School of Earth Sciences and Engineering

CIVE 416

Highway Engineering & Pavement Design: AASHTO MEPDG ( USA)

Group members

Names Surname Student ID

Jeremia Ntwayamodimo 22 000 268
Levite Mokoka 18 001 103
Tebogo L. Moalosi 22 000 390

Objectives
 Discusses the details (step by step process) of your selected method

 Identify key parameters and analysis involved

 The applicability of your method (inputs, outputs,

strengths/weaknesses)

 Identify the advantages and disadvantages of your selected method

INTRODUCTION

Pavement engineering has experienced a major paradigm change, moving

away from conventional empirical design techniques and towards more
advanced and predictive strategies. The shortcomings of earlier guides,
like the 1993 AASHTO Empirical Guide, which relied on scant
observational data from a single, historic road test and found it difficult to
appropriately account for present-day traffic loads, novel materials, and
the significant influence of changing climatic conditions, have contributed
to this evolution. The result of this change is the Mechanistic-Empirical
Pavement Design Guide (MEPDG), created under the National Cooperative
Highway Research Program (NCHRP) Project 1-37A. Approved by the
American Association of State Highway and Transportation Officials
(AASHTO), it is the most recent accepted standard for pavement design in
the United States.

The fundamental idea of the MEPDG is its main innovation: the

combination of empirical calibration and mechanistic modelling. The
"mechanistic" component, as the name suggests, uses theoretical models
to calculate the physical pavement responses to environmental and traffic
loads, including stresses, strains, and deflections. Models that forecast
distresses and smoothness from these computed responses are then
calibrated by the "empirical" component using observed performance
data from actual pavements. This contrasts sharply with purely empirical
approaches, which compute a necessary layer thickness directly using
equations derived from regression models of historical performance. (R.L.
Baus, 2010) .

One of the MEPDG's unique characteristics, and the primary source of its
superior predictive power, is its fundamental dependence on a large, high-
fidelity dataset covering four crucial areas: weather, traffic patterns,
material composition, and past pavement performance. This need for
thorough input is the fundamental, enabling framework that makes the
method's analytical depth possible, not just a logistical challenge.
Agencies like the South Carolina Department of Transportation (SCDOT)
serve as prime examples of how the implementation process turns into a
coordinated effort in data synthesis, where various information streams
are combined into a single model. This method is formally implemented
through the AASHTOWare Pavement ME Design® software, which has
become the standard platform for its application across U.S. state
DOTs. This process involves harvesting material specifications and mix
designs from materials research offices, compiling complex traffic spectra
including vehicle classification distributions, axle load distributions, and
growth factors—from traffic engineering divisions and networks of
Automatic Traffic Recorders (ATRs) and Weigh-in-Motion (WIM) stations,
and integrating historical performance metrics from pavement
management systems.

The main benefits of the MEPDG are unlocked by this process, even
though it requires a much larger initial investment in data management
and acquisition than empirical approaches. The ability of the methodology
to use hierarchical data levels, ranging from accurately measured Level 1
inputs to estimated Level 3 defaults, guarantees that it can still function in
the absence of perfect data, albeit with corresponding adjustments in
predictive certainty. The design process is changed from being based on
broad regional generalisations to a project-specific engineering analysis
through the strategic incorporation of localised data. The MEPDG can
simulate the intricate, time-dependent relationship between a particular
pavement structure, the environmental conditions of its location, and the
precise traffic loads it bears by processing this data using its integrated
climate model and mechanistic algorithms. Consequently, this data-driven
approach yields a highly calibrated and optimized design outcome,
enhancing long-term performance, improving resource allocation by
avoiding both over-design and under-design, and supporting compliance
with U.S. federal and state-level performance-based infrastructure
planning mandates. Ultimately, it provides a more reliable, cost-effective,
and sustainable pavement infrastructure asset tailored to its precise
operational context. (Md Mostaqur Rahman, 2017)
Step-by-Step Design Procedure in AASHTO Mechanistic–Empirical
Pavement Design Guide (MEPDG)
This report presents a detailed, step-by-step description of the design
procedure outlined in the AASHTO Mechanistic–Empirical Pavement
Design Guide (MEPDG), widely implemented in the U.S. through the
Pavement ME Design software. Each step highlights the essential
activities, input requirements, and outcomes as defined in the MEPDG
Manual of Practice and associated AASHTO guidelines.

Step 1: Define Project Information

· Establish the basic project context: location, roadway classification,
design life (typically 20–50 years), traffic level, and functional
classification.

· Define design reliability (commonly 90–95%) and acceptable levels of

risk.

· Select the general pavement type (new, rehabilitation, overlay).

Step 2: Establish Traffic Inputs

· Gather axle load spectra from weigh-in-motion (WIM) stations or regional

defaults.

