficulty in handling
mixed traffic in the loss of PSI equations, therefore they decided to apply an equivalent
applications approach and came up with the 18‐kip equivalent single‐axle load or ESAL.
Coefficient multipliers are used to convert other axle weights to ESALs.
Image description: Chart the depicts ESALs per vehicle. A car is 0.0007; a delivery truck is
0.10; a loaded 18‐wheeler is 1.35; a loaded 40‐ft. bus is 1.85; and a loaded 60‐ft.
articulated bus is 5.11.
Why are flexible and rigid pavement ESALs different rather than the same value?
Because, the ESAL coefficient multiplier is based on the following factors:
• Specific flexible layer thickness (SN);
• Specific rigid slab thickness (D);
• Specific axle group (single, tandem, or tridem);
• Specific truck weight range; and
• Specific serviceability loss (ΔPSI).
Therefore, rigid and flexible ESALs are different because their coefficient multipliers were
derived from different empirical pavement experiments.
This graph demonstrates one of the most serious drawbacks to using the AASHO Road Test
as a basis for the pavement design. The AASHO Road Test data was based upon less than 2‐
million axle applications. This information can be projected several different ways as
illustrated by projection A’s straight line (the increase in traffic gets a proportional increases
in thickness. Or by projection B that says that as the traffic increases, the required
thickness required increases at a substantial rate. Or projection C that says the thickness
requirement varies with different levels of traffic.
The real answer is that we don’t know exactly which way it will go, which is why a more
accurate description of traffic (through load spectra) is so important.
Now we are designing some roads approaching 100‐million load applications in this area.
Which of these projections is correct, based upon the original data produced in the AASHO
Road Test?
The bottom line is that we are way beyond the traffic limits represented in the AASHO Road
Test.
Image description: Graph demonstrating drawbacks to using the AASHO Road Test as a

basis for the pavement design.
There are essentially four approaches to pavement design: experience, empirical,
mechanistic empirical, and mechanistic. Let’s review the differences between these
methods of design so that you can understand how the old AASHTO design based on
empirical models is different from the new ME design concept. The other two extremes—
design based on “experience” and “mechanistic concepts alone”—are provided here to
give you the right perspective about where ME design stands in the big picture.
Select each design methodology to learn more.
Image description: Photo of pavement.
In the experience methodology, the pavement design decision is made based on
what is typically done in the area.
In the empirical methodology, models were developed based on observations of
performance and then were extended to similar pavement types.
In the mechanistic empirical methodology, mechanistic theory is used to predict the
pavement’s response to loading, and then empirical models are used to relate these
responses to distress levels.
In the mechanistic methodology, distresses are calculated directly from loading by using
calculated stresses, strains, and deflections. In addition, crack propagation theory and
particle displacement predictions are used to determine the appropriate pavement design.
This continuum qualitatively summarizes pavement design practice as it exists today. Note
the distance between the current practice and state‐of‐practice and between the state‐of‐
practice and state‐of‐the‐art. Most agencies (approximately 90%) currently use AASHTO‐
based empirical design procedures (72, 86, or 93). The state‐of‐practice is the most
advanced practice being used by DOTs at this point (some use mechanistic empirical
design). The state‐of‐the‐art is where current research activities in pavement design are
leading us.
Recall the empirical design (such as the AASHTO 1993 version) is based on statistical

models from road tests. Then, mechanistic empirical design includes the calculation of
stresses/strains combined with empirical pavement performance models, like the MEPDG
approach.
Finally, a mechanistic design would include mechanistic‐based pavement performance

