Concrete-filled steel tubes (CFST) are widely utilized in civil engineering due to their superior load-bearing capacity, ductility, and seismic resistance. However, existing design codes, such as AISC and Eurocode 4, tend to be excessively conservative as they fail to account for the composite action between the steel tube and the concrete core. To address this limitation, this study proposes a hybrid model that integrates XGBoost with the Pied Kingfisher Optimizer (PKO), a nature-inspired algorithm, to enhance the accuracy of shear strength prediction for CFST columns. Additionally, quantile regression is employed to construct prediction intervals for the ultimate shear force, while the Asymmetric Squared Error Loss (ASEL) function is incorporated to mitigate overestimation errors. The computational results demonstrate that the PKO-XGBoost model delivers superior predictive accuracy, achieving a Mean Absolute Percentage Error (MAPE) of 4.431% and R² of 0.9925 on the test set. Furthermore, the ASEL-PKO-XGBoost model substantially reduces overestimation errors to 28.26%, with negligible impact on predictive performance. Additionally, based on the Genetic Algorithm (GA) and existing equation models, a strength equation model is developed, achieving markedly higher accuracy than existing models (R² = 0.934). Lastly, web-based Graphical User Interfaces (GUIs) were developed to enable real-time prediction.

Data-driven research on recycled aggregate concrete (RAC) has long faced the challenge of lacking a unified testing standard dataset, hindering accurate model evaluation and trust in predictive outcomes. This paper reviews critical parameters influencing mechanical properties in 35 RAC studies, compiles four datasets encompassing these parameters, and compiles the performance and key findings of 77 published data-driven models. Baseline capability tests are conducted on the nine most used models. The paper also outlines advanced methodological frameworks for future RAC research, examining the principles and challenges of physics-informed neural networks (PINNs) and generative adversarial networks (GANs), and employs SHAP and PDP tools to interpret model behaviour and enhance transparency. Findings indicate a clear trend toward integrated systems, hybrid models, and advanced optimization strategies, with integrated tree-based models showing superior performance across various prediction tasks. Based on this comprehensive review, we offer a recommendation for future research on how AI can be effectively oriented in RAC studies to support practical deployment and build confidence in data-driven approaches.