In this study, a novel digital twin (DT) modeling approach is developed to enhance the reliability assessment of civil structures subjected to multi-source uncertainties. Referring to the proposed procedure, a modular Bayesian inference (MBI) with the Transitional Markov Chain Monte Carlo (TMCMC) sampling algorithm is first used to calibrate the DT model of structures based on measured structural responses. Distinguished from the classic Bayesian inference approach, the uncertainty caused by modeling errors is added to the extended likelihood function by using a bias function. Then, the DT model calibration of structures can be achieved by three modules. Based on the calibrated DT model of structures, a finite number of discrete structural response samples can be generated by performing nonlinear dynamic analysis. Then, the generalized extreme value distribution (GEVD) is used to fit the probability density function (PDF) of these discrete response samples based on the maximum likelihood estimation (MLE) algorithm. Subsequently, the earthquake-induced failure probability of structures can be assessed by directly integrating the fitted GEVD and predefined structural failure thresholds. Numerical simulations on a three-story steel frame structure subjected to seismic excitations are developed to validate the feasibility of the proposed approach. The shake table tests on a scaled reinforced concrete (RC) column structure are further conducted to verify the effectiveness of the proposed approach. Both numerical and experimental results demonstrate that the proposed approach is reliable and highly effective for structural reliability assessment.

The scarcity of high-quality granular materials has led to exploring alternatives like fly ash (FA), an industrial byproduct of coal-fired power plants. Utilizing FA in pavement construction can reduce the reliance on crushed stone aggregates, conserving energy and protecting the environment. However, its use in structural layers (base and subbase) is limited due to fine particles that make it brittle when stabilized. To address this, stone dust (SD) and aggregates (AG) were mixed with FA (noted FA-SA) before cement stabilization to improve gradation and strength. Polypropylene fiber (FI) was added to enhance reinforcement and reduce brittleness. Fatigue failure, caused by repeated loading, is the primary mode of failure in stabilized layers, making the development of a fatigue performance model critical for Mechanistic-Empirical pavement design. This study investigated the fatigue performance of cement-stabilized FA-SA mixtures using cyclic indirect tensile tests. Experimental evaluations, conducted with and without fiber reinforcement, included indirect tensile strength, resilient modulus, and fatigue behavior using a servo-hydraulic testing machine. Results showed three stages of stiffness reduction, with a final modulus drop of 15%-30% compared to the initial modulus, indicating brittle behavior under stress-controlled conditions. A strain-based fatigue model was developed, with strain damage exponents for FA-SA composites stabilized with 4%-6% cement and 0.25%-0.35% fiber ranging from 4.37 to 4.55. These findings enhance the understanding of fatigue performance in FA-based stabilized layers, supporting the design of sustainable road infrastructure. • Three stages of stiffness reduction was identified for fiber-reinforced cement-stabilized fly ash aggregate mixture. • Stiffness modulus reduced to 15%-30% compared to the initial modulus, indicating brittle behavior under stress-controlled conditions. • Strain-based fatigue models were developed.

Porous silicon carbide (SiC) ceramic membranes require extremely high sintering temperatures, limiting their large‑scale adoption in cost-effective water and wastewater treatment. Incorporating NaA zeolite residue as a sintering aid markedly lowers the sintering temperature and promotes industrial waste valorization, but the coupled effects of multiple compositional and processing parameters remain poorly understood. In this study, we developed a skip-connection multi‑path multilayer perceptron (skip@M‑MLP) framework, together with several conventional machine learning models, to predict porosity and bending strength using a curated dataset from the literature. A hybrid data augmentation strategy combining Gaussian noise and linear interpolation was applied to address small‑sample limitations, and model interpretability was achieved via Shapley Additive Explanations (SHAP) and partial dependence plots (PDPs). The skip@M‑MLP achieved the highest accuracy among all tested models (R² up to 0.8716) and revealed that activated carbon content is the dominant factor governing the porosity–strength trade‑off, followed by sintering temperature and SiC fraction. These insights link data‑driven predictions with membrane engineering, providing a theoretical basis and quantitative guidance toward the potential scalable for the low-carbon fabrication of robust SiC membranes explicitly tailored for advanced water treatment processes. • ML optimizes NaA-assisted low-temp sintering of SiC water filtration membranes. • Hybrid data augmentation overcomes small-sample limits in membrane engineering. • Skip-connection MLP accurately predicts the porosity-strength of membranes. • Interpretable AI guides cost-effective design of water treatment membranes.

Hemp fibers as polymer matrix composites offer sustainability, cost-effectiveness, adaptability, and exceptional mechanical characteristics in materials science and engineering. This study examines hemp fiber-reinforced thermoset matrix composites, integrating walnut shell powder as filler and epoxy resin as matrix, assessing creep, fatigue, wear, and dielectric characteristics according to ASTM standards. The investigation varies the hemp fiber and epoxy resin ratios while keeping the walnut shell content constant. Four laminates (L1, L2, L3, L4) with unique mixtures of hemp fiber, resin, and walnut shell filler were specifically designed. L1 comprises 30% hemp fiber and 70% resin, without walnut filler. L2, L3, and L4 have progressively higher hemp fiber content, 10% walnut filler and varying the resin concentration appropriately. The amount of Epoxy resin in L1, L2, L3 and L4 are 70, 80, 70 and 60%, respectively. The results indicated that composites with walnut shell filler exhibit better dynamic characteristics than composites without it. Creep testing revealed reductions of 85.05%, 87.73%, and 91.41% for L2, L3, and L4 compared to L1. Fatigue life for L2, L3, and L4 increased by 17.98%, 12.97%, and 10.2%, and dielectric constant values rose by 17.02%, 34.04%, and 57.45%, respectively. Wear loss decreased by 12.49%, 13.51%, and 4.39%, while the coefficient of friction increased by 27.14%, 20%, and 8.57% for L2, L3, and L4 compared to L1. • Mix of 10% walnut shell powder filler, 30% hemp fiber, 60% epoxy resin is advised. • The more the proportion of hemp, the higher its creep and fatigue strength. • Mix with walnut shell filler has better dynamic performance than those without it. • Mix of 30% hemp fiber and 70% epoxy resin yields less dielectric value.

