Structural Design of Flexible Highway Pavements

Basis for design

A flexible pavement is a structure composed of one or more layers of bound or unbound materials with or without a bituminous surfacing.

Since the intensity of stresses from traffic loads diminishes with depth the conventional form of construction for a flexible pavement is a layered system in which the quality of the materials decreases progressively from the surface layer towards the sub-grade.

The basis of the design of flexible highway pavements is to provide:

  • Adequate total thickness of materials over the sub-grade to prevent the sub-grade from failing in compression
  • Sufficient strength within each layer to resist stresses induced by traffic loading and
  • Protection against volume changes in the sub-grade

Design parameters

These are factors that affect the design and determine the thickness of the pavement structure required for supporting the traffic loading without unacceptable deterioration in structural strength,  surface and riding quality over the design period. 

These factors are:

  • Traffic
  • Sub-grade strength
  • Pavement material properties

Traffic

Traffic on highway facilities varies in the number of vehicles and vehicle types. Vehicles may also vary in the number and configuration of axles and the magnitude of the loads on the axles.

Each of the different types of vehicles or categories of vehicles such as private cars, buses, rigid trucks, and articulated trucks, etc. within a given traffic stream is weighted differently in terms of the structural damage they cause to a  pavement.

For the purposes of the structural design of flexible highway pavements, only commercial vehicles (vehicles whose un-laden weight is 1500kg or more) and the magnitude of the loads on their axles are of interest.

Vehicles with un-laden weights below 1500kg cause no significant structural damage to pavements and are, therefore, ignored in routine traffic analysis for pavement design.

The figure below provides the different types of commercial vehicle axle configurations that are likely to be encountered on any roadway.

The volume and nature of commercial vehicles using or expected to use a highway facility may be expressed in a number of ways depending on the level of detail required or the details of traffic flow and vehicles that are available. 

Traffic data on commercial vehicles may be expressed in one of the following forms in order of increasing level of detail or accuracy:

  • number of commercial vehicles
  • number of commercial vehicles by type
  • number of commercial  vehicle by axle configuration
  • number of commercial axles by type and load on the axle
Axle configuration

Expressing design traffic in terms of the number of commercial axles by type and load on the axle allows the most reliable estimates of the damaging effect of traffic on pavements to be made.

Except in axle load studies, such a form of data is of course impractical to collect in routine traffic surveys for design purposes.

On the other hand, design traffic data in terms of the number of commercial vehicle types are easy to collect and presupposes that there is adequate information or data on the damaging potential of the different types of commercial vehicles to permit the total effect of the anticipated traffic on the pavement in terms of the number of standard axles to be evaluated.

Such data will exist in the form of average load equivalency factors for commercial vehicles in general, for a given class of road within a given part of the road network.

Load equivalency factors

An axle load equivalency factor gives the quantitative measure of the structural damage potential of a commercial axle relative to the damaging potential of the standard 80kN single axle.

The standard way of taking into account the effect on the pavement’s structure of any combination of axle types and magnitude of axle loads is to equate the loads to the number of 80 kN (8.2 tonnes or 18,000 lb) single axle loads required to produce an equivalent effect.

A simplified approach to expressing the cumulative effects of traffic loads in terms of the equivalent standard 80 kN single axle loads (ESAL) is to base all conversions on truck factors.

A truck factor is the damage per pass of a commercial vehicle relative to the damage per pass of the standard single axle.

Considerable information about equivalent axle load estimation was made available through the AASHTO (American Association of State Highway and Transportation Officials) Road Test.

A given single axle loaded to L tonnes may be converted to the equivalent standard axle load (ESAL) by the expression;

ESAL

where c is an exponent ranging between 3 and 6, with a value of 4.5 being typical.

In Australia, c=4 is commonly adopted and the standard axle load used as a denominator in the Equation above varies according to whether the axle is single with single wheels, or single with dual-tired wheels, or tandem with dual-tired wheels.