· Define annual average daily truck traffic (AADTT), growth factors, traffic
wander, lane distribution, and truck percentages.

· Assign traffic input level (Level 1: site-specific, Level 2: regional, Level 3:

defaults).

Step 3: Define Climate and Environmental Data

· Use Integrated Climatic Model (ICM) embedded in Pavement ME to

simulate temperature, moisture, frost, and freeze-thaw cycles.

· Input local weather station data (precipitation, air temperature, wind,

solar radiation).

· Consider groundwater table depth and seasonal variations.

Step 4: Characterize Subgrade and Foundation

· Determine subgrade type and strength (resilient modulus, MR) from
laboratory or field testing.

· Identify bedrock depth, seasonal variations, and soil-water characteristic

curves.

· Define unbound layer properties for base and subbase (gradation,

strength, moisture sensitivity).

Step 5: Characterize Pavement Materials

· Input material properties depending on pavement type:

· Asphalt concrete: Dynamic modulus (|E*|), Poisson’s ratio, thermal

cracking properties.

· Portland cement concrete: Modulus of rupture, elastic modulus,

coefficient of thermal expansion.

· Unbound materials: Resilient modulus, nonlinear behavior models.

· Select Level 1–3 inputs based on data availability.

Step 6: Construct Initial Pavement Structure

· Define trial layer thicknesses (surface, base, subbase).

· Enter sequence of construction for each pavement layer.

· Assign material input properties to each layer.

Step 7: Perform Structural Response Analysis

· Use mechanistic models to simulate stresses, strains, and deflections in

pavement layers.

· Analysis performed at discrete time increments (hourly/daily) across

design life.

· Output used to estimate fatigue, rutting, and cracking potential.

Step 8: Predict Pavement Distress and Performance

· Employ calibrated empirical transfer functions (from LTPP data) to

convert mechanistic responses into predicted distresses:

· Asphalt: rutting, fatigue cracking, thermal cracking.

· Concrete: joint faulting, transverse cracking, slab cracking.

· Smoothness: International Roughness Index (IRI) over time.

Step 9: Compare Results Against Performance Criteria

· Compare predicted distresses and smoothness against agency

thresholds:

· IRI ≤ 172 in/mi (commonly used).

· Asphalt rutting ≤ 0.75 in.

· PCC faulting ≤ 0.12 in.

· Evaluate results under reliability criteria (e.g., 90% reliability means

distress thresholds not exceeded in 9 of 10 simulated cases).

Step 10: Iteration and Optimization of Design

· If design fails to meet thresholds, adjust layer thickness, materials, or

reliability level.

· Re-run simulation until all criteria are satisfied.

· Consider cost-effectiveness and constructability in selecting final design.

Step 11: Document the Design

· Prepare a comprehensive design report including:

· Project details and assumptions.

· Input data (traffic, climate, materials).

· Performance predictions and reliability checks.

· Final pavement structure.

· Submit documentation for review and approval by agency or owner.

Key Parameters & Analysis Involved

The MEPDG works by taking in a set of key parameters that describe the project, the traffic,
the environment, and the materials. At the project level, you define things like the location,
the design life of the pavement, the reliability level, and the type of pavement structure. For
traffic, you use data such as axle load spectra, truck volumes, growth rates, and how trucks
are distributed across lanes. Climate inputs come from the Integrated Climatic Model, which
uses local weather data to simulate temperature changes, precipitation, freeze–thaw cycles,
and groundwater effects.

The foundation and subgrade are described using tests for resilient modulus and soil-water
properties, while materials are characterized using values such as dynamic modulus for
asphalt, modulus of rupture for concrete, and resilient modulus for unbound layers. The
analysis process then uses mechanistic models to calculate stresses and strains inside the
pavement and links these to predicted distresses like rutting, cracking, faulting, and
roughness over time

Application Of AASHTO MEPDG

In practice, the method takes inputs such as traffic spectra, climate data, subgrade properties,
and detailed material information. From these, it produces outputs that show how the
pavement will perform, including predictions of distresses, recommended layer thicknesses,
and the reliability of the design. One of its strengths is that it produces more accurate and
project-specific results than older empirical methods. It also gives flexibility by allowing
engineers to work with different levels of data quality, from detailed site-specific tests to
default values.

Another strength is that it helps balance costs by reducing the chances of over-designing or
under-designing the pavement. On the downside, it requires a lot of high-quality input data,
which can be costly and time-consuming to collect. It also needs specialized software and
expertise, and the results are only reliable if local calibration is done properly to reflect actual
field conditions
Key References:

- AASHTO (2007). Mechanistic–Empirical Pavement Design Guide: A

Manual of Practice.

- AASHTOWare Pavement ME Design User Manual.

- NCHRP 1-37A, 1-40D, 1-40B project reports.

- FHWA Long-Term Pavement Performance (LTPP) Program data.