Asphalt mortar is a key constitutive phase in asphalt-mixture mesoscale modeling, but its thermo-rheological characterization usually requires repeated testing over multiple temperatures and loading frequencies, especially across diverse binder systems including rubber-containing modified binders. This study proposes a physics-informed generative adversarial network (PI-GAN) and a physics-regularized multilayer perceptron (MLP) to learn asphalt-mortar viscoelastic master curves from limited experimental data and reduce the burden of repeated mortar-scale testing. A dual-sigmoidal representation with temperature-frequency shifting is adopted, and physical admissibility is enforced during both data generation and prediction through constraints on modulus behavior, phase response, and Prony realizability. A differentiable surrogate is introduced to connect master-curve parameters with relaxation spectra during training. Results from the asphalt-mortar dataset show that about 87% of the raw outputs generated by PI-GAN pass the prescribed physical screening and are retained for downstream training. The physics-regularized MLP substantially improves physical validity while maintaining comparable regression accuracy. Specifically, 96% of the predicted responses satisfy the requirement that the dynamic modulus remains greater than the storage modulus over the full temperature-frequency domain, and 99.7% satisfy Prony realizability. The proposed method provides an efficient route for predicting asphalt-mortar viscoelastic master curves, reducing repeated laboratory characterization, and generating direct constitutive inputs for asphalt-mixture mesoscale finite element modeling.

Conventional gradation design for hydraulic asphalt concrete does not explicitly control coarse-aggregate skeleton formation and internal load-transfer paths, limiting the coordinated optimization of multiple performance requirements for dam facings. To address this limitation, a two-stage mesostructure-informed gradation design method integrating Particle Flow Code (PFC) simulation, X-ray computed tomography (CT) characterization, and systematic performance tests was developed. PFC was first used to optimal coarse-to-fine aggregate ratio based on particle packing and force-chain uniformity, followed by CT-based optimization of the internal coarse-aggregate fractions through contact characteristics, spatial uniformity, and compaction orientation. The results show that a 70:30 coarse-to-fine aggregate ratio achieves the densest packing and the most uniform force-chain distribution.Compared with the specification-recommended hydraulic asphalt concrete gradation, the optimized gradation reduces the strength contribution of mortar-mortar contacts by 21.62% and increases that of aggregate-mortar contacts by 65.63%, reconstructing the internal load transfer network from mortar-dominated bearing to aggregate-mortar collaborative bearing. A 3:1 mass ratio (G2 gradation) of 13.2–9.5mm to 9.5–4.75mm aggregates produced the most favorable mesostructure. The G2 gradation increased residual Marshall stability and low-temperature flexural failure strain by 3.75% and 4.56%, respectively, while reducing the 70 °C slope-flow value by 17.84%. These findings confirm that rationally designing mixture gradation based on quantifiable mesostructural parameters can provide a replicable reference for the proportion design of hydraulic asphalt mixtures.

During asphalt pavement construction, compaction is conventionally assessed by coring to measure density, with the destructive and time-consuming nature of this method limiting its frequent application. In contrast, intelligent compaction (IC) technologies enable real-time quality control and have driven significant advances in asphalt pavement construction. Building on these advancements, this study presents a comprehensive review of research published between 2010 and 2025, synthesizing developments in IC and integrated multi-sensor systems, which combine data from accelerometers, temperature sensors, and GPS to monitor compaction quality and spatial variability, and highlighting their implications for pavement performance. Recent developments demonstrate the growing use of machine learning (ML) for processing IC data to enhance predictions of stiffness, modulus, and density. Meanwhile, Digital Twin (DT) frameworks have emerged as powerful tools for data-driven decision-making for preventive maintenance and rehabilitation of pavements. The integration of IC with Building Information Modeling (BIM) and the Internet of Things (IoT) further enhances construction management and quality monitoring. Additionally, the fusion of LiDAR and camera data offers millimeter-level precision in paving operations. Despite these advances, key research gaps persist, particularly in establishing quantitative relationships between Intelligent Compaction Measurement Values (ICMVs) and asphalt modulus evolution, and in understanding the effects of temperature variation and underlying layer conditions. This review concludes that continued innovation in Artificial intelligence (AI)-driven modeling, multi-sensor fusion, and mechanistic interpretation is essential to fully realize the potential of intelligent, data-centric pavement construction.