Using c=4.5, the Equation becomes

ESAL 2

In general the ESA for a commercial vehicle with n single axles becomes

ESAL 3

In spite of the differences in commercial vehicles by way axle configuration type and load on the axles, it is possible, for the purposes of analysis to place the vehicles into a finite number of commercial vehicle categories by axle configuration type irrespective of the magnitude of the axle loads.

Consider the case in which a population of N commercial vehicles is placed into M categories. Let the kth category be made up of Mk vehicles and characterized by Nk single axles per vehicle without due regard to axle configuration.

Assuming that the commercial axles could all be considered single, then the ESAL per pass of all the commercial vehicles within this category is given as;

ESAL K

The average ESAL per pass of a commercial vehicle (truck factor-TF) within this category, TFk, therefore, becomes;

Truck factor

In the case where legal load limits on axles are well respected by commercial vehicle operators within the trucking industry, large variations in axle loads within a given category are unlikely and TFk as given by the Equation above would characterize the structural damage potential of each vehicle within this category fairly accurately.

By extending the above concept to cover all categories of commercial vehicles within the traffic stream, an equivalency factor may be obtained that applies to any commercial vehicle irrespective of axle configurations and loads.

Thus based on the commercial vehicle population and categories above, TF applicable to any commercial vehicle within the population is evaluated thus;

TF

The accuracy of the above expression for the truck factor is further enhanced if the evaluation is based on a large sample of data collected over a substantial part of the road network and over a fairly long period of time.

Such equivalency factors are useful parameters for estimating the ESAL for design on the basis of just the estimates of the number of commercial vehicles without necessarily knowing the type or category of commercial vehicle nor the magnitude of the loads on the axles. 

Design ESAL

If the average daily commercial vehicle traffic is made up of nk vehicles of category k, then the initial daily commercial traffic in terms of standard axles loads, ESALo, is obtained thus

ESALo

It is important to know at the onset that for a multi-lane carriageway, traffic on the most heavily-traveled lane is what is used for design.

For a single carriageway with opposing traffic, there must be simultaneous traffic data collection for the two directions of travel in order to establish the lane that carries the higher traffic and hence the design traffic.

Once the number of equivalent standard axles corresponding to the average daily commercial traffic for the design lane has been evaluated, the cumulative number of standard axles for the first year is obtained by aggregating the daily value over 365 days

i.e.

The cumulative number of standard axles, ESALcum,  expected during the design life of the pavement is obtained by taking into account the traffic growth during the design period.

Given a traffic growth rate of r% per annum and a design life of n years, the traffic growth during the design period is taken care of by a parameter called the traffic growth factor, G,  which is given by the following expression.

Traffic Growth Factor

The traffic growth factor is then applied to the first year number of standard axles to obtain the cumulative number of standard axles for design ESALcum , i.e.,

ESALcum =365GESALo                                  

Sub-grade strength

This is a very important factor in pavement design in that it is the sub-grade that the design should seek to protect from shear failure and excessive deflection.

The strength of the sub-grade provides a basis for the selection of the structure for the pavement. The strength of the sub-grade is obtained from sub-grade strength evaluation along the proposed road alignment for a new road or along the existing alignment for an existing road. T

he investigation may be carried out by taking samples from appropriate depths along the road alignment at intervals that may range from 250m to 2km depending on the scale of the road project and the characteristics of the terrain and determining the CBR values in the laboratory in accordance with the appropriate standards.

CBR values are in percentages and it is customary to quote the values as whole numbers.

CBR values may also be evaluated in-situ using the DCP tests or any appropriate tests but such values are affected by the prevailing moisture content and do not represent the appropriate values needed for design.

Due to the natural variability of soil deposits, the strength of the sub-grade will vary along the road alignment.

This means that several and different CBR values will be obtained from the strength investigation along the proposed route.

Depending on the value selected for analysis purposes, the design can represent under-design or over-design. If the average value is selected, about one half of the road is over-designed and one-half under-designed.

