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.

Structural steel reuse can gain large embodied-carbon savings, yet it is still not widely adopted since approval depends on the quality of the evidence, how uncertainty is handled, and if the design requirements are followed, not just on resistance. Reclaimed members frequently lack dependable documentation regarding material grade, loading history, boundary conditions, connection status, and degradation. For reuse decisions, conservative default assumptions protect safety but frequently eliminate qualified reuse options. This research examines data-driven and physics-informed computational methods from a decision-making standpoint, contending that their significance resides in facilitating an auditable approval process, not in supplanting deterministic verification. We differentiate feasibility, acceptability, and approval as distinct engineering phases. Data-driven models are thought of as tools for quickly screening candidates, surrogate evaluation, inverse reasoning, and stock-to-demand matching. Their goal is to reduce the list of candidates and prioritize evidence collection. Physics-informed approaches are examined as admissibility filters that impose restrictions of equilibrium, compatibility, stability, and plausible boundary-condition envelopes; therefore, minimizing mechanically invalid predictions under partial information. Next, we consider uncertainty quantification and explainability to be essential for reuse decisions. We suggest practical outputs for approval packages, such as resistance bounds within specified assumption envelopes, sensitivity rankings of decision-critical unknowns, low-support flags, and evidence actions for conditional acceptance. This document is organized into a process from audit to approval. It also states the open issues in reuse-specific datasets, standardized evidence capturing, decision-relevant validation under degradation, and regulatory acceptance. The resulting framework clarifies how advanced computational tools can enable adaptable, conservative, and transparent steel reuse in practice.