Explore the vast range of titles available.

In steel structural engineering, artificial intelligence (AI) and machine learning (ML) are improving accuracy, efficiency, and automation. This review explores AI-driven approaches, emphasizing how AI models improve predictive capabilities, optimize performance, and reduce computational costs compared to traditional methods. Inverse Machine Learning (IML) is a major focus since it helps engineers to minimize reliance on iterative trial-and-error by allowing them to identify ideal material properties and geometric configurations depending on predefined performance targets. Unlike conventional ML models that focus mostly on forward predictions, IML helps data-driven design generation, enabling more adaptive engineering solutions. Furthermore, underlined is Explainable Artificial Intelligence (XAI), which enhances model transparency, interpretability, and trust of AI. The paper categorizes AI applications in steel construction based on their impact on design automation, structural health monitoring, failure prediction and performance evaluation throughout research from 1990 to 2025. The review explores challenges such as data limitations, model generalization, engineering reliability, and the need for physics-informed learning while examining AI’s role in bridging research and real-world structural applications. By integrating AI into structural engineering, this work supports the adoption of ML, IML, and XAI in structural analysis and design, paving the way for more reliable and interpretable engineering practices.

Composite modular panels are increasingly used in modern buildings, yet their layered behavior makes mechanical characterization and modeling difficult. This study presents a novel hybrid framework that integrates analytical, numerical, and AI-driven approaches for the mechanical characterization of composite panels. The system combines a layered concrete configuration with embedded steel reinforcement, and its performance was evaluated through experimental testing, analytical formulation, finite element simulations, and artificial intelligence techniques. Full-scale bending and shear tests were conducted and results in terms of displacements were compared with in silico simulations. The equivalent elastic modulus and thickness were suggested via a closed-form analytical procedure and validated numerically, showing less than 3% deviation from experiments. These equivalent parameters were used to simulate the dynamic response of a two-storey prototype building under harmonic excitation, with simulated modal periods differing by less than 10% from experimental data. To generalize the method, a parametric dataset of 218 panel configurations was generated by varying material and geometric properties. Machine learning models including Artificial Neural Network, Random Forest, Gradient Boosting, and Extra Trees were trained on this dataset, achieving R2 > 0.98 for both targets. A graphical user interface was developed to integrate the trained models into an engineering tool for fast prediction of equivalent properties. The proposed methodology provides a unified and computationally efficient approach that combines physical accuracy with practical usability, enabling rapid design and optimization of composite panel structures.

The present work proposes a comprehensive study of the safety evaluation of a reinforced concrete bridge in terms of the actual and expected degradation status. A bridge constructed in the early 1930s over the Cassibile River in Sicily, Italy, is selected for consideration. The study starts with a safety assessment based on the original design codes in use at the time of construction and then follows, by an exegetical approach, the evolution over time of the prescribed loading and design rules, both under the hypothesis of the undamaged and degraded structure. Therefore, the study examines the effects of increased traffic loads with the evolution of the code rules and, specifically, as defined by the Italian Technical Standards (NTC18), on a theoretically undamaged bridge. The results show a reduction in the safety factor across all critical structural components, apart from the lateral beams. These latter beams, in fact, benefit from a more comprehensive consideration of the overall resistance of the section, rather than just the localized stress values. Notably, the transition from the stress-based approach (Allowable Stresses Method, ASM) to a capacity-based evaluation of the full cross-section, as prescribed by NTC18, has minimal impact on elements subjected to purely axial loads, such as the hangers. Overall, the study aims to contribute to the understanding of the expected behaviour of an old structure facing the evolution of acting loads and allows the authors to restate, among other facts, the need to consider the expected degradation phenomena beyond the simple visual findings from surveys.

The goal of paper is the development and demonstration of efficiency of algorithm for form finding of a slack cable notwithstanding of the initial position chosen. This algorithm is based on product of two sets of coefficients, which restrict the rate of looking for cable geometry changes at each iteration. The first set restricts the maximum allowable change of absolute values of positions, angles and axial forces. The second set takes into account whether the process is the converging one (the signs of maximal change of parameters remain the same), so that it increases the allowable changes; or it is a diverging one, so that these changes are discarded. The proposed procedure is applied to two different methods of simple slack cable calculation under a number of concentrated forces. The first one is a typical finite element method, with the cable considered as consisting of number of straight elements, with unknown positions of their ends, and it is essentially an absolute coordinate method. The second method is a typical Irvine’s like analytical solution, which presents only two unknowns at the initial point of the cable; due to the peculiarity of implementation it is named here a shooting method. Convergence process is investigated for both solutions for arbitrary chosen, even very illogical initial positions for the ACM, and for angle and force at the left end for SM as well. Even if both methods provide the same correct convergent results, it is found that the ACM requires a much lower number of iterations.

