Accurate prediction of the residual drift ratio of reinforced concrete bridge piers is challenging because conventional methods are computationally expensive, time-consuming, and unable to effectively capture complex nonlinear interactions among multiple influencing factors. To address these limitations, this study proposes an interpretable machine learning framework for predicting the residual drift ratio of reinforced concrete bridge piers. A comprehensive database containing 261 quasi-static experimental datasets was established, incorporating key structural and material parameters, including axial com-pression ratio, shear span ratio, stirrup ratio, longitudinal reinforcement ratio, material strengths, and geometric dimensions. Based on this database, six representative machine learning models were developed and systematically compared. Their predictive performance, robustness, and generalization capability were evaluated using multiple statistical metrics and Monte Carlo simulations. The results show that the CatBoost model consistently outperformed the other models, achieving an R2 value of 0.9629 on the test set while maintaining excellent stability under random data partitions. Furthermore, SHAP analysis was employed to interpret the trained model and quantify the contributions of individual input variables. Eight key factors influencing the residual drift ratio were identified, with the loading displacement ratio (θ) exhibiting the greatest influence. These findings demonstrate that the proposed framework provides an accurate, reliable, and interpretable tool for predicting the post-earthquake residual drift ratio of reinforced concrete bridge piers, offering valuable support for performance-based seismic design, post-earthquake damage assessment, and resilience-based bridge engineering.
A comprehensive data-driven framework integrating ensemble machine learning models with systematic hyperparameter sensitivity analysis and explainable artificial intelligence techniques is proposed, demonstrating that the XGB model significantly outperforms the other approaches, achieving superior accuracy and robust generalization.
Qaim Shah, Waheed Ali Khoso, Fawad Iqbal et al.· Discover Artificial Intellig...· 0 citations
Accurate prediction of shear capacity in reinforced concrete beams is crucial for structural safety assessment. Conventional theoretical methods exhibit significant variability due to the complexity of shear failure mechanisms. This study presents an interpretable machine learning (ML) framework to enhance shear capacity prediction. A comprehensive database of 1175 beam specimens was developed, including normal concrete (NC) and ultra-high-performance concrete (UHPC) beams across three distinct cross-sectional geometries. The ML algorithms–support vector regression, artificial neural network, K-Nearest neighbors, decision tree, random forest, gradient boosting machine, light gradient boosting machine, adaptive boosting, categorical boosting, and extreme gradient boosting (XGBoost)–were optimized using 10-fold cross-validation and random search. The XGBoost algorithm demonstrated superior performance, achieving an R2 of 0.986 on the aggregated data set. Interpretability analysis with Shapley additive explanations identified beam depth (h), shear-span ratio (m), cross-sectional area (Ac) and fibre factor (λf) as critical features, highlighting their individual and interactive contributions. Moreover, a unified ML-based shear strength prediction model was developed that simultaneously captures the shear behaviour of both NC and UHPC beams, incorporating physically meaningful input features derived from the data set, thereby overcoming the limitations of separate empirical formulations. The proposed ML-based model significantly improved the accuracy of shear strength predictions compared to traditional empirical methods, enhancing reliability in structural design.
Qizhi Xu, Yan Tang, Shimin Ding et al.· Proceedings of the Instituti...· 0 citations
The proposed AFELA–machine learning framework provides a computationally efficient, reliable, and interpretable approach for tunnel stability assessment in sloping rock mass conditions.
A. Kumar, V. Chauhan, Aayush Kumar et al.· Transportation Infrastructur...· 1 citation
An interpretable and uncertainty-aware machine-learning framework for estimating the shear capacity of FRCM-strengthened beams enables accurate, transparent, and uncertainty-aware assessment of shear capacity in FRCM-strengthened concrete beams.
Xiangsheng Liu, G. Figueredo, G. Gordon et al.· Journal of composites for co...· 0 citations
A comparative framework evaluating nine regression algorithms using the UCI Concrete Compressive Strength dataset, jointly integrating correlation-corrected statistical validation, multi-model Bayesian optimization, and domain-informed feature engineering with SHAP interpretation, rarely combined in prior concrete-strength studies.
Musthafa 'Abduh Fakhruddin, Sri Winarno, Acun Kardianawati· IDEALIS : InDonEsiA journaL...· 0 citations
To overcome the dual bottlenecks of scarce actual strong earthquake records and high computational costs of nonlinear time history analysis, this study proposes a fast prediction method for structural nonlinear response that integrates systematic seismic sample expansion and machine learning technology by studying mature methods in the industry. A total of 500 ground motion records were created through the application of the amplitude scaling approach. Subsequently, the development of the steel frame structure was carried out through the application of the Abaqus software(Abaqus 2021 Edition) for the purpose of carrying out the nonlinear time history analysis to obtain the maximum interstory drift ratio (IDR) as the target response parameter. The XGBoost model was optimized to obtain improved results through the application of various evaluation criteria. Subsequently, the Shapley Additive exPlanations (SHAP