The shear behaviour of Ultra High-Performance Fiber Reinforced Concrete (UHPFRC) is complex due to fiber contributions alongside conventional reinforcement and concrete effects. This paper presents a novel Machine Learning (ML) approach to predict the shear capacity of UHPFRC beams, offering an alternative to conventional empirical and analytical methods. A database of 282 experimental data points was used to train multiple ML models, including Support Vector Regression (SVR), k-Nearest Neighbour (kNN), Decision Tree Regressor (DTR), Random Forest Regressor (RFR), Artificial Neural Network (ANN), Gradient Boosting Regressor (GBR), Light Gradient Boosting Regressor (LGBR), and Extreme Gradient Boosting Regressor (XGBR). Among these, boosting-based models performed best, with XGBR achieving the highest accuracy with an R 2 score of 0.91. Further, a feature importance analysis identified the shear span-to-depth ratio as the most influential parameter on the shear capacity, with longitudinal and transverse reinforcement ratios also playing key roles. A SHapley Additive exPlanation (SHAP) analysis was conducted, offering an interpretable framework for understanding the contribution of each feature to the model’s prediction. This analysis further clarified feature contributions, identifying optimal parameter ranges for shear capacity. This data-driven approach bridges empirical methods and computational modelling enhancing confidence in ML-based predictions and guiding efficient UHPFRC design.

As a typical representative of infrastructure, tunnels are indispensable carriers for the normal operation of cities, with their safe and efficient operation directly influencing urban efficiency. However, the various data supporting tunnel operation and maintenance (O&M) exhibit significant diverse sources and structural differences, which pose substantial challenges to tasks such as tunnel structural health assessment. To address these challenges, this paper proposes an ontology-based multi-source heterogeneous O&M data integration framework to support the assessment of tunnel structural health, thereby improving decision-making efficiency in tunnel maintenance. The framework consists of four layers: data layer, ontology layer, mapping layer, and application layer, enabling the unified modeling, integration, and comprehensive application of multi-source heterogeneous tunnel O&M data. Additionally, the proposed framework is applied to a practical engineering project, the Tanglang Mountain Tunnel. Compared with existing methods, the framework demonstrates improvements in data fusion accuracy, data completeness, and operational efficiency.

Wall piers are widely used to enhance lateral stability in bridges with tall piers and relatively narrow decks. For this type of structure, the longitudinal direction of the bridge is commonly acknowledged as the governing direction for seismic performance, with wall piers serving as the seismic critical members. However, progressive scouring reduces foundation strength and stiffness, leading to increased transverse seismic deformation. This deformation amplifies the risk of pile damage and may shift the seismic critical member to the foundation. This study investigates the seismic performance of wall pier bents in bridges, explicitly focusing on the effects of riverbed scouring. Through a Taiwan-based case study, seismic performance is evaluated at various scour depths, identifying the seismic critical member through capacity spectrum analysis and the peak ground acceleration corresponding to the performance limit of the wall pier bent. The findings highlight that seismic performance is frequently controlled by the transverse direction, emphasizing the foundation as the seismic critical member. The effect of employing foundation strengthening as a retrofit strategy is also assessed, revealing that it provides only limited improvement in seismic performance. Even after retrofit, the seismic performance of wall pier bents remains primarily governed by the pile foundation.

The load-bearing mechanism in reinforced concrete beams with corroded rebar has not been clearly elucidated, impeding good shear strength predictions. A thorough literature review reveals the impact of rebar corrosion on the load-bearing mechanism of beams. A systematic validation confirms deficiencies in the current prediction approaches such as the failure to reflect changes in the load-bearing mechanism caused by rebar corrosion and insufficient consideration of the effects of key variables. Corrosion impact factors that reflect the enhancement of arch-action caused by main bar corrosion and the section loss of corroded stirrups are derived to better account for changes in the shear strength. The superiority of incorporating these impact factors into fundamental methods for intact beams is confirmed using ample published experimental data. Incorporating corrosion impact factors into Niwa’s formula is shown to yield the best predictions.

This study explored effective techniques for segmenting bridge-point clouds to facilitate the automated creation of 3D bridge models. By employing a projection-based segmentation method, superior mean Intersection over Union (mIoU) and F1-score were achieved compared with the point-based method, PointNet++, across a bridge point-cloud dataset. A simple but powerful filter was introduced that suppresses the transparency effect in the point-cloud projection process to minimize misclassifications during training. Additionally, for image semantic segmentation, employing Axial-DeepLab and an architecture that separates the images into top and bottom views and training from pretrained weights yielded the highest mIoU, improving by 0.4031 over the baseline mIoU and outperforming the projection-based method. The results confirmed the potential of projection-based methods for efficiently segmenting bridge-point clouds, thereby assisting the automated creation of 3D models. These models can be utilized for inspection and maintenance, facilitating effective management and understanding of the structural state of bridges during operation.