As a key design and construction control indicator of asphalt mixtures, void content significantly affects the freeze-thaw (FT) deterioration performance of asphalt mixtures in cold regions. To reveal the mechanism between initial void content, pore structure evolution and interfacial transition zone (ITZ) micromechanical degradation under FT cycles, this study employed an integrated approach using X-ray computed tomography (XCT), mercury intrusion porosimetry (MIP), and nanoindentation tests to examine the pore structure of asphalt mixtures after 0, 5, 15, 25 and 35 FT cycles. The change in pore morphology, structure and micro-mechanical properties of asphalt mixtures with initial void contents of 1%, 3% and 5% after FT cycles were evaluated. The results demonstrate that the initial void content considerably influences the FT damage mode: low-void content samples primarily develop new pores, whereas relatively high-void content samples exhibit both pore formation and expansion, and higher initial void content accelerates the initiation of pore expansion. The difference in pore evolution path further leads to completely different damage modes of the ITZ: the new pore generation in low-void samples produces micro-cracks, whose tip stress causes tearing damage to the ITZ; while the pore expansion in high-void samples generates continuous extrusion stress on the surrounding ITZ, leading to extrusion damage. The ITZ modulus and hardness, which are the most sensitive mechanical indicators to FT damage, show significantly different degradation rates under the two damage modes. Pearson correlation analysis confirms strong correlations (P <0.01) among void content, hardness, modulus, and pore fractal dimension (D-f). The Levenberg-Marquardt optimization algorithm effectively characterizes the FT damage deterioration process in asphal t mixtures. These findings provide valuable insights for enhancing frost resistance and predicting FT damage in asphalt mixtures employed in cold regions.

Styrene-butadiene-styrene copolymer and crumb rubber modified asphalt (SBS/CRMA) is widely used in infrastructure structure. However, its durability is significantly influenced by creep behavior, particularly under low-temperature conditions and freeze–thaw cycles, making accurate and explainable prediction essential for design and application. This study explores a predictive framework for low-temperature creep behavior that integrates spectral information coupling encoding with physics-informed neural networks (PINNs). By combining Gramian Angular Field (GAF) transformation with AutoEncoders (AE), the framework develops a feature extractor capable of coupled FTIR spectral information and capture micro-component inputs. Two PINNs with distinct tasks and architectures were developed for prediction. To advance interpretability, SHapley Additive exPlanations (SHAP) and a developed interpretation strategy for the encoded spectral are employed, providing multi-scale insights into the influence of both macroscopic and microscopic features. The results demonstrate that the feature extraction successfully captures functional group coupling information, improves prediction accuracy by 9.4%, and the PINNs models exhibit clear advantages over conventional approaches. Furthermore, the framework was integrated into a material design optimization workflow, and its feasibility was verified through application in a specialized engineering project. This study advances intelligent characterization and design of SBS/CRMA, offering a novel perspective for modeling and design of complex materials.

Accurately evaluating the high-temperature performance of asphalt mixtures can provide useful performance-oriented support for their design and production. However, conventional performance testing and mixture design methods often involve long cycles and high costs. This study proposed a data-driven and efficient framework for predicting high-temperature performance and enabling automated design assistance of asphalt mixtures. A database was established from laboratory tests covering asphalt binder properties, volumetric characteristics, gradation-related descriptors, and dynamic stability (DS), followed by Grey Relational Analysis (GRA) to screen highly relevant 21-dimensional feature set. A Gaussian Process Regression (GPR) machine learning (ML) model was then developed to predict the dynamic stability (DS) of asphalt mixtures and benchmarked against Support Vector Regression (SVR) and Artificial Neural Network (ANN) models. The results show that the GPR model achieved superior performance in prediction accuracy, generalisation capability, and computational efficiency, with a coefficient of determination (R-2) of 0.9889, and a normalised root mean square deviation (NRMSD) of 0.0325. To enhance interpretability, explainable machine learning (XML) techniques were applied to analyze feature contributions within the GPR model, with results compared to those from GRA. Subsequently, the GPR model was integrated with Bayesian Optimization (BO) to develop an automated design optimisation approach for high-temperature performance. Experimental validation yielded a design error of 6.95%, confirming the method's practical applicability. Furthermore, a multi-module software system was implemented to support intelligent design and optimisation of asphalt mixtures for high-temperature performance. These advancements provide both theoretical foundations and practical tools for promoting digital transformation in pavement engineering.