models and is the area in the continuum where researchers are pushing forward toward for
the future of pavement design.
Image description: Graphic of the pavement design continuum.
There are a number of differences between the 1993 AASHTO design and the new ME
design approach. Let’s take a quick look at how the two approaches differ in order to
understand why and how pavement designers are changing the way that they do business.
Select each term to learn more.
The AASHTO 1993 design is based on field measurements and observations taken from the
AASHO Road Test, which were then used to create empirical models including statistical
correlations. The AASHTO 1993 design further refined the statistical models and empirical
measurements for use as inputs in the estimation of pavement performance and
calculation of required pavement layer thicknesses.
One of the major differences in the two approaches is how the traffic loads are
characterized. The AASHTO 1993 design is still founded on the traffic loads originally
modeled from the AASHO Road Test.
One of the major differences in design approach lies in how traffic loads are characterized.
The ME approach only considers FHWA vehicle classes 4 and higher and relies on
information such as truck speed, axle configuration, etc.
ME design defaults are based on measurements from the FHWA long‐term pavement
performance field sections all across the United States.
Structural responses (stresses, strains, and deflections) are mechanistically‐calculated in
ME design and are based on material properties, environmental conditions, and loading
characteristics.
There are two types of empirical models used in the MEPDG. One model predicts the
distress directly (e.g., rutting model for flexible pavements, and faulting for rigid
pavements). A second model predicts the damage, which is then calibrated against
measured field distress (for example, fatigue cracking for flexible pavements and punch
outs for continuously reinforced rigid pavements).
There are a number of benefits to using the MEPDG design approach. One benefit includes
the improved potential for designs reaching their intended service life through improved
levels of design reliability. The MEPDG allows for better design of rehabilitations for
pavements through either flexible or rigid overlays. It has the capability to address unique
pavement sections, such as composition pavement structures as well as rehabilitation. The
program uses the same principles to analyze existing pavements.
There is a method for locally calibrating the distress models in the Pavement ME Design
software to conditions in the State or location where the design will eventually be built.
Also, the MEPDG approach permits for better evaluation of the impacts of both loading
types (that is, traffic and climate loads) through applying various traffic levels, seasonal
traffic, or seasonal drainage effects, and through yearly changes in material properties.
The ME design approach also integrates the effects of current and future loadings, site‐
specific climate (ICM), and the material changes that occur over time.
Image description: Construction crew applying a cape seal to an existing flexible pavement.
Cape seal treatments prevent the occurrence of loose stones in chip seals, used in flexible
pavements with high traffic volumes.
Two surveys were conducted in the past 10 years to determine the state of the practice of
pavement design in the US, in 2003 and 2010.
Select each survey year to review the survey results.
The FHWA first queried its division offices in 2003 as to the practices used by the DOTs for
pavement design. It was observed that 0% of the States were using MEPDG approach for
their design procedure.
A second question common between the two surveys was related to whether DOTs were
pursuing local calibration or implementation of the MEPDG design approach. Select the
forward arrow to learn more.
Image description: Pie chart showing the 2003 survey results of the state of the practice in
pavement design.
Source: FHWA Design Guide Implementation Survey (2003)
In 2003, there were 28 States who were actively collecting data to be able to support local
calibration.
pavement design.
Source: FHWA Design Guide Implementation Survey (2003).
Seven years after the 2003 survey was conducted, FHWA again asked its division offices as
to the practices used by the DOTs for pavement design. Two States were using a
combination of MEPDG and their DOT design approach, and six States were using a
combination of AASHTO 1993 and the MEPDG.
Select the next arrow to learn about whether DOTs were pursuing local calibration or
implementation of the MEPDG design approach.
pavement design.
Source: Transportation Research e‐c155 (2010)
By 2010, 22 States reported that they had a complete local calibration plan in place for
using MEPDG as their partial or full design approach.
pavement design.
Source: Transportation Research e‐c155 (2010)
Let’s review some real‐world examples to help you apply the information presented in this
lesson. This screen introduces four different States and corresponding sample pavement
design projects (both flexible and rigid), which will be used in this lesson, and the following
instructor‐led lessons in this module, to illustrate concepts in the Pavement ME Design
software.
Select each highlighted State to review the current state of the practice in terms of
pavement design for that State.
Image description: Map of the United States with the states of Washington, Utah, Indiana,
and Florida highlighted.
Review the pavement design information for Florida. This information, and a streamlined
version of the entire FDOT standard, can be found in the Module E, Lesson 1 Exercises PDF
document. To download this document, select the paperclip icon.
Image description: Documents from sample pavement projects, Florida.
Review the pavement design information for Indiana. This information, and a streamlined
version of the entire Indiana DOT standard, can be found in the Module E, Lesson 1
Exercises PDF document. To download this document, select the paperclip icon.
Image description: Documents from sample pavement projects, Indiana.
Review the pavement design information for Utah. This information, a streamlined version
of the entire UDOT pavement design policy, and the layer coefficient (a) tables used by the
Utah DOT, can be found in the Module E, Lesson 1 Exercises PDF document. To download
this document, select the paperclip icon.
Image description: Documents from sample pavement projects, Utah.
Review the pavement design information for Washington. This information, a reduced,
streamlined version of the entire WS DOT pavement design method, and the layer
coefficient (a) tables used by the Washington DOT, can be found in the Module E, Lesson 1
Exercises PDF document. To download this document, select the paperclip icon.
Image description: Documents from sample pavement projects, Washington.
Now let’s answer a series of questions on the information you reviewed related to design
methodology in the example States.
Select the best answer. Based on your review of the information provided, what type of
design methodology do you think Florida DOT is using?
a) AASHTO Guide for Design of Pavement Structures (1993);
b) Mechanistic empirical; or
c) MEPDG along with FDOT pavement historical data and experience used in the
development and validation of the design tables.
The correct answer is a) AASHTO Guide for Design of Pavement Structures (1993).
design methodology do you think Indiana DOT is using?
c) MEPDG along with Indiana DOT pavement historical data and experience used in the
