Dynamic traffic maneuvers such as acceleration, braking, and steering induce spatially varying vertical and horizontal tire loads, resulting in complex multiaxial stress states within layered asphalt pavements. These responses are governed by the viscoelastoplastic material behavior of asphalt and by induced anisotropies. In the context of emerging digital twin concepts for road infrastructure, these effects motivate the combination of high-fidelity numerical studies with experimental traffic data from weigh-in-motion (WIM) systems and established tire load models to construct physically informed, data-driven virtual representations of pavement behavior over large spatial scales. This paper presents a dynamic multiscale ALE–microlayer framework within the finite element method for asphalt pavement structures. The approach builds on Arbitrary Lagrangian–Eulerian (ALE) formulations for moving loads and their extension to inelastic material behavior. Asphalt is described by a thermodynamically consistent multiscale microlayer model that has recently been extended to viscoelastoplastic features at finite strains. Microscale constitutive relations are evaluated analytically and homogenized to the macroscale, allowing essential material mechanisms to be captured efficiently while avoiding the computational cost of classical FE² approaches. Within the ALE formulation, the load application region remains fixed on the pavement surface, while the pavement structure flows underneath. This enables the simulation of long traveled distances using a limited numerical domain and is particularly well suited for digital shadow representations as a prerequisite for future digital twin applications. The framework is demonstrated by two data-informed numerical examples. In the first example, a tire rolls over 700 m under dynamic driving conditions including acceleration, braking, and cornering, with tire load characteristics based on established vehicle dynamics models, and the resulting stress and strain states are evaluated for a pavement structure designed according to German guidelines. In the second example, real WIM traffic data are employed. Assuming constant vehicle velocities and a sufficiently long pavement segment, accumulated viscoelastoplastic effects due to repeated vehicle passages are numerically identified, providing a basis for long-term durability predictions using time-homogenization techniques.