How to Calculate Asphalt Pavement Rutting Using Road Vehicle Dynamic Load Coupling
Rutting is easy to see and hard to explain with a single number. A wheel path sinks, the lane collects water, steering gets worse, and the pavement may still look intact outside the loaded strip. The usual first answer is “heavy trucks caused it,” which is true but incomplete.
A truck does not apply a perfectly steady load to the pavement. Tires bounce. Suspensions move. Pavement roughness feeds motion back into the vehicle. The result is a tire force that changes from moment to moment as the wheel travels. That changing force can make rutting worse than a static axle load would suggest.
A better calculation treats the vehicle and pavement as a coupled system. The pavement surface excites the vehicle, the vehicle changes the tire load, and the tire load changes the stress and strain inside the pavement. This article explains a practical method for calculating asphalt pavement rutting from that road-vehicle dynamic load coupling.
Why static axle load is not enough
A static rutting calculation often starts with axle load, tire pressure, contact area, pavement layer thickness, temperature, and material properties. That is useful for early design, but it misses the motion of the vehicle.
In the field, the force at the tire-road contact is not constant. It changes because of:
Pavement roughness
Tire stiffness and damping
Suspension stiffness and damping
Vehicle speed
Axle configuration
Load distribution
Braking, acceleration, and steering
Temperature-dependent asphalt stiffness
The tire may press harder than the static load at one point and lighter at the next. The pavement also responds differently depending on the duration of loading. A slow truck gives the asphalt more time to deform. A faster truck applies shorter load pulses, but roughness can produce larger dynamic peaks.
That is why dynamic tire load matters. Rutting is not only a question of how much load crosses the pavement. It is also a question of how that load varies in time and where the peaks occur.
A simple way to express the idea is:
\[
F(t) = F_s + F_d(t)
\]
Where:
\(F(t)\) is the tire load at time \(t\)
\(F_s\) is the static wheel load
\(F_d(t)\) is the dynamic part caused by vehicle-road interaction
Engineers often describe the severity of this variation with a dynamic load coefficient:
\[
DLC = \frac{\sigma_F}{F_s}
\]
Where \(\sigma_F\) is the standard deviation of the dynamic tire force. A higher coefficient means the wheel load is more variable. For rutting, those repeated high-load moments can be more damaging than a smooth passage at the same mean load.
Build the coupled road-vehicle model
A coupled model connects two physical systems: the vehicle and the pavement. The vehicle model predicts tire force. The pavement model predicts stress, strain, and permanent deformation from that force.
For many rutting studies, the vehicle does not need to be modeled as a full truck with every suspension detail. A quarter-car model or half-car model can be enough if the goal is to capture vertical dynamic loading.
A common quarter-car model includes:
Sprung mass The vehicle body supported by the suspension
Unsprung mass The wheel, tire, axle end, and related components
Suspension spring and damper These control body motion
Tire stiffness and damping These control how the tire transmits load to the pavement
Road profile input The pavement elevation under the moving tire
The road profile is usually written as \(r(x)\), where \(x\) is distance along the road. Since the vehicle moves at speed \(v\), the tire experiences that profile as a time input:
\[
r(t) = r(vt)
\]
The vertical tire force can then be estimated from tire deflection:
\[
F(t) = k_t[z_u(t) - r(t)] + c_t[\dot{z}_u(t) - \dot{r}(t)]
\]
Where:
\(k_t\) is tire stiffness
\(c_t\) is tire damping
\(z_u(t)\) is unsprung mass displacement
\(r(t)\) is road surface elevation
Dots represent time derivatives
This force becomes the moving load applied to the pavement model.
Choose the pavement response model
The pavement model can range from simple to detailed.
A layered elastic model treats the pavement as horizontal layers with assigned stiffness and Poisson’s ratio. It can estimate vertical compressive strain and shear strain under the tire. This approach is fast and useful for repeated calculations.
A viscoelastic model is better for asphalt because asphalt stiffness depends on temperature and loading time. At high temperature or slow speed, asphalt behaves softer and rutting risk rises.