The correct answer is b) Mechanistic empirical.
design methodology do you think Utah DOT is using?
c) MEPDG along with Utah DOT pavement historical data and experience used in the
The correct answer is a) AASHTO Guide for Design of Pavement Structures (1993).
There are many States deriving their own State‐specific data to be used in populating
databases that serve to provide inputs to the Pavement ME Design software and thereby
“calibrating” the designs more to their particular conditions. Let’s take a look at how these
four sample States are doing this.
Select each highlighted State to learn more.
Image description: Map of the United States with the states of Washington, Utah, Indiana,
and Florida highlighted.
In Washington, there is a traffic input catalog with State‐specific traffic reports for different
types of roadways in the State.
Image description: Map of the United States with the state of Washington highlighted.
In Utah, there is a materials catalog which contains frost‐depth data for soils and
aggregates because of the frost‐heave issues experienced in much of the State.
Image description: Map of the United States with the state of Utah highlighted.
In Indiana, there is a materials catalog with State‐specific inputs as well as a traffic input
catalog with traffic data for different types of roadways.
Image description: Map of the United States with the state of Indiana highlighted.
In Florida, there is a materials catalog with State‐specific inputs.
Image description: Map of the United States with the state of Florida highlighted.
Let’s take a moment to consider what is being done in terms of pavement design in your
State. Please answer the questions on the screen in the Module E, Lesson 1 Exercises PDF.
We’ll discuss your answers in the instructor‐led portion of this module.
Once you have completed this exercise, advance to the next screen to continue this lesson.
The ME design procedure accounts for all factors that change with time and that affect
pavement response. The mechanistic empirical design is much more than thickness design.
Structural response models in ME design help determine the critical responses in the
pavement structure. These responses are directly used to calculate damage, which is in
turn related to individual distress modes.
Select the mechanical and empirical analysis icon to take a peek inside the engine.
Image description: Illustration of a rain cloud blocking the sun.
Image description: A highway traffic jam.
Image description: Illustration of cracked soil.
Image description: Illustration of an engine.
What is meant by mechanistic empirical? This means that for the first time the principles of
engineering mechanics will be applied to the pavement design problem on a national basis.
Mechanistic analysis is based on a given material’s response to load and environment,
through the definition and evaluation of its stress and strain relationships. This figure
shows a wheel load being applied to a pavement. The pavement materials will
mechanically respond in three ways:
1. The layers will deform when the load passes over a given point in a pavement
(represented by the deflection symbol, δ);
2. The compressive strain response at the surface of the subgrade soil will indicate how
effective the soil is under the pressure of a heavy vertical load being applied to the system;
and
3. The tensile strain at the bottom of the surface layer (e.g., asphalt or concrete) will
indicate how effective the material’s strength is in reacting to the bending under the
vertical load applied.
Empirical analysis is based on field observations and physical measurements of how the
pavement materials strengthen (as in the case early on for concrete pavements) or
deteriorates over time.
Image description: Illustration of engineering stress/strain relationships.
The structural response model should be able to consider a variety of loading situations
and material properties in predicting the pavement response. Just as with the inputs, the
response models used to analyze pavements can also vary over a wide range of
sophistication. Although a sophisticated pavement analysis model is desirable for accurate
pavement response calculation, the degree of sophistication should always be weighted
against practicality.
Tremendous research has been conducted over the last century in developing suitable
structural response models for pavement analysis. Today, this component of the Pavement
ME Design process is perhaps the best developed. The process includes:
• Traffic loading characteristics, such as average daily truck traffic (ADTT), growth factors,
axle distributions, hourly distribution, wander, and tire pressures.
• Climate loading, such as through measuring hourly temperatures, rainfall, and the water
table depth.
• Materials properties through measurement of the layer moduli, master curves,
temperature coefficients, and aging of the properties over time and temperature
conditions.
• Pavement structure characterization, such as the thickness and existing pavement
structural conditions.
Image description: Illustration of a rain cloud blocking the sun.
Image description: Illustration of a block of road above two layers of sub‐base.
Image description: Illustration of the front of a semi truck.
Image description: Illustration of an engine.
Image description: Graphic of a sound wave.
Image description: Graph of Damage/Time.
Image description: Illustration of cracked soil.
These are some of the factors that affect material properties in Pavement ME Design.
These will be discussed in further detail as part of the instructor‐led part of this module,
but it is important to keep these in mind during this lesson.
The load frequency applied to the pavement will influence the material’s response (such as
elastic, viscoelastoplastic, etc.), particularly if you are considering a low‐volume pavement
exposed to low operational speeds as opposed to a freeway pavement that experiences
high frequency loading (from high operational speeds).
The gradation of a soil or base material is one factor, or the gradation used as part of an
asphalt or concrete pavement mix design, which will impact the material properties.
The level of compaction, constructed properties, consolidation and aging rate are all factors
that come from field exposure but can alter how effective the material responses are over
time.
Finally, the temperature (seasonally, monthly, and hourly) can certainly affect the material’s
strength and resiliency at any given time and traffic load level.
Image description: Interstate with light traffic.
The performance of pavements is highly dependent upon several factors. Pavement ME
software checks a pavement design to see if it will meet determined criteria at a
designated reliability. Understanding the interrelationships of the factors can help be better
prepared to adjust a pavement design as needed to obtain the most successful and
adequate possible design.
Select each factor to learn more.
Image description: Graphic of a flowchart.
Traffic loading has a significant effect on the predicted pavement performance
indicators. Higher truck traffic volumes may result in higher predicted distresses.
Image description: Static plate load testing on geogrid reinforced base layers.
Climate will also affect pavement performance. High temperatures may cause
rutting or permanent deformation in flexible pavements; whereas cold
temperatures may cause cracking in flexible pavement.
Image description: Photo of a stretch of road with very low storm clouds hovering
above.
Material properties, foundation, and existing pavement conditions greatly contribute to