Specimens with binary (SL, FA) and ternary (T1, T2) concrete mixes (30.5 cm x 12.7 cm x 7.6 cm) were prepared without any chlorides, using a w/cm ratio of 0.41 or lower. Each specimen, reinforced with #3 rebar, have a 0.75 cm concrete cover. The specimens had reservoirs of varying lengths on their top surface. A 10% NaCl solution by weight was introduced into the reservoirs, and electromigration was applied for a period ranging from few weeks to several months to accelerate chloride transport. Corrosion current values were monitored for approximately 1600 days using galvanostatic pulse techniques and converted to mass loss using Faraday’s law. The SL mix specimens showed the highest average corrosion current values, followed by FA, T1, and T2 mix specimens. Despite the prolonged exposure, no visible corrosion such as cracks or surface-reaching corrosion products were observed over the monitoring period.

Optimization is the key to obtaining efficient utilization of resources in structural design. Due to the complex nature of truss systems, this study presents a method based on metaheuristic modelling that minimises structural weight under stress and frequency constraints. Two new algorithms, the Red Kite Optimization Algorithm (ROA) and Secretary Bird Optimization Algorithm (SBOA), are utilized on five benchmark trusses with 10, 18, 37, 72, and 200-bar trusses. Both algorithms are evaluated against benchmarks in the literature. The results indicate that SBOA always reaches a lighter optimal. Designs with reducing structural weight ranging from 0.02% to 0.15% compared to ROA, and up to 6%–8% as compared to conventional algorithms. In addition, SBOA can achieve 15%–20% faster convergence speed and 10%–18% reduction in computational time with a smaller standard deviation over independent runs, which demonstrates its robustness and reliability. It is indicated that the adaptive exploration mechanism of SBOA, especially its Levy flight–based search strategy, can obviously improve optimization performance for low- and high-dimensional trusses. The research has implications in the context of promoting bio-inspired optimization techniques by demonstrating the viability of SBOA, a reliable model for large-scale structural design that provides significant enhancements in performance and convergence behavior.

This study investigates the structural resilience of skylights integrated with perforated aluminum panels (Mashrabiya) in healthcare facilities under arid climate conditions. This study is based on a real, constructed skylight system installed in a healthcare facility, offering a rare field-based assessment rather than a purely theoretical or simulated model. The motivation behind this work stems from the need for energy-efficient, structurally robust, and aesthetically appealing skylight systems capable of withstanding extreme wind loads. Unlike conventional skylight studies, this research introduces a comprehensive numerical modelling approach in SAP2000 that incorporates nonlinear geometric effects, second-order buckling analysis, and stress concentration factors for perforated panels. The study evaluates deflections, stresses, and demand-to-capacity ratios (DCRs) under a wind load of 1.2 kPa, in accordance with local Standards. A key novelty of this study is the integration of perforated panels as both aesthetic and functional elements, enhancing structural performance by dissipating wind-induced stresses. The results indicate that the maximum DCR is 0.46, ensuring a significant safety margin, while the perforated panels exhibit a maximum stress of 41.05 MPa, well below the allowable limit of 160 MPa. Additionally, a mesh sensitivity analysis was conducted to optimise computational accuracy while balancing efficiency. The perforated aluminium panels (Mashrabiya) serve a dual function, enhancing aesthetics and acting as structural elements to redistribute and dissipate wind-induced stresses, a novel approach in skylight system design. This research advances wind-resistant façade design in arid climates by offering practical recommendations for optimising skylight configurations. Future work should focus on experimental validation through wind testing, real-time load monitoring, and parametric studies to support further structural optimisation. Unlike prior parametric or purely analytical works, this paper is based on a field‑installed skylight structure, using full‑scale geometry and validated against site‑measured boundary conditions. The study bridges the gap between experimental contexts and numerical modelling, thereby offering both practical and methodological novelty.

Abstract This study developed machine learning models to predict the swelling pressure (P S ) and swelling potential (SP) of fibre-reinforced expansive soils using a compiled dataset of 187 experimental specimens collected from published literature. The input variables included fibre characteristics (length, aspect ratio, content, tensile strength), soil consistency and compaction parameters (liquid limit, plasticity index, fine content, maximum dry density, optimum moisture content), and swelling characteristics of untreated soil (P S0 and SP 0 ). Tree-based gradient boosting models and artificial neural networks were implemented and evaluated using an 80% training and 20% testing data split. Model performance was assessed using determination coefficient (R²), root mean square error (RMSE), mean absolute error, and additional statistical indices. The gradient boosting model achieved the highest predictive accuracy, with testing R² values of 0.974 and 0.972 and RMSE values of 14.2 kPa and 0.82% for P S and SP, respectively. Feature importance analysis showed that untreated soil swelling parameters (P S0 and SP 0 ) were the dominant predictors governing swelling behaviour after fibre reinforcement. The results demonstrated that machine learning models, particularly gradient boosting methods, provided reliable and accurate prediction of swelling characteristics, enabling efficient evaluation and design of fibre-reinforced expansive soils.