Overconsolidation ratio (OCR) is an important geotechnical parameter that plays a crucial role in the analysis and design of foundations and structures on clay deposits. In this study, five machine learning (ML) algorithms, including gradient boosting machine (GBM), random forest (RF), artificial neural network (ANN), support vector machine (SVM), and eXtreme gradient boosting (XGB) were developed to estimate the OCR of clays based on the piezocone penetration test data. The ‘GridSearchCV’ function from the Scikit-learn package was employed to perform hyper-parameter tuning and k-fold cross-validation, ensuring the best possible model performance. Vertical total stress (σ_v0), hydrostatic pore water pressure (u₀), corrected cone resistance (q_t), pore pressure elements at the cone tip (u₁), and above the cone base (u₂), along with the type of clay (intact or fissured) were selected as the main features of input data. Sensitivity analysis revealed that q_t was the most influential parameter for RF, GBM, XGB, and SVM predictions, while all inputs affected the ANN model. It was found that the SVM model delivered the lowest accuracy in predicting OCR. In contrast, the XGB model showed the best performance, while the remaining models achieved reliable results, each with a coefficient of determination of 0.90 or higher.

This study adopts a reliability-based optimization approach for the failure mechanism analysis and design of the retaining wall considering nonlinear soil backfills. The assumed failure mechanism is represented by rigid blocks within a kinematically admissible framework in a rotational coordinate system. Then the active and passive earth pressures are derived from the optimization procedure. A convenient way for incorporating seepage effects is proposed and implemented in the nonlinear upper bound analysis. Finally, a novel response surface method is employed to calculate the failure probability considering different probabilistic scenarios and distribution types with high calculation efficacy. The accuracy of the proposed method is evaluated using the Monte Carlo simulations with 1 million trials. Sensitivity analysis indicated that soil unit weight and initial cohesion are the critical factors dominating the failure probability of passive and active mechanism, respectively. The reliability-based design can be performed to obtain the safe range of the lateral force against nonlinear soil backfills with a target failure probability.

Site effects study has always been a key research topic in earthquake engineering. This study proposes a hybrid method to analyze large-scale three-dimensional sedimentary basin under Rayleigh (R) wave incidence. The proposed hybrid method includes two steps: 1) calculate the free field responses of layered sites subjected to R-wave using the frequency-wavenumber method; 2) Simulate the local site region using spectral element method with the equivalent forces input computed from the free field responses. A comprehensive verification study is conducted demonstrating the accuracy of this method. To investigate the effect of sedimentary basin on R-wave propagation, a parametric study is performed on the medium impedance contrast ratio of sedimentary basins and the incident seismic wave predominant frequency, revealing the scattering patterns of sedimentary basins under R-wave incidence. Finally, a practical case of the Wudu Basin in the Tibetan Plateau region of China is simulated. Results indicate significant amplification of R-wave by sedimentary basin, and the proposed hybrid method could serve as a reliable and efficient approach for large-scale R-wave propagation simulation.

This study investigates the horizontal load-bearing behavior of the Caisson-type Diaphragm Wall with Variable Cross-sections (CDWVC), a novel foundation system that integrates closed and open wall segments to improve performance and material efficiency. A series of 1g model tests was conducted using instrumented plexiglass models under controlled soil conditions to evaluate the bearing behavior under lateral loading. Finite difference simulations were also performed to complement the experimental findings and provide additional insights. The performance of CDWVCs with varying closed segment heights and open segment widths was analyzed. Results indicate that taller closed segments reduce horizontal displacements under equivalent loads and shift the rotation point closer to the surface, thereby improving overall stability. While increasing the height of the closed wall segment leads to modest improvements in horizontal ultimate bearing capacity, these gains are often outweighed by significant increases in material consumption. Conversely, expanding the width of the open wall segment results in a more substantial increase in horizontal ultimate bearing capacity relative to material usage, improving load transfer and overall stability. This design strategy achieves a favorable balance between load capacity and material efficiency compared to increasing closed wall height. The findings underscore the importance of design choices in the performance of CDWVCs under horizontal loading conditions.