A finite element model can represent:
Moving tire loads
Nonuniform contact pressure
Temperature gradients
Layer interfaces
Nonlinear asphalt behavior
Subgrade response
Dual tires or wide-base tires
The right model depends on the study goal. If the goal is network-level screening, a layered model may be enough. If the goal is a research-grade rutting estimate for a specific pavement structure, finite element modeling gives more control.
Represent the tire contact pressure
Do not reduce everything to a point load. Rutting develops under the contact area, so pressure distribution matters.
A basic model uses uniform circular or rectangular pressure:
\[
p = \frac{F(t)}{A}
\]
Where:
\(p\) is contact pressure
\(F(t)\) is dynamic tire load
\(A\) is tire contact area
A more detailed model uses nonuniform pressure. Real tires often create pressure peaks near tread ribs or shoulders. That detail can matter near the surface, especially for asphalt layers that rut in the top few inches.
For many practical calculations, a rectangular contact area with time-varying pressure offers a good balance. The load moves along the pavement at vehicle speed, and the pressure rises or falls as \(F(t)\) changes.
Convert dynamic load into pavement rutting
The calculation has two main parts. First, find pavement strain from the moving dynamic load. Next, convert that strain into permanent deformation.
The vertical rut depth is the sum of permanent vertical strain through the rutting-susceptible layers:
\[
RD = \int_0^H \varepsilon_p(z)\,dz
\]
Where:
\(RD\) is rut depth
\(H\) is the depth of the layer or layers being evaluated
\(\varepsilon_p(z)\) is permanent vertical strain at depth \(z\)
In a layered calculation, this becomes a summation:
\[
RD = \sum_{i=1}^{n} \varepsilon_{p,i}h_i
\]
Where:
\(h_i\) is the thickness of sublayer \(i\)
\(\varepsilon_{p,i}\) is permanent strain in that sublayer
The challenge is getting \(\varepsilon_p\). Asphalt does not deform permanently in a single step under normal traffic. Rutting accumulates through thousands or millions of load repetitions. A permanent strain model usually relates plastic strain to stress, recoverable strain, load repetitions, temperature, and material constants.
A general form is:
\[
\varepsilon_p = f(\varepsilon_r, N, T, \sigma)
\]
Where:
\(\varepsilon_r\) is resilient or recoverable strain
\(N\) is number of load applications
\(T\) is temperature
\(\sigma\) is stress state
Some rutting models use a power law form:
\[
\varepsilon_p = aN^b\varepsilon_r^c
\]
Here, \(a\), \(b\), and \(c\) are material parameters found from laboratory repeated load tests or calibrated field data. The exact values should come from the asphalt mixture and conditions being studied, not from a generic assumption.
Account for temperature and speed
Temperature can dominate rutting behavior. A pavement that resists rutting well in cool weather may deform much faster on a hot summer afternoon.
The asphalt modulus should change with temperature and loading frequency. Vehicle speed affects loading time. A slow load has a lower effective frequency, which makes asphalt behave softer. A fast load has a higher effective frequency, although roughness can increase dynamic load peaks.
A useful workflow assigns pavement temperature by depth, then updates asphalt stiffness and permanent deformation parameters for each layer. Near-surface asphalt often sees the largest temperature swings, so it may need finer layer divisions.
Count traffic in load groups
A real pavement carries many vehicles, not one truck. To calculate cumulative rutting, traffic should be grouped by axle type, axle load, tire type, tire pressure, and speed range.
For each group:
Calculate the dynamic tire load history.
Apply the moving load to the pavement response model.
Compute permanent strain for the relevant number of passes.
Add the rutting contribution to the total.
The total rut depth becomes:
\[
RD_{total} = \sum_{j=1}^{m} RD_j
\]
Where \(RD_j\) is rutting caused by traffic group \(j\).
This approach avoids treating all trucks as equal. A lightly loaded truck on a smooth surface may be less damaging than a heavily loaded truck crossing a rough section at a speed that excites suspension bounce.
A practical calculation workflow
The full method can be organized into a repeatable process.