Metallic thin-walled beams with continuously varying cross-sections loaded in compression are particularly sensitive to instability problems due to lateral-torsional buckling. Such a phenomenon depends on several parameters, including the cross-sectional properties along the entire length, material properties, load distribution, support, and restraint conditions. Due to the difficulty of obtaining analytic solutions for the problem under consideration, the present study takes a numerical approach based on a variational formulation of the lateral-torsional buckling problem of tapered C-beams. Numerical simulations are compared with experimental results on the buckling of a physical model of at thin-walled beam with uniformly varying cross-section, with the aim of assessing the accuracy of the proposed approach. The good agreement between numerical and experimental results and the reduced computational effort highlight that the proposed variational approach is a powerful tool, provided that the geometry of the structure and the boundary conditions are accurately modeled.

Stainless-steel provides substantial advantages for structural uses, though its upfront cost is notably high. Consequently, it's vital to establish safe and economically viable design practices that enhance material utilization. Such development relies on a thorough understanding of the mechanical properties of structural components, particularly connections. This research advances the field by investigating the behavior of stainless-steel connections through the use of a four-parameter fitting technique and explainable artificial intelligence methods. Training was conducted on eight different machine learning algorithms, namely, Decision Tree, Random Forest, K-nearest neighbors, Gradient Boosting, Extreme Gradient Boosting, Light Gradient Boosting, Adaptive Boosting, and Categorical Boosting. SHapley Additive Explanations was applied to interpret model predictions, highlighting features like spacing between bolts in tension and end-plate height as highly impactful on the initial rotational stiffness and plastic moment resistance. Results showed that Extreme Gradient Boosting achieved a coefficient of determination score of 0.99 for initial stiffness and plastic moment resistance, while Gradient Boosting model had similar performance with maximum moment resistance and ultimate rotation. A user-friendly graphical user interface (GUI) was also developed, allowing engineers to input parameters and get rapid moment-rotation predictions. This framework offers a data-driven, interpretable alternative to conventional methods, supporting future design recommendations for stainless-steel beam-to-column connections.

The collapse factor is a significant parameter in the framework of the safety assessment and economical design of ductile structures. This fact draws attention to the necessity of a careful assessment of the limit analysis approaches. The kinematics in these structures arises in fact from the actual rotation of the plastic hinges under axial force and bending moment. It can be shown that it is possible to obtain a reliable tool capable of competing with computationally expensive methodologies. The application of the methods of limit analysis involves a simplified and idealised model of the structure and, notwithstanding the fact that hundreds of papers have been devoted to the topic, some consequences of apparently unimportant simplifications still seem to have not been properly and firmly highlighted. This paper investigates the ultimate load and collapse modes of steel frames under combined vertical and horizontal forces through limit analysis.

In recent years, considerable attention has been paid to research and development methods able to assess the seismic energy propagation on the territory. The seismic energy propagation is strongly related to the complexity of the source and it is affected by the attenuation and the scattering effects along the path. Thus, the effect of the earthquake is the result of a complex interaction between the signal emitted by the source and the propagation effects. The purpose of this work is to develop a methodology able to reproduce the propagation law of seismic energy, hypothesizing the \"transmission\" mechanisms that preside over the distribution of seismic effects on the territory, by means of a structural optimization process with a predetermined energy distribution. Briefly, the approach, based on a deterministic physical model, determines an objective correction of the detected distributions of seismic intensity on the soil, forcing the compatibility of the observed data with the physical-mechanical model. It is based on two hypotheses: (1) the earthquake at the epicentre is simulated by means of a system of distortions split into three parameters; (2) the intensity is considered coincident to the density of elastic energy. The optimal distribution of the beams stiffness is achieved, by reducing the difference between the values of intensity distribution computed on the mesh and those observed during four regional events historically reported concerning the Campania region (Italy).