The development of appropriate inspection and maintenance strategies is essential for the operational safety of metro tunnels. This study aimed to propose a calculation method to characterize the structural performance degradation of metro tunnels by considering the non-stationary deterioration process, structural resilience, and the impact of adjacent construction. Accordingly, a method for calculating the life-cycle maintenance cost of tunnels considering resilience was devised using the life-cycle cost as an evaluation index. The inspection and maintenance parameters in the metro tunnel were further optimized based on this method. The study shows that the cost comparison of inspection parameters leads to an optimum interval for the inspection cycle, depending on the inspection accuracy. A comparison of the parameter disassembly and cost-mapping models for the three types of maintenance tools shows that there are different suitable intervention times for aggressive and conservative maintenance tools during the deterioration of metro tunnels. By analyzing the ductility parameters, it is concluded that considering ductility allows the deterioration model to reflect the actual deformation of tunnel structures better. The methodology and conclusions of this study will help practitioners develop more effective maintenance strategies.

Bond between pre-stressing tendons and concrete is an important characteristic which allows the stresses to transfer from steel to concrete. Corrosion-induced expansion stresses the concrete, leading to cracking, bond reduction, and structural safety concerns. In this paper, a life cycle analysis of a real case study, Ynys-y-Gwas bridge, has been performed considering the effect of bond deterioration due to chloride attack. The bridge has been modeled and analyzed using ABAQUS software. A numerical method is developed to assess bond deterioration from uniform and localized corrosion. Connector elements are used to simulate the interface layer between the pre-stressing steel and concrete. The behavior of the connector elements were considered according to the CEB mathematical bond-slip relationship.The results show that the connector elements successfully simulate bond deterioration since there was a good agreement between the mathematical input and output behavior. The effect of bond degradation associated with localized and generalized corrosion is analyzed and its effect is very significant in case of uniform corrosion. The probabilistic non-linear method that is used in this study can be successfully used in the assessment of the existing segmental post-tensioned bridges, since they need for a regular inspection due to their risk to corrosion.

Construction, damage and conservation are intertwined in the life of Milan Cathedral. The evolution of the structural system and the restorations carried out throughout the centuries are described with a particular focus on the case study of the tiburio, with the supporting pillars; these underwent dramatic damage in the XX century, close to collapse, followed by repair and strengthening interventions and the rebirth of the Cathedral. This same part of the monument is currently the object of the latest ongoing restoration interventions, related to service state conditions. The organization of the Veneranda Fabbrica del Duomo di Milano is outlined by the worksite management. The inspection and maintenance are set in relation to the technical activities of other institutions. The aim of all operations for the preservation of the structural safety and cultural identity of the monument up to the present day, is discussed. A discussion is provided for monumental heritage buildings, encompassing structural assessment, reliability evaluation, damage survey, maintenance and repair operations, in relation to the underlying social values.

The study presents a reinforcement learning (RL) method called the Q-learning algorithm to determine the best maintenance policy for equipment. This involves an artificial intelligence agent making decisions and an environment representing the equipment. The agent creates maintenance policies and takes actions, while the environment determines state transitions and rewards based on the actions chosen. The optimization of the maintenance policy starts with predicting equipment reliability using sensor data. This prediction method combines Back Propagation Neural Network (BPNN) algorithms with the Boxing Match Algorithm (BMA), an evolutionary meta-heuristic algorithm. A novel weight update strategy enhances the performance of artificial neural networks in reliability prediction. This integrated model, BMA-BPNN, aims to improve the accuracy of forecasted reliability. The study involves forecasting equipment reliability to determine critical levels, incorporating cost and risk considerations into decision-making for optimizing maintenance policies. As a result, the Q-learning algorithm is used to identify the best maintenance actions based on equipment reliability. Implementing an automated maintenance system that considers equipment reliability and costs can help reduce accidents resulting from maintenance program deficiencies. This study thus contributes to the field by providing an efficient approach to equipment maintenance.

This paper explores the behaviour of cable-stayed bridges (CSBs) when subjected to intentional explosion events. An example CSB was proposed and analysed using a three-dimensional (3D) finite-element (FE) model; a nonlinear time-history analysis taking into account the geometric and material nonlinearities was conducted. Six blast scenarios were considered; the explosion took place on the bridge deck. Response quantities, such as forces in cables, movements of deck and pylons, states of stress in main girders, and reactions on bearings and pylon foundations were examined. The critical blast scenario for each of these quantities was identified. Some design measures for mitigating cable failure were proposed and examined. The results of analyses showed that considering several blast scenarios for the design and evaluation of CSBs is mandatory. Non-symmetric (relative to the bridge longitudinal axis) blast loading scenarios are generally more critical than symmetric ones. Increasing the area of some long cables in the main span, and increasing spacing of the cables nearby the pylon together with strengthening the main girder nearby the pylon could be useful measures to reduce failure potential. This study may help better understand the response of CSBs to extreme events.