The severe climatic conditions and intense economic activities in the Hexi region of China pose significant challenges to the durability of asphalt pavements. To address this issue, the study developed an epoxy resin/ activated rubber powder modified asphalt (EARMA) binder. To accelerate the application of this material in the Hexi region of China, a thorough evaluation was conducted encompassing the optimization of the preparation process, high-, low-, and intermediate-temperature performance, aging resistance, the reaction mechanism, and cost-effectiveness of EARMA. The results indicate that the optimal formulation of EARMA for the Hexi region of China comprises an activated rubber powder dosage of 21.30 % (by weight of asphalt), epoxy resin dosage of 4.25 % (by weight of asphalt), the ratio of epoxy resin to curing agent of 1.27, curing at 170 degrees C for 3 h, followed by 2 days at 60 degrees C. The results of assessments on high-, low-, and intermediate-temperature performance, as well as aging resistance suggest that EARMA not only shows exceptional resistance to rutting but also exhibits favorable low-temperature performance, fatigue resistance, and aging resistance. Reaction mechanism studies demonstrate that epoxy groups within epoxy resins undergo ring-opening reactions to crosslink with polar functional groups (such as secondary amines, tertiary amines, hydroxyl groups, and carboxyl groups) present in curing agent, activated rubber powder, and base asphalt, forming ether compounds. Raw material economic evaluations reveal that the raw material cost of EARMA is reduced by 21.4-84.1 CNY/ton in comparison to styrene-butadiene-styrene modified asphalt and styrene-butadiene rubber modified asphalt.

Semi-Flexible Pavement (SFP) is a composite material composed of an asphalt matrix and cement grout, widely used in heavy-load and channelized areas due to its excellent deformation resistance. However, due to the significant differences in thermal sensitivity between asphalt and cement, as well as its complex mesostructure, SFP often encounters challenges such as high-temperature cracking during service. This study investigates the mechanical responses of SFP under multi-temperature conditions. Industrial computed tomography (CT) scanning and three-dimensional reconstruction techniques were employed to quantitatively characterize key mesostructural features, including connected voids, asphalt mortar, and aggregate skeleton. Integrated correlation analysis was used to reveal the evolution mechanisms between meso-structure and macro-performance across different temperature conditions. The results indicate that the high-temperature mechanical properties of SFP are strongly associated with the connectivity of the cement skeleton, the asphalt-cement interface area, and the total number of aggregate contacts. As the temperature decreases, the contribution of the cement skeleton and aggregate interlock diminishes, while the bonding effect of asphalt mortar becomes more pronounced. Mechanism analysis suggests that increasing the design void ratio and reducing the aggregate size can promote the formation of a stronger cement skeleton and improve high-temperature performance. However, excessively high void ratios may reduce asphalt mortar thickness, significantly reducing crack resistance at medium and low temperatures. To balance multi-temperature performance, this study proposes an optimized gradation design, including a nominal maximum aggregate size of 13.2 mm, a target void ratio of 20%, and a mass ratio of 2:1–3:1 between 9.5–13.2 mm and 4.75–9.5 mm aggregates. This research elucidates the mesostructure–performance relationship of SFP under multi-temperature conditions and provides theoretical insights for material design and engineering applications.