the longevity of pavements.
Design and construction factors, such as pavement thickness, maintenance level, and
quality of construction have a considerable effect on the pavement performance.
Image description: Photo of concrete reinforced using grade 60 bars.
Other factors that affect the predicted distresses are geometric features, such as
the lane width. The width controls the closeness of the edge of the tires traversing
the jointed plain concrete pavement (JPCP) and continually reinforced concrete
pavement (CRCP).
The further a wheel is from the edge of the pavement the lower the fatigue
damage, which affects the amount of transverse cracking. Moving the wheel away
from the edge of the pavement will result in reduced fatigue damage.
Image description: Photo of four‐lane road with a widened slab width (shown as 2
ft. wider than the 12‐ft. lane, so a 14‐ft. slab width) indicated.
Now let’s take a moment to check your knowledge.
True or false? Load frequency and temperature affect material properties.
a) True; or
b) False.
The correct answer is a) True.
The Pavement ME Design analysis consists of three main elements: the selection of a trial
design, the mechanistic calculation of structural responses to traffic and climatic loads, and
the empirical estimation of resulting pavement distress amounts over the target design life
of the pavement.
Once the distress amounts are obtained, the user can compare the trial pavement’s
predicted distresses to the damage threshold set at a defined reliability level and can
decide the best course of action for altering the trial pavement structure for better
performance.
Image description: Arrow moving from left to right on screen.
This flowchart illustrates in details the Pavement ME Design analysis process.
The process begins with gathering of user‐defined inputs including structure, materials,
traffic, and climate.
The user will then select a trial design.
Based on this trial pavement design, the Pavement ME Design software will calculate the
structural responses in terms of stress, strain, and deflection.
The calculated mechanical responses will then be incorporated into the calibrated damage‐
distress models to predict pavement distresses and smoothness at a certain reliability level
over the design life.
At this point, the predicted pavement performance will be verified by comparing the
predicted distresses to the threshold criteria.
If the predicted distresses are less than the threshold criteria and the design life is met,
then the trial design might be considered final.
Otherwise, the trial design will need to be revised and the process will start again.
Image description: Flowchart illustrates in details the Pavement ME Design analysis
process.
Let’s pause for a knowledge check.
Select all that apply. What best describes the MEPDG analysis process?
a) It calculates pavement responses of trail design to compute incremental damage over
time;
b) It addresses mixture durability issues; and/or
c) It is a trial‐and‐error process.
The correct answers are a) It calculates pavement responses of trail design to compute
incremental damage over time, and c) It is a trial‐and‐error process.
As mentioned previously, the Pavement ME Design process is iterative and consists of three
steps:
1. Create a trial design for the project;
2. Run the Pavement ME Design to predict the key distresses and smoothness for the trial
design; and
3. Compare the predicted performance of the trial design against performance criteria. The
user must note whether the predicted performance exceeds or is less than the
performance life set by the user.
If necessary, the user might need to modify the trial design as needed in order to produce a
feasible design that satisfies the performance criteria.
Image description: Interchange on the interstate with the words, “Tools to allow easy
comparison between pavement designs.”
Let’s review an example using the Pavement ME Design software. We’ll explore the
Pavement ME Design analysis process using a JPCP located in Ohio. First, we’ll review the
information that a user inputs into the software.
Select each input to learn more.
The user will input the general information related to a project in the first window that
appears when the software is launched and a new project is selected. The user will start by
selecting the pavement type as well as the design life, pavement construction date, and
traffic opening date. The performance criteria and related reliability are also defined in this
screen.
Recall that this project consists of a JPCP located in Ohio. After selecting a trial design
consisting of 9‐in. PCC surface over a 6‐in. AASHTO classified A‐1‐a base on top of an
AASHTO classified A‐6 subgrade soil, we want to determine if this trial structure performs
within the set threshold criteria over a 30‐year design life.
Image description: Screenshot of the general information screen of the Pavement ME
Design software.
The next step is to input the traffic data. The software requires the annual average daily
truck traffic (AADTT) as an input. The AADTT has a significant effect on the
predicted pavement performance indicators and represents a weighted average
between weekday and weekend truck traffic. AADTT are obtained from weigh‐in‐
motion (WIM) data, automated vehicle counters, or manual traffic counts.
This window is also where other traffic information such as speed, traffic growth,
monthly adjustments, vehicle class distribution, and axle configuration are defined.
Image description: Screenshot of the traffic data screen of the Pavement ME Design
software.
From the traffic input screens a user then proceeds to select the climate station and
climate inputs. This example project is located in Marion, Ohio. In this window is where the
depth to the groundwater table is also defined.
Image description: Screenshot of the climate station and climate screen of the Pavement
ME Design software.
The next step is to input the JPCP design properties, such as the PCC joint spacing, sealant
type, dowel spacing and diameter, and erodibility index.
Image description: Screenshot of the JPCP design properties screen of the Pavement ME
Design software.
Once the JPCP design properties are defined, the next step consists of selecting the
material properties for each pavement layer.
For the surface layer or PCC, the main inputs are layer thickness, unit weight, Poisson’s
ratio, thermal, and mix properties.
For the unbound A‐1‐a base layer the main inputs are layer thickness, Poisson’s ratio,
coefficient of lateral earth pressure, and resilient modulus.
Finally, for the A‐6 subgrade layer the main inputs are Poisson’s ratio, coefficient of lateral
earth pressure, and resilient modulus.
Image 1 description: Screenshot of the PCC inputs screen of the Pavement ME Design
software.
Image 2 description: Screenshot of the unbound A‐1‐a base layer screen of the Pavement
ME Design software and screenshot of the A‐6 subgrade layer screen of the Pavement ME
Design software.
After the user runs the Pavement ME Design software, several outputs will be available for
analysis. Let’s review some of the outputs from the example Ohio pavement scenario.
Select each output to learn more.
Image description: Photo of a local street.
A PDF file summarizing the user‐defined inputs (including design structure and traffic) and
the predicted distresses is automatically generated when the Pavement ME Design
software is run.
Based on the set performance thresholds and reliability, the software will determine if each
criterion was satisfied. As noted in the “Distress Prediction Summary”, the only criterion
