The increasing number of waste rubber tires has attracted the attention of more and more researchers. Rubber asphalt has better performance compared with original asphalt. However, the compatibility between rubber powder and asphalt is poor because of the difference in physical and chemical properties and the improvement of high-temperature performance of asphalt by rubber powder is very limited. The compatibility between rubber powder and original asphalt plays a key role in the storage stability and rheological performance of rubber asphalt. This paper provides a comprehensive overview of rubber asphalt, factors influencing compatibility, compatibility evaluation methods, and improvement approaches. Desired compatibility results from the fact that rubber powder is cross-linked with molecules of original asphalt and evenly distributed in the original asphalt, forming a homogeneous system. Starting from the preparation process of rubber asphalt, the best preparation process was summarized. Then the activity of rubber powder is improved by physical and chemical methods, or other additives are added to finally promote the formation of a cross-linking network structure between rubber powder and original asphalt. Rheological method is the most widely used method in compatibility evaluation, but it is gradually accepted by researchers to evaluate the compatibility by observing the molecular morphology of rubber asphalt.

Steel bridge deck pavements (SBDPs) are susceptible to complex mechanical and service environmental conditions, yet current design methods often struggle to simultaneously capture global bridge system behavior and local pavement responses. To address this issue, this study develops a multi-scale finite element modeling framework that integrates a full-bridge model, a refined girder-segment model, and a detailed pavement submodel. The framework is applied to an extra-long suspension bridge to evaluate the mechanical responses of five typical pavement structural configurations—including double-layer SMA, double-layer Epoxy Asphalt (EA), EA-SMA combinations, and a composite scheme with a thin epoxy resin aggregate overlay. By coupling global deformations from a full-bridge model to the local pavement submodel, the proposed method enables a consistent assessment of both bridge-level effects and pavement-level stress concentrations. The analysis reveals that pavement structures significantly alter the stress and strain distributions within the deck system. The results indicate that while the composite configuration with a thin overlay effectively reduces shear stress at the pavement–deck interface, it results in excessive tensile strain, posing a high risk of fatigue cracking. Conversely, the double-layer EA configuration exhibits the lowest fatigue-related strain, demonstrating superior deformation coordination, while the optimized EA-SMA combination offers a robust balance between fatigue control and interfacial stress distribution. These findings validate the effectiveness of the multi-scale approach for SBDP analysis and highlight that rational structural configuration selection—specifically balancing layer stiffness and thickness—is critical for enhancing the durability and long-term performance of steel bridge deck pavements.