Accurate assessment of asphalt aging is fundamental to durability evaluation and maintenance decision-making for asphalt concrete waterproofing layers in high-speed railway subgrades. This study developed a laboratory accelerated aging system to simulate the combined effects of thermal, radiation, moisture, and oxygen exposure. Using three representative modified asphalt binders, the physicochemical mechanisms of asphalt aging were investigated through the combined analysis of rheological indicators, Fourier-transform infrared (FTIR) spectra, and atomic force microscopy (AFM) parameters. An indoor–field aging equivalence relationship was subsequently established by matching macroscopic and microscopic indicators obtained from laboratory-aged binders with those binders recovered from 5-year in-service field samples. The results showed that environmental exposure progressively hardened the asphalt binders, promoted the formation of oxygen-containing functional groups, increased surface roughness and stiffness, and continuously elevated cracking susceptibility. Among the three modified binders, DCR/SBSCMA exhibited the greatest resistance to environmental aging. Its aging process could be divided into three stages: an initial slow stage (0–20 d), an accelerated stage (20–80 d), and a stabilization stage (80–100 d). By jointly considering indicator sensitivity and the fitting accuracy of the Verhulst aging models, 5 years of field service was determined to be equivalent to 29.3 d of indoor accelerated aging. Furthermore, the waterproofing layer was predicted to enter the severe cracking-risk stage after approximately 11–13 years of service. The proposed multi-scale aging assessment and indoor–field aging equivalence framework provides a scientific basis for durability evaluation, service-life prediction, and maintenance decision-making for asphalt concrete waterproofing layers in high-speed railway subgrades.

The aging phenomenon of asphalt binder is a critical factor that observably affects pavement durability. Utilizing the desulfurization process to develop desulfurized rubber-modified asphalt (DRMA) has emerged as an effective method for improving the aging resistance of asphalt. This study systematically evaluated the rheological properties of rubber-modified asphalt (RMA) and DRMA, before and after both thermo-oxidative and pressure aging. The swelling behavior of crumb rubber (CR) and alterations in the chemical structure of asphalt binder were also detected to elucidate the aging mechanism. Both the deformation resistance and fatigue performance of RMA and DRMA were enhanced after thermo-oxidative aging, which is attributed to the swelling of undissolved rubber particles, reinforcing the internal network structure in asphalt binder. Conversely, pressure aging was characterized by rubber degradation, leading predominantly to the breakdown of the network structure, which in turn resulted in deteriorated rheological properties. When subjected to equivalent aging conditions, DRMA consistently demonstrated superior rheological performance compared to RMA. This superiority was primarily due to the improved compatibility of desulfurized rubber with asphalt, which facilitated the formation of a more uniform and robust internal network structure. Furthermore, the greater integration of carbon black and aging inhibitors into the asphalt from the desulfurized rubber further enhanced the resistance to aging.

Effective compaction quality significantly impacts pavement durability and performance, with uneven compaction often resulting from traditional empirical approaches that adjust vibration modes and rolling periods. This paper investigates the effects of vibratory roller amplitude optimization on asphalt pavement compaction and develops predictive models for intelligent compaction (IC) parameters and in-place density using machine learning (ML) methods. Results indicate that higher-amplitude vibration passes yielded higher average intelligent compaction measurement values (ICMVs) and in-place density. Predictive models were developed using XGBoost and CatBoost, optimized through the Bayesian Optimization Algorithm (BOA). Among them, the XGBoost models achieved the best performance in predicting ICMVs and non-nuclear density (NND) values, demonstrating high accuracy (R2 = 0.9362, 0.9916) and low error (RMSE = 4.5678, 1.1934) during validation. These findings have significant implications for compaction quality control and provide a foundation for future research to enhance predictive models for ICMVs and NND under diverse construction conditions.