that failed in the terminal IRI. The predicted IRI was estimated to be higher than the target
threshold.
Image description: Pavement ME Design PDF file summarizing the user‐defined inputs and
the predicted distresses.
The Pavement ME Design software will also generate plots illustrating the predicted
pavement performance over the design life.
For all plots, the red line represents the performance criteria threshold, the blue dotted
line represents the predicted distress at the specified reliability, and the black dashed line
represents the predicted distress at 50% reliability.
In this case, in the “predicted IRI” plot you can see that the predicted IRI surpassed the
threshold value at approximately 20 years out of the intended 30‐year design life.
Image description: Pavement ME Design PDF screenshot of Predicted IRI graph.
Recently the NCHRP has published the NCHRP Project 1‐47: Sensitivity Evaluation of
MEPDG Performance Prediction. The sensitivity study was conducted on MEPDG
software version 1.0 for flexible pavements, JPCP, and CRCP.
This report documents and presents the results of a study to evaluate the sensitivity
of pavement performance predicted by the Mechanistic Empirical Pavement Design
Guide to the values of the design inputs. Global sensitivity analyses were performed
for five pavement types under five climate conditions and three traffic levels.
Design inputs evaluated in the analyses included traffic volume, layer thicknesses,
material properties (e.g., stiffness, strength, HMA and PCC mixture characteristics,
subgrade type), groundwater depth, geometric parameters (e.g., lane width), and
others.
The NCHRP Project 1‐47 sensitivity study is available at:
http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP01‐47_FR.pdf.
Image description: Screenshot of Research Results Digest 372.
A normalized sensitivity index (NSI) was adopted as the quantitative metric. The NSI
is defined as the percentage change of predicted distress relative to its design limit
caused by a given percentage change in the design input.
The “mean plus/minus two standard deviations” value of NSI was adopted as the
best and most robust ranking measure because it incorporates both the mean
sensitivity and the variability of sensitivity across the problem domain. Based on
this, design input sensitivity categories were defined.
Select each sensitivity category to learn more.
Very sensitive (VS): those that have very significant effect on one or more outputs; it
roughly corresponds to a mean‐normalized sensitivity index > 0.5.
Sensitive (S): those that have a moderate effect on one or more outputs; it roughly
corresponds to a mean‐normalized sensitivity index in the range of 0.1–0.5.
Not sensitive (NS): those that have only minor effect on one or more outputs; it roughly
corresponds to a mean‐normalized sensitivity index < 0.1.
Let’s take a few moments for you to consider some critical thinking questions related to
pavement performance. In the Module E, Lesson 1 Exercises PDF, complete Exercise 3.
When you have finished, return to the training to advance to the next screen and review
the questions you have answered.
The factors affecting JCPC performance include traffic, climate, design feature,
layer/general, layer/PCC, layer/subbase (base), layer/subgrade, and JPCP/HMA (rehab).
Select each factor to view its sensitivity rating from the triage summary for JCPC pavements
resulting from the NCHRP 1‐47 research project.
Image description: JPCP inputs graph.
The factors affecting HMA, or flexible, pavement performance include general, traffic,
climate, layer/general, layer/HMA, layer/subbase (base), layer/subgrade, HMA/HMA
(rehab), and HMA/JCPC (rehab).
Select each factor to view its sensitivity rating from the triage summary for flexible
pavements resulting from the NCHRP 1‐47 research project.
Image description: HMA pavement Inputs graph.
The factors affecting CRCP performance include traffic, climate, design feature,
layer/general, layer/PCC, layer/subbase (base), and layer/subgrade.
Select each factor to view its sensitivity rating from the triage summary for flexible
pavements resulting from the NCHRP 1‐47 research project.
Image description: CRCP inputs graph.
Let’s review the most sensitive parameters for each pavement type.
Select each type of pavement to learn more.
Image description: Diagrams of typical flexible and rigid pavements.
According to the findings from the NCHRP 1‐47 project, the most sensitive parameters for
flexible pavements are:
• Traffic volume;
• HMA thickness; and
• Surface shortwave absorptivity.
The most sensitive parameters for JPCP are:
• Initial two‐way AADTT;
• Climate;
• Permanent curl/warp effective temperature difference;
• Joint spacing;
• Edge support;
• Surface shortwave absorptivity;
• PCC layer thickness;
• Coefficient of thermal expansion; and
• Granular base material/stiffness.
The most sensitive parameters for CRCP are:
• Initial two‐way AADTT;
• Coefficient of thermal expansion;
• PCC layer thickness;
• 28‐day PCC modulus of rupture;
• Percent steel; and
• Bar diameter.
Select the best answer. What do engineers need to adjust if the predicted distress is
significantly less than the threshold criteria?
a) Reliability level;
b) Trial design; or
c) Distress criteria thresholds.
The correct answer is b) Trial design.
True or false? The final step in the iterative design process with Pavement ME Design is
comparing the predicted performance of the trial design against the performance criteria.
a) True; or
b) False.
The Pavement ME Design software can be used to analyze the expected
performance of new pavements or rehabilitated pavements. New pavements can be
flexible pavements, jointed plain concrete pavements, or continuously reinforced
concrete pavements.
Rehabilitated pavements can feature either asphalt concrete (AC) or Portland
cement concrete (PCC) overlays.
AC overlays can be paved on existing AC or JPCP overlays.
PCC overlays can be constructed on top of JPCP or CRCP.
Image description: Diagram displaying pavement types analyzed in Pavement ME
Design.
Flexible and rigid pavements are subjected to a number of distresses throughout their
design life. The ability to predict key distresses makes it possible to control them to
reasonable levels over the life of the pavement. This aspect makes the MEPDG extremely
valuable.
Select each of the distresses that pavements may experience to learn more.
Image description: Photo of a crack running across a road.
Rutting is a surface depression in the wheel path resulting from plastic or permanent
deformation in each pavement layer. It could be a result of structural consolidation
rutting (which initiates at the surface of the subgrade layer or unbound base layer)
or by result of instability rutting (which initiates at the surface of the asphalt
mixture layer).
Image description: Photo of a stretch of road with rutting.
Transverse cracking is a non‐wheel load related cracking that is predominately
perpendicular to the pavement centerline and caused by low temperatures or
thermal cycling.
Image description: Photo of a stretch of road showing transverse cracking.
Alligator cracking is a form of fatigue or wheel load‐related bottom‐up cracking and is
defined as a series of interconnected cracks (characteristically with an “alligator”
pattern) that initiate at the bottom of the HMA layers.
Image description: Close up photo of alligator cracking.
Image description: Photo of alligator cracking on the road.
Longitudinal cracking is a form of fatigue or wheel load‐related top‐down cracking
that occurs within the wheel path and is defined as cracks predominantly parallel to