The research investigates ensemble machine learning techniques to forecast high-performance concrete (HPC) compressive strength through analysis of Gradient Boosting Machines (GBM) together with Random Forest (RF) and Deep Neural Network (DNN) performances. Previous experiment data served as model inputs for the machine learning systems that comprised cement, fly ash, blast furnace slag, water, superplasticizer, coarse aggregate, and fine aggregate for HPC compressive strength prediction. The research study utilizes input parameters and direct bypassing of dimensionality reduction to evaluate the performance of models that capture intricate nonlinear patterns from concrete compressive strength data. RF produced the most accurate results during training by establishing 0.9650 R² measurements and 0.0798 RMSE indicators, thus demonstrating exceptional accuracy at a minimal error level. In testing, RF maintained its lead with an R² of 0.9399, followed closely by GBM, while DNN showed slightly higher error rates. A comprehensive ranking analysis across multiple statistical metrics highlighted RF as the most dependable concrete compressive strength prediction model. Further, Regression Error Characteristic (REC) curves visually assessed model performance relative to error tolerance, revealing RF and GBM’s reliable accuracy across different thresholds. A Graphical User Interface (GUI) with user-oriented features connected to the prediction models was created for smooth system usage. The results indicate that RF provides accurate predictions for concrete compressive strength because of the effectiveness of ML models, according to this study. Predictions of tensile strength, modulus of elasticity, and fracture energy parameters in concrete materials become possible when categorized based on their compressive strength values. This approach significantly enhances structural analysis by reducing both cost and time requirements.

Detection of cracks in concrete structures is critical for their safety and the sustainability of maintenance processes. Traditional inspection techniques are costly, time-consuming, and inefficient regarding human resources. Deep learning architectures have become more widespread in recent years by accelerating these processes and increasing their efficiency. Deep learning models (DLMs) stand out as an effective solution in crack detection due to their features such as end-to-end learning capability, model adaptation, and automatic learning processes. However, providing an optimal balance between model performance and computational efficiency of DLMs is a vital research topic. In this article, three different methods are proposed for detecting cracks in concrete structures. In the first method, a Separable Convolutional with Attention and Multi-layer Enhanced Fusion Network (SCAMEFNet) deep neural network, which has a deep architecture and can provide a balance between the depth of DLMs and model parameters, has been developed. This model was designed using a convolutional neural network, multi-head attention, and various fusion techniques. The second method proposes a modified vision transformer (ViT) model. A two-stage ensemble learning model, deep feature-based two-stage ensemble model (DFTSEM), is proposed in the third method. In this method, deep features and machine learning methods are used. The proposed approaches are evaluated using the Concrete Cracks Image Data set, which the authors collected and contains concrete cracks on building surfaces. The results show that the SCAMEFNet model achieved an accuracy rate of 98.83%, the ViT model 97.33%, and the DFTSEM model 99.00%. These findings show that the proposed techniques successfully detect surface cracks and deformations and can provide practical solutions to real-world problems. In addition, the developed methods can contribute as a tool for BIM platforms in smart cities for building health.

In the past years, shape memory alloys (SMAs) have found applications in numerous fields. The unique energy dissipation capacity of SMAs under cyclic loading and their notable super elasticity, along with their high reliability, make them an ideal candidate for use as an energy dissipation material and a passive vibration control solution. In vibration control implementation, because of the cyclic essence of applied dynamic forces, it is crucial to examine the mechanical behavior of SMAs under cyclic loading. Numerous researchers have dedicated their efforts to studying the cyclic behavior of superelastic SMAs, delving into the effect of various loading characteristics such as frequency and strain amplitude as well as the effect of specimen size and ambient temperature. While the primary focus of this paper is to review existing research on the mechanical behavior of NiTi SMA under strain-controlled cyclic loading, it also includes an overview of the phase transformation mechanism, which manifests itself as shape memory effects and superelasticity. This additional information is provided to deepen the understanding of SMA material behavior. The perspective and data presented in this paper are tailored to appeal to civil engineers, with a specific emphasis on the information of interest in this field.