The skid resistance of runway is critical for ensuring safety. To address durability assessment for runway material selection, this study presents a comprehensive method that couples indoor accelerated wear testing with fullspectrum texture analysis. Three types of asphalt mixtures with different binders (SBS-modified asphalt, polyurethane-modified asphalt (PU), and epoxy asphalt (EP)) were selected for tests under a ground contact pressure of 1.5 MPa at both high-temperature and normal-temperature conditions. The tests were carried out using a self-developed indoor accelerated wear device. Mean Profile Depth (MPD), Power Spectral Density (PSD), and fractal dimension were adopted as full-spectrum texture features, and the British Pendulum Number (BPN) was used to evaluate skid-resistance durability. Grey relational analysis was further applied to examine the relationship between surface texture and skid performance. The results indicated that SBS-modified asphalt exhibited early plastic deformation exceeding 5 mm under heavy aircraft loads. This thermoplastic behavior accelerated the loss of skid resistance. In contrast, the thermoset properties of epoxy and polyurethane asphalt mixtures ensured superior skid and wear resistance. Additionally, Stone Matrix Asphalt (SMA-13) graded mixtures outperformed Asphalt Concrete (AC-16) in wear resistance, indicating better texture retention capabilities. Fractal dimension showed strong correlation with MPD, proving valuable for characterizing surface texture evolution in asphalt mixtures.

Uniformly dispersed in the asphalt mixture, fibers can adsorb asphalt, reinforce asphalt concrete, and improve its pavement performance. It can be applied to recycled asphalt mixture to address the performance defects of RAP material. To clarify the influence of fiber properties on the fatigue performance of recycled asphalt mixture, the four-point bending fatigue test and scanning electron microscope(SEM) were used to analyze the fatigue performance of lignin fiber (LF), basalt fiber (BF), BF and polyester fiber (PF) mixed recycled asphalt mixture. The main conclusions are as follows: Because the three-dimensional network structure formed by basalt fiber improves the strength of asphalt mixture, the deformation resistance ability of polyester fiber increases the toughness of asphalt mixture. BF and PF can significantly improve the fatigue performance of asphalt mixture, especially under low strain conditions. However, under large strain, due to the synergistic effect of the two fibers, the inherent performance limitations of the fiber were magnified, resulting in an increased decay range of the fatigue life of the AC20-BF + PF asphalt mixture. Both the strain level and fiber type can influence the decay trend of the bending stiffness modulus of asphalt mixtures. Selecting an appropriate strain level is crucial for accurately assessing the effect of fibers on the fatigue performance of asphalt mixtures. The phase angle of asphalt mixture under load shows different fluctuation trends due to different fibers, but the change trend of accumulative dissipated energy was not affected by materials and external loads. According to the grey correlation degree, the strain level is the main factor affecting the change rate of cumulative dissipated energy. BF and PF act as the role of improving strength and toughness in asphalt mixture respectively. As N/Nf gradually increased, the cumulative stiffness modulus attenuation ratio (CSMDR) of asphalt mixture increased logarithmically. Through parameter transformation and the least squares methods, the CSMDR calculation equation for asphalt mixtures under different strain levels was derived, incorporating fiber modulus and strain level. The ratio of the CSMDR of asphalt mixture under N loads to that after Nf loads can characterize the damage to asphalt mixture after N loads.

Epoxy resins (ERs) are esteemed for their mechanical robustness and adhesive qualities, particularly in steel bridge deck applications. Nonetheless, their intrinsic brittleness limits broader utility. This study addresses this limitation by modulating ER crosslink density through adjustments in curing agent concentration, incorporation of hyperbranched polymers (HBPs), and optimization of curing conditions. Employing a multi-objective optimization strategy, this research aims to enhance toughness while minimizing strength degradation. Non-isothermal curing kinetics, realized using the differential scanning calorimetry (DSC) method, attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR), tensile testing, and thermogravimetric analysis (TGA), were employed to investigate the effects of curing agent and HBP content on the curing reaction, mechanical properties, and thermal stability, respectively. Response surface methodology facilitated comprehensive optimization. Findings indicate that both curing agent and HBP contents significantly influence curing dynamics and mechanical performance. Curing agent content below 40% or above 50% can induce side reactions, adversely affecting properties. While a curing agent content exceeding 45% or an HBP content exceeding 5% improves the toughness of ER, these increases concurrently reduce mechanical strength and thermal stability. The study identifies an optimal formulation comprising 45.21% curing agent, a curing temperature of 60.45 °C, and 5.77% HBP content.