the pavement centerline.
Image description: Collage of photos of longitudinal cracking.
IRI is a parameter used to define pavement smoothness and derived from the
simulation of a “quarter‐car” traveling along the longitudinal profile of the road. IRI
is calculated from the mean of the longitudinal profiles in each wheel path.
Image description: Collage of photos of roadway smoothness.
Now let’s review some of the distresses that rigid pavements are subjected to throughout
their design life.
Select each of the distresses that rigid pavements may experience to learn more.
Image description: Photo of a stretch of road with pavement distresses.
Mean transverse joint faulting is the differential elevation across the joint measured
approximately 1 ft. from the slab’s edge (longitudinal joint for a conventional lane
width), or from the right‐most lane paint stripe for a widened slab.
Image description: Photo of mean transverse joint faulting.
Bottom‐up transverse cracking occurs when the truck axles are near the longitudinal
edge of the slab, midway between the transverse joints, and a critical tensile
bending stress occurs at the bottom of the slab under the wheel load.
Image description: Photo of bottom‐up transverse cracking.
Top‐down transverse cracking is a fatigue damage at the top of the slab, which
eventually results in a transverse or diagonal crack that is initiated on the surface of
the pavement. It occurs when a repeated loading by heavy truck tractors with
certain axle spacing is applied when the pavement is exposed to high negative
temperature gradients.
Image description: Photo of top‐down transverse cracking.
The load transfer efficiency (LTE) is defined as the ratio of deflection of the
unloaded side to the loaded side of the joint multiplied by 100. When crack LTE is
reduced the potential for punch outs to develop increases greatly.
In joint load transfer deterioration, a transverse crack happens at a short distance
from a transverse joint, usually at the end of joint load transfer dowels.
The joint deterioration is due to extensive corrosion or misalignment of dowel
system.
Image description: Screenshot of the pavementinteractive.org article regarding joint load
transfer deterioration.
Hyperlink description: http://www.pavementinteractive.org/article/load‐transfer/
In CRCP pavements, a punch out occurs when truck axles pass near the longitudinal
edge of the slab between two closely‐spaced transverse cracks, and a high‐tensile
stress occurs transversely across the pavement at the top of the slab within a
distance of 48 in. from the edge.
Image description: Collage of photos of CRCP punch outs.
Just as with flexible pavement distresses, IRI is a parameter used to define pavement
smoothness and derived from the simulation of a “quarter‐car” traveling along the
longitudinal profile of the road. IRI is calculated from the mean of the longitudinal
profiles in each wheel path.
Image description: Collage of photos of roadway smoothness.
This table summarizes the recommended performance threshold values for Pavement ME
Design. The threshold values change at the end of the design life depending on the
functional classification of the roadway. These are design target thresholds and do not
consider the scheduling of preventative or routine maintenance activities.
Design performance criteria and design reliability greatly affect the determination
of an adequately performing pavement. Reliability is the probability that the
predicted performance indicator of the trial design will not exceed the design
criteria within the design‐analysis period. The design criteria and design reliability
levels could be selected in balance with each other. A low level of distress should
not be selected in conjunction with a high level of reliability because this may make
it impossible or costly to obtain an adequate design. A designer may specify the
desired level of reliability for each distress type and smoothness.
The next series of slides are included to provide you with a demonstration of the Pavement
ME Design software. These are built around a jointed plain rigid pavement design in Ohio,
which is based on the LTPP Section Site 39‐0263.
The purpose of this activity is to show how a selected number of factors affect the
predicted pavement performance. We want to observe what happens to the predicted
pavement performance when few parameters are changed.
Image description: PvD AASHTOWare Pavement ME Design logo banner.
We will now change one parameter at a time from the 9‐inch baseline rigid pavement
thickness and then compare the predicted performance of the modified structure to the
predicted performance of the baseline.
This slide shows some of the basic inputs for the design of the example rigid pavement.
Image description: Pavement ME Example for Rigid Pavement in Ohio: Baseline
Information.
This slide shows the predicted performance of the original pavement in the example. The
IRI was predicted to fail in terms of pavement smoothness prior to its target value of 172
inches per mile at the 90% reliability. The design performed sufficiently in terms of faulting
at the joints and transverse cracking.
Assuming the Ohio DOT desired to optimize the design for costs, let’s see what would
happen if we attempted a design that is 1.5” thinner, so a 7.5‐inch rigid pavement instead.
Image description: Pavement ME Output example for Rigid Pavement in Ohio:
Baseline Information.
We will now change one parameter at a time from the 9‐inch baseline rigid pavement
thickness and then compare the predicted performance of the modified structure to the
predicted performance of the baseline.
When we decrease the PCC layer thickness from 9 in. to 7.5 in., we notice an increase in all
distresses and smoothness.
As can be seen in the plots at the bottom, the IRI fails after approximately 8 years, which is
12 years before this failure was observed in the baseline structure.
In addition, cracking and faulting were observed to exceed their respective threshold values
after approximately 18 and 9 years, respectively.
In comparison, the baseline structure doesn’t experience any cracking and faulting within
the 30 years of design life.
When we decrease the dowel diameter from 1.5 in. to 1.25 in., we notice an increase in all
distresses and smoothness.
The IRI fails after approximately 15 years. The IRI of the baseline structure surpasses the
threshold after approximately 20 years.
The cracking and faulting fail after approximately 18 and 24 years, respectively.
On the other hand, the baseline structure doesn’t experience any cracking and faulting
within the 30 years of design life.
In this case, we changed the subgrade type from A‐7‐6 (with a resilient modulus of 13,000
psi) to A‐6 (with a resilient modulus of 14,000 psi), through the application of a cut/fill with
more superior soil.
When we increase the subgrade strength, we notice a slight decrease in cracking, IRI, and
faulting compared to the baseline structure.
This table summarizes the results from the four analyzed pavement structures.
As you can see, changing subgrade strength, dowel size, and PCC thickness affected the
predicted pavement performance of this particular pavement structure.
Increasing subgrade stiffness resulted in slightly lower terminal IRI, joint faulting, and
transverse cracking as compared to the predicted distresses of the baseline pavement
structure.
Decreasing the dowel bar diameter size resulted in higher predicted distresses and
smoothness (compared to the baseline pavement structure).
Decreasing the PCC thickness resulted in even higher levels of distresses predicted and a