To investigate the long-term performance of a 32 m prestressed simply supported box girder, a 1:4 scale prestressed concrete simple supported box girder was cast. The casting procedure adheres to the principle of stress equivalence within the concrete in the middle span after tensioning of prestressed tendons. Utilizing the CEB-FIP 90 model as a foundation, we established a long-term deformation calculation model for the box girder. Subsequently, the reliability of the long-term deformation model was confirmed by employing data from a 96 d long-term deformation test conducted on the box girder. Meanwhile, a new database was created by integrating shrinkage and creep experiment data with the shrinkage and creep database developed by Bazant. The shrinkage and creep uncertainty coefficients were introduced to complete the modeling of concrete shrinkage creep uncertainty calculations. The results demonstrate that the long-term deformation prediction model can effectively characterize the tendency of the mid-span upward deflection in the box girder. At 988 d, the upward deflection at the mid-span of the 1:4 scale model was expected to reach approximately 3.67 mm. It is worth noting that the CEB-FIP 90 model tends to slightly overestimate long-term deformation compared with experimental results. Additionally, it significantly underestimates the shrinkage strain observed in the test results. The uncertainty associated with the long-term deformation prediction of the structural system increased as the prediction time extended.

Accidental surcharge, a type of uncertain manmade hazard, poses a huge threat to the safe operation of shield tunnels. In this regard, a vulnerability assessment framework is proposed in this paper to evaluate the damage state of a shield tunnel subjected to sudden extreme surcharges, accounting for the effect of soil uncertainty and tunnel burial depths. A two-dimensional numerical model of the shield tunnel in soft soil under surcharge loading is established and verified by the field monitoring data. Then, joint opening and horizontal convergence of the shield tunnel are chosen as damage indices, and the corresponding fragility curves and vulnerability curves are established based on the Monte Carlo calculation. The influences of surcharge loading and buried depths on the vulnerability are discussed. Finally, the proposed vulnerability assessment framework is applied in a real case in Shanghai to make a quick judgement on the dangerous sections of shield tunnels. The research results show that the vulnerability of shield tunnels increases with the surcharge loading. Deep shield tunnels have higher initial vulnerability but are not sensitive to surcharge loading. The study sheds light on the robust design, post-hazard decision-making, and rapid risk identification for shield tunnels subjected to surcharge loads.

The increasing focus on health and hygiene has expanded the need for protective measures on material surfaces. In this regard, developing antibacterial concrete and mortar capable of eliminating viruses and bacteria is crucial. However, a key challenge in cementitious systems is the inability to maintain long-term antibacterial effectiveness when titanium dioxide (TiO₂) is used as the sole photocatalyst. To address this limitation, this study aimed to enhance the antibacterial properties of TiO₂ by modifying it with silver (Ag) using a planetary ball mill. Concrete and mortar samples incorporating the modified material were produced, and their antibacterial performance was evaluated over both short and long durations. So the originality of this study was to evaluate the performance of cementitious system surfaces against repeated bacterial attacks using a specific mechanical alloying method in the modification of TiO₂ with Ag. Additionally, the modified products were characterized through X-ray diffraction (XRD), fourier transformed infrared spectroscopy (FTIR), scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) imaging, grain size analysis, and band gap energy measurements. The impact of the components on antibacterial efficiency was statistically analyzed using analysis of covariance (ANCOVA). The results demonstrated that Ag-containing samples achieved a 100% bacterial killing rate in all experimental replicates. These findings confirm that Ag-TiO₂ alloying was successfully achieved via planetary ball milling, providing concrete with sustained antibacterial properties in both early and long-term applications.

This study enhances the application of cross-sectional warping considered mixed finite element (W-MFE) formulation to accurately determine natural vibration, static displacement response, and shear and normal stress evaluation with very close to the precision of solid finite elements (FEs) in two-phase/multi-phase functionally graded (FG) laminated composite beams strength using carbon nanotubes (CNTs). The principles of three dimensional (3D) elasticity theory are used to derive constitutive equations. The mixed finite element (MFE) method is improved by accounting for warping effects by displacement-based FEs within the cross-sectional domain. The MFE with two nodes has a total of 24 degrees of freedom. The two-phase material consists of a polymer matrix reinforced with aligned CNTs that are FG throughout the beam thickness. The multi-phase FG beam is modeled as a three-component composite material, consisting of CNTs, a polymer matrix, and fibers. The polymer matrix is reinforced by longitudinally aligned fibers and randomly dispersed CNT particles. The fiber volume fractions are considered to change gradually through the thickness of the beam following a power-law variation. The W-MFE achieves satisfactory results with fewer degrees of freedom than 3D solid FEs. Benchmark examples examine the effects of ply orientation, configuration, and fiber gradation on FG beam behavior.