If the minimum value is selected most of the road section is over-designed. Some level of risk must be taken in the selection of the CBR for design.

Since the sensitivity of sub-grade strength increases as traffic increases, the design subgrade CBR value is selected on the basis of the anticipated traffic levels; a higher percentile value must be selected for higher traffic to minimize the risk of failure.

When this approach is adopted, then at high traffic levels, the design subgrade CBR approaches the minimum CBR value.

The table below provides a guide to the selection of design sub-grade CBR based on traffic levels.

Sub-grade design CBR limits

Strength of pavement materials

The strength of the materials forming the various pavement layers is generally expressed in terms of their 4-day soaked CBR values.

The higher the CBR value, the higher the strength, and hence the higher the quality of the material. The soaked CBR value is used to characterise the strength of the layer materials because it represents the strength of the materials in the weakest state when totally inundated (flooded).

Where the elastic modulus, E,  of the layer is required, the  value may be estimated from the CBR using the following simple relationship:

E=10CBR MPa                            

where the CBR is in %.

For the Poisson’s ratio,  the values in the Table below provide a useful guide.

Typical Poisson’s ratio for pavement materials

Other parameters related to the strength of the layer materials are:

  • Particle size distribution
  • Atterberg limits.

Specific limits in respective of the CBR, particle size distribution, and Atterberg limits exist that must be satisfied by the layer materials to be acceptable for use in a particular layer.

Design methods

Acceptable methods for the design of flexible highway pavements are many and differ from each other. While some are theoretically based, others are empirically based.  

Each design procedure has its own basis for development and may be used for design where applicable. Most of the procedures have been field-verified and used by highway agencies for several years.

The selection of one procedure over another is usually based on the highway agency’s experience and satisfaction with the design results. 

The American Association of State Highway and Transportation Officials (AASHTO) design method and the Transport and Road Laboratory (TRL) CBR design method will be presented here.                

AASHTO Design method

This design method is empirical and was based on the AASHTO Road Test conducted in the United States in the early sixties.

The pavement system structural requirements needed to sustain the design traffic loadings for the design period is expressed in terms of a parameter called the structural number (SN). The structural number may be considered to be the required bearing capacity of the pavement.

The structural number may be estimated from nomographs  but for design ESAL greater than 500,000, the following empirical relationship  may be used for its estimation;

The structural number

where,

  SN= structural number in mm

CBR=design sub-grade CBR in % 

.   R=regional adjustment factor  with a value =1 for areas with rainfall throughout  the year and value =0.1 for dry and arid conditions, and intermediate values for intermediate rainfall conditions.

The following equation is then used to relate SN to the to the individual material types and thickness of the pavement:

Structural number

where Di=thickness of layer i in mm and

ai=layer coefficient of layer i

The data in the table below provides an indication of the magnitude of layer coefficients for different pavement materials.

Typical layer coefficients for layer materials

For any given structural number, there is an infinite number of pavement material combinations and thicknesses that will provide satisfactory service.

In view of this, there are guidelines that can be used to narrow down the number of solutions. Experience has shown that the wearing layer can be 50 to 100 mm thick while the sub-base can be 100 to 200mm thick.

Because the cost of the construction must be minimized, a knowledge of which of the materials available is the most costly per unit thickness will also assist with the solution of initial layer thickness.

TRL CBR Design method

The design categorises traffic in terms of equivalent axles into eight classes (T1-T8)  and the sub-grade strength in terms of CBR into six classes (S1-S6).

For any combination of traffic and sub-grade class, the appropriate pavement structure is selected from a structural catalogue. 

The structural catalogue provides the thickness of the sub-base, base, and wearing course or a surface dressing combination that will suit the loading conditions and sub-grade strength.   

Materials to be used for each layer are specified in quality or character in the corresponding design chart.  The accompanying charts provide the structural catalog for the design by this method.

This is the end of this post on Structural Design of Flexible Highway Pavements.

Leave a Comment