reduction in smoothness.
Let’s pause to check your understanding of performance parameters and the Pavement ME
Design software.
Select all that apply. Which performance parameters are not analyzed by Pavement ME
Design software?
a) Pavement deformation;
b) Surface disintegration distresses;
c) Faulting;
d) Skid resistance;
e) Fatigue cracking; or
f) IRI.
The correct answers are b) Surface disintegration distresses and d) Skid resistance.
The concept to be taken from this schematic is that one can use the performance
prediction models (with adjustments) from the Pavement ME Design software to assist in
the design of paving mixtures and possibly tie performance into the mixture design
process. However, you cannot tie the design of paving mixtures into the Pavement ME
Design trial pavement process. The software tool can be used to provide a different
perspective for materials engineers in predicting the quality of materials in a
performance‐based sense, prior to final mixture design occurring.
The procedures used in the analysis are based upon fundamental engineering properties
and behavior based upon ME principles to predict specific pavement distress. The models
can also be used as a forensic tool for evaluating changes in specification and/or legal load
limits that we are periodically asked by legislators and lobbyists to assess.
However, when it comes to materials design there are some specifics with the capabilities
of the Pavement ME Design software to be aware of. In order to help participants visualize
the use of Pavement ME Design software prediction models for materials design, two
examples or “proofs of concept” are provided: one for rigid pavements and one for flexible
pavements. The concept to be proved is that sensitivity to the pavement performance is
not stemming from the materials or mixture design.
For example, in rigid pavements, there is a section of the design software that allows for
the input of the concrete’s coefficient of thermal expansion (or CTE). However, the value
that is input by the user is an average CTE or in other words, a value that is averaged over
all sublots and lots that would be placed in a given pavement project.
Image description: Photo showing paving operation on Interstate 80 with dowel bar
baskets.
Another example comes from looking at flexible pavements, in which the dynamic modulus
stiffness, or E‐star, can be input for the Level 1 section of the asphalt layer’s definition.
However, the value that is input by the user is an average E‐star for each temperature and
load frequency combination. In addition, it is a value that is averaged over all sublots and
lots that would be paved in a given pavement project.
Image description: Photo of flexible pavement rehabilitation.
This schematic presents the process of altering materials properties in order to take a trial
pavement design that fails to one that meets all performance criteria and can advance to
the final design and construction phase.
At the top you start with a trial pavement design and trial mixture and material properties
for an example flexible pavement.
The analysis is then run in the Pavement ME Design software.
Results from the distress prediction models show that the design fails in multiple distress
modes, for example rutting in both the asphalt surface and in the subgrade.
Therefore, the designer elects to change the subgrade properties by increasing its stiffness
through stabilization of the subgrade with lime.
The analysis is then run again and the results show that now the pavement only exceeds
the threshold value for surface rutting.
The designer then elects to attempt to stiffen the surface layer by bumping up the binder
by one PG grade.
The analysis is run one more time.
And finally in this case, the pavement structure now passes all for the given traffic and
environmental loads.
Image description: Schematic showing the process of altering materials properties in order
to take a trial pavement design that fails to one that meets all performance criteria and can
advance to the final design and construction phase.
There are three recent tools developed either through the National Cooperative Highway
Research Program and the Federal Highway Administration that have been based on using
the Pavement ME Design performance models to provide a different perspective on how
materials engineers can predict the quality of materials in a performance‐based sense prior
to final mixture design occurring.
Select each tool to learn more about it.
In the NCHRP 9‐22 project, entitled the concept of quality‐related standard specifications
(QRSS) for flexible pavements, was enhanced through the development of a stand‐alone
program that calculates the predicted performance of an HMA pavement from the
volumetric and materials properties of the as‐designed HMA and compares it with that of
the as‐built pavement calculated from the contractor's lot or sublot quality control data.
The QRSS accounts for the climate, traffic, pavement structure, and desired or expected
service life.
Reference: Moulthrop, J., James, M., Witczak, M., Jeong, M., McCarthy, L., and D.
Mensching, Evaluation of the Quality Related Specification Software (QRSS) Version 1.0.
NCHRP Project No. 9‐22A Final Report, Transportation Research Board, National Research
Council, 2012, http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP09‐22A_FR.pdf.
As a follow up to the previously discussed project, a second effort entitled “Evaluation of
the Quality‐Related Specification Software (QRSS) Version 1.0” was conducted (known as
NCHRP Project 9‐22A) to evaluate the QRSS Version 1.0 with quality assurance (QA) data
obtained from State DOTs in three parts of the nation (southern Midwest, west, and
northeast). In this project, the QRSS results that were obtained were compared with
various levels and types of measured mix volumetric and performance data.
Reference: National Cooperative Highway Research Program (NCHRP), Quality‐Related
Specification Software (QRSS), 2010,
http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=958.
For rigid pavements, the Federal Highway Administration Office of Research and
Technology Development has developed the PaveSpec software tool. This tool is
anticipated for release in Summer 2014 and incorporates the distress prediction models
from the Pavement ME Design into JPCP performance‐related specifications (PRS). The
concept is that this tool can be applied in the development of a plan for the State‐wide
implementation of PRS.
Reference: Federal Highway Administration PaveSpec 3.0 (need to update this Reference
once PaveSpec 4.0 is released in Summer 2014),
http://www.fhwa.dot.gov/publications/research/infrastructure/pavements/pccp/pavespec
/
Image description: PaveSpec software screenshot.
It is important to make a clear distinction between the design and the analysis purposes
when using the AASHTO Pavement ME Design software. When the software is used as a
pavement design tool, the trial pavement layers have to be practical and based on
constructability. Practitioners will design the pavement layers with a realistic approach at
reasonable accuracy levels by implementing strategic thickness iterations to the process. A
designer should design what the contractors can construct, especially the pavement layer
thickness.
When the software is used for pavement analysis purposes and academic exercises, the
layers thicknesses are not governed by constructability. This tool can be used for forensic or
exploratory analysis purposes. As a pavement analysis tool, the AASHTO Pavement ME