1. Collect pavement and traffic inputs
Start with physical inputs that can be measured or reasonably estimated:
Input group | Examples |
Pavement structure | Asphalt thickness, base thickness, subgrade type |
Material behavior | Modulus, Poisson’s ratio, permanent strain parameters |
Surface condition | Longitudinal profile, roughness, rut depth history |
Traffic | Axle loads, tire pressure, speed, axle spacing |
Environment | Pavement temperature, seasonal conditions |
The road profile is central to coupling. Smooth pavement produces smaller dynamic tire force changes. Rough pavement can amplify tire load and speed up rutting, especially if heavy trucks repeat the same wheel path.
2. Solve the vehicle dynamic response
Use the road profile and vehicle speed as inputs to the vehicle model. The output is the tire force history \(F(t)\).
For a single wheel path, the tire force record may look like a fluctuating line around the static load. Peaks occur where surface elevation changes excite the tire and suspension. These peaks are then translated into contact pressure on the pavement.
3. Apply the moving pressure to the pavement model
Convert \(F(t)\) into pressure over the contact area. Move that pressure across the pavement model at the vehicle speed.
At each time step, calculate the stress and strain at selected depths and lateral positions. Rutting is usually evaluated under or near the wheel path centerline, but lateral distribution matters for dual tires and wander.
4. Calculate incremental permanent strain
Use a rutting model to convert response strain into permanent strain for each pass or group of passes.
A common approach is layered accumulation:
\[
\Delta RD_i = \Delta \varepsilon_{p,i}h_i
\]
Then sum the layer contributions. Asphalt layers, granular base, and subgrade can all contribute to rutting. The dominant layer depends on pavement structure, material quality, temperature, and moisture.
5. Repeat for traffic and time
Run the calculation across traffic groups and climate periods. Hot months may receive separate treatment because asphalt rutting sensitivity rises with temperature. Wet subgrade periods may also need special handling if lower layers contribute to permanent deformation.
6. Compare with measured rutting
No model should stand alone. Compare calculated rut depth with field measurements or accelerated pavement testing when possible.
Useful checks include:
Does the model match early rutting trends?
Does it overpredict rutting in cool periods?
Does it show higher rutting where roughness is higher?
Does the predicted rut shape match the field wheel path?
Are calculated strains reasonable for the material type?
Calibration can adjust permanent strain parameters, layer moduli, or contact pressure assumptions. The goal is not to force a perfect fit. The goal is to make the model physically credible and useful for design decisions.
What the coupled method reveals
A coupled road-vehicle method adds work, but it also gives better answers than a static load check.
It can show why two pavements with the same truck traffic rut at different rates. One may be smoother, cooler, or stiffer. Another may have roughness that increases dynamic tire force at the same location again and again.
It can also show why maintenance timing matters. Surface roughness is not only a comfort issue. Once roughness increases dynamic loading, it can raise stress in the wheel path and contribute to faster rut growth. That creates a feedback loop: rutting and roughness increase load variation, which can increase deformation.
The method is also useful for comparing design choices, such as:
Thicker asphalt layers
Stiffer or more rut-resistant mixes
Improved base support
Lower tire pressure assumptions
Different speed conditions
Surface smoothing before structural damage grows
For pavement management, the most useful output is not just a final rut depth. It is the predicted rutting trend over time. A trend helps identify when rutting may reach a service limit and which factor drives the growth.
The core idea is simple: rutting depends on repeated pavement strain, and repeated pavement strain depends on the actual tire force moving across the road. When the tire force is calculated from vehicle-road coupling, the rutting estimate reflects field behavior more closely than a static load method.
A sound calculation links the chain from roughness to vehicle motion, from vehicle motion to tire force, from tire force to pavement strain, and from pavement strain to permanent deformation. If any link is weak, the rutting prediction becomes less reliable.
The best next step is to start with a simple coupled model, calibrate it against measured rutting, and add detail only where it improves the result. For many pavement studies, that balance gives a method that is practical, repeatable, and strong enough to guide design and maintenance decisions.
Comments