Earthwork productivity analysis is essential for successful construction projects. If productivity analysis results can be accessed anytime and anywhere, then project management can be performed more efficiently. To this end, this paper proposes an earthwork productivity monitoring framework via a real-time scene updating multi-vision platform. The framework consists of four main processes: 1) site-optimized database development; 2) real-time monitoring model updating; 3) multi-vision productivity monitoring; and 4) web-based monitoring platform for Internet-connected devices. The experimental results demonstrated satisfactory performance, with an average macro F1-score of 87.3% for continuous site-specific monitoring, an average accuracy of 86.2% for activity recognition, and the successful operation of multi-vision productivity monitoring through a web-based platform in real time. The findings can contribute to supporting site managers to understand real-time earthmoving operations while achieving better construction project and information management.

To explore the applicability of three-dimensional (3D) peridynamics (PD) in complex stress, 36 reinforced concrete (RC) beams without web reinforcement were designed and tested, and investigated the effects of shear span ratio, longitudinal reinforcement, and flange width on the shear strength of beams. A 3D discretization model of all specimens in the test was established, and the specimens were simulated using the PD method. To consider the heterogeneity of concrete, a non-homogeneous PD model considering aggregate size was established, and the simulation results were compared with the original model. The results indicate that the shear span ratio, longitudinal reinforcement, and flange width have a significant impact on the shear strength of RC beams without web reinforcement. The 3D PD model has a good applicability for RC beams under complex stress. Without considering computational costs, heterogeneous models can obtain more accurate results than homogeneous models and better reflect the process of concrete beam failure.

This study addresses the application of advanced boosting-based ensemble machine learning techniques such as extreme gradient boosting (XGBoost), random forest (RF), category-aware gradient boosting (CATBoost), and adaptive boosting (ADABoost) algorithms to study the bond behavior of fiber-reinforced polymer (FRP) bars in reinforced concrete (RC) beams. To forecast the peak load (P_max) of the bond behavior between the FRP bars and concrete, five total input variables, namely, the elastic modulus of the bar (E_f), the tensile strength of the bar (F_f), the compressive strength of the concrete (f_{\mathrm{c}}^{\prime }), the diameter of the bar (d_{\mathrm{b}}), and the bar embedment length (l_{\mathrm{b}}), were selected for machine learning model construction. The accuracy of the constructed predictive machine learning models was compared using several metric performances. However, rank analysis has also been used to ascertain which models perform the best. According to the findings of rank analysis using several metric performances, XGBoost outperformed RF, ADABoost, and CATBoost. Utilizing the developed advanced machine learning methods to examine the bond behavior of FRP bars in RC beams yields tangible advantages for the construction sector. This approach refines the design precision, minimizes expenses, and elevates the overall effectiveness and longevity of structures reinforced with FRP.

This study presents a multi-scale deep-learning framework that integrates several advanced neural models to optimize hybrid three dimensional (3D) printed self-sensing nano-carbon cementitious composites. The first step was undertaken by Multi-Scale Graph Neural Network, where special conductive pathways were taught ensuring the uniform work on nano-carbon learning patterns, improving electrical conductivity by 25%–35%. four-dimensional Spatiotemporal Transformer Network decoded printing parameters achievements with an interlayer conductivity improvement of 40%–50%, avoiding anisotropic print by aiming for defects prediction on print Induced anisotropic behavior. High-fidelity artificial microstructures have been generated with Physics Informed Generative Adversarial Networks; these synthetic methods realize an experimental cost-cutting of about 50% while conserving about 98% fidelity to the characteristics of real microstructures. Fifth, Self-Supervised Contrastive Learning automatically classifies small macro and microdefects with over 95% detection reliability. There has been reduction of as much as 35% in the number of false positives. Predicted kinetics of hydration and long-term electrical stability can now be predicted with speed improvements of 15% and resistance drift reduction by 20% over six months. This approach for the first time combines different hybrid models of deep learning for the self-sensing cementitious composites, thus significantly increasing percolation of electrical networks, accuracy in crack detection, and predictions on long-term durability. The developed framework creates a new paradigm in the real-time structural health monitoring world, providing enhanced reliability in structures while also reducing costs at a level for the next generation of smart infrastructure sets.