During the construction of Hot Mix Asphalt (HMA) pavements, compaction is crucial for ensuring the pavement's ultimate performance. Traditionally, in-place density is monitored using in situ spot tests with non-destructive density gauges (nuclear or non-nuclear) calibrated against core sample measurements. However, these density spot tests are often conducted at different intervals, hindering timely assessment. Intelligent compaction (IC) has emerged to enhance this process, equipping rollers with sensors to improve HMA compaction. This study aims to monitor the density of the asphalt pavement and develop predictive models for HMA density during construction, utilizing intelligent compaction measurement values (ICMVs), lanes, and pass counts. Data were collected from a construction project in Pakistan, including IC data from vibratory rollers and non-nuclear gauge (NND) density measurements. Machine learning methods—specifically support vector regression, random forest, extreme gradient boosting, and light gradient boosting—were applied and enhanced with a chaotic firefly algorithm to develop these predictive models. The robust model was interpreted using Shapley’s adaptive explanations (SHAP) to assess the influence of ICMVs on NND. The results revealed that the model's predictions closely aligned with measured NND values, with light gradient boosting demonstrating superior performance during validation and achieving an R2 value of 0.989 and RMSE less than 1.172. The SHAP analysis demonstrated a significant influence of ICMVs, particularly the compaction control value (CCV) and resonant meter value (RMV) with NND. Furthermore, the machine learning model's results outperform previous studies' findings. These findings suggest that utilizing ICMVs and machine learning predictions may effectively replace traditional NND density tests, providing easier and more accurate access for density determination during asphalt construction.

The full-section asphalt concrete waterproof layer (FACWL) has garnered significant attention for its outstanding ability to reduce frost heave and thaw-related weakening in railway track beds, particularly in seasonally frozen regions. To explore the dynamic properties of the FACWL, a fractional-order constitutive model was utilized to characterize the viscoelastic behavior of asphalt concrete. Additionally, a vehicle–track coupled finite element (FE) model and the numerical approach incorporating the fractional-order constitutive model were developed and validated via experimental and field testing. Simulation results indicate that applying the FACWL reduces the vertical dynamic response of each structural layer, vertical peak accelerations across the subgrade surface layer exhibited reductions exceeding 30% in both positive and negative directions. Moreover, the tensile strain at the bottom of the FACWL remained relatively low, less than 100 με. Compared with conventional waterproof sealing layers, the viscoelastic nature of the FACWL facilitates energy dissipation, effectively decreasing the overall vibrational amplitude and vertical deformation within the track structure by more than 20%. Consequently, the FACWL plays a crucial role in ensuring the long-term stability of the subgrade and minimizing vibrations in the track system.

Devulcanized rubber-modified asphalt (DRMA) possesses broad application potential due to its environmental friendliness and exceptional performance. However, the relationship between the rubber devulcanization degree, rubber content, and the modification effect of asphalt binder has not been clearly revealed. In this work, DRMA with different rubber devulcanization degrees and contents was prepared. The storage stability and rheological properties of several asphalt binders were evaluated, and the interaction mechanism between different devulcanized rubber (DVR) and asphalt was also detected. The storage stability of DRMA increased with the rubber devulcanization degree and decreased with the DVR content. The rheological test results demonstrated that under the same rubber devulcanization conditions, the elasticity ratio, rutting resistance, and fatigue resistance of the ML40 series was enhanced continuously, while the above properties of DRMA of the other series were enhanced and then weakened with the rubber devulcanization degree. Additionally, the DRMA of the ML60 series had a superior elasticity ratio and rutting resistance when the DVR content was the same. The low-temperature performance of DRMA increases with the rubber devulcanization degree and DVR content.

Microwave heating can be applied for the ice and snow melting and self-healing of asphalt pavement. The efficiency of microwave heating is subject to the dielectric constant of the material, the working frequency, and the depth of penetration of the electromagnetic wave. The purpose of this paper is to reveal the effect of steel slag content and frequency of microwave on the heating efficiency penetration depth. Compared with the mineral aggregate asphalt mixture, the heating rate of steel slag asphalt mixture dramatically increases under microwave heating. Microwaves at a frequency of 2.45GHz are recommended for asphalt surface layers over 40cm thick, while those at 5.8GHz are recommended for layers over 20 cm.