Design software will predict pavement distresses predictions for either flexible or rigid
pavements.
For materials engineers, one of the major benefits of the MEPDG will be its ability to tie all
the design elements and stakeholders together in one integrated package. For example, in
earlier guides HMAC mix design and structural capacity were only indirectly related.
Through the use of the mechanistic approach, that connection becomes direct and directly
related to performance of the pavement.
The most important is the way we design pavement now is a "more integrated" approach
that is based on performance criteria. We are not designing based on the thickness of the
upper layers anymore. We are designing the whole system.
The ME analysis procedures allows an agency to compare the expected performance of
their as‐constructed pavements to the expected performance under assumed design
conditions. This enables an agency to rationally set the pay factors on their construction
specifications and to determine the effectiveness of their warranty and performance‐
related specifications.
Image description: Photo of two men in suits wearing hard hats observing a construction
site.
Image description: Photo of Heavy equipment on a construction site.
As previously mentioned, there are five sample pavement projects that we will use to help
illustrate different MEPDG concepts. These five pavement projects will serve as the
recurring theme throughout the instructor‐led session. They include a new flexible
pavement in an area with low truck traffic volumes, a new flexible pavement in an area
with high truck traffic volumes, a new jointed plain concrete pavement in an area with high
truck traffic volumes (a CRCP study will also be included in the Lesson 5 of the instructor‐
led training session), a flexible pavement overlay over an existing flexible pavement in an
area with high truck traffic volumes, and a flexible pavement overlay over an existing JPCP
pavement in an area with high truck traffic volumes (a CRCP study will also be included in
the Lesson 5 of the instructor‐led training session).
These sample pavement projects are being used to illustrate the important factors to
consider as part of forming a pavement design with the Pavement ME Design, with a
particular emphasis on materials.
The details of the pavement sections that are used were extracted from real long‐term
pavement performance sections that are in the same climatic region as the four States
used for illustration purposes in this training.
Select each example project for a project overview.
New HMA (Low Traffic Volume): The first sample pavement section is a new flexible
pavement in an area with low truck traffic volumes and based in the State of Washington.
Image description: Sample of pavement design.
New HMA (High Traffic Volume): The second sample pavement section is a new flexible
pavement in an area with high truck traffic volumes and based in the State of Florida.
New JPCP (High Traffic Volume): The third sample pavement section is a new jointed plain
concrete pavement in an area with high truck traffic volumes and based in the State of
Utah. It should be noted that in Lesson 5, a CRCP pavement will also be substituted in the
place of the JPCP to demonstrate the design properties and performance predictions for
CRCP pavements.
Rehab HMA over HMA (High Traffic Volume): The fourth sample pavement section is a
flexible pavement overlay over an existing flexible pavement in an area with high truck
traffic volumes and based in the State of Washington.
Rehab HMA over JPCP (High Traffic Volume): The fifth sample pavement section is a flexible
pavement overlay over an existing JPCP pavement in an area with high truck traffic volumes
and based in the State of Indiana. It should be noted that in Lesson 5, a CRCP pavement will
also be substituted in the place of the JPCP to demonstrate the design properties and
performance predictions for asphalt overlays of CRCP pavements.
Before you attend the instructor‐led portion of this module, complete Exercise 4: Review
and Comparison. In this exercise, you will search for, identify, and consult the pavement
design manual, asphalt, and concrete mixture specifications, and other information as
applicable one (in this case, review those of Indiana DOT only) of the four States, and your
own local area.
Use the exercises that were provided from the paperclip icon. There you will find the steps
to complete this exercise.
Again, make sure you have completed this exercise in advance of Lesson 2. Be ready to
discuss your findings with other participants and the instructor when you attend the
instructor‐led portion of this module.
Image description: Sample of the pavement design and materials specifications handout.
Before you attend the instructor‐led portion of this module, complete Exercise 5: Case
Study. Read the handout of a case study of a pavement design planned near Johnstown,
Pennsylvania and complete the related questions. Identify the important considerations
for pavement design and comment on the traffic levels, climate considerations, and
types of critical materials properties that should be considered in improving this
fictional pavement section’s performance.
Use the exercises that were provided from the paperclip icon. There you will find the steps
to complete this exercise.
You should complete this exercise in advance of the instructor‐led session of this
module.
Image description: Collage of images of predicted faulting, predicted cracking, and
predicted IRI for the Johnstown, PA case study.
Now we’ll answer one last knowledge check before completing this lesson.
True or false? Material engineers might use the Pavement ME Design software tool to
predict the quality of materials in a pavement performance‐based sense.
a) True; or
b) False.
You have completed Module E, Lesson 1: Introduction. You are now able to:
• List the basic factors required in designing a pavement structure;
• Compare mixture design and pavement design related to materials in terms of predicted
pavement performance;
• Distinguish between empirical and mechanistic empirical design methods;
• List the factors that may affect the predicted performance of pavements in MEPDG;
• Describe the steps of a MEPDG analysis;
• Describe the iterative design process used with Pavement ME Design software;
• Describe the performance parameters analyzed by Pavement ME Design software;
• Explain how pavement prediction models from MEPDG are being incorporated into the
evaluation of asphalt and concrete paving mixtures; and
• Using an example, identify important considerations for pavement design.
Close this lesson, and return to the module curriculum. To close this window, select the “X”

in the upper right‐hand corner of your screen.

Lesson01 WBT

Uploaded by

Copyright:

Available Formats

Lesson01 WBT

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

Lesson01 WBT

Uploaded by

Copyright:

Available Formats

Welcome

A printer-friendly version of the lesson materials can be downloaded by selecting the

Image description: People sitting in front of computers in a training setting.

This lesson will take approximately 180 minutes to complete.

Drainage is accounted for in a general sense from the EICM by the inclusion of an

Image description: Graph demonstrating drawbacks to using the AASHO Road Test as a

Recall the empirical design (such as the AASHTO 1993 version) is based on statistical

Finally, a mechanistic design would include mechanistic‐based pavement performance

Close this lesson, and return to the module curriculum. To close this window, select the “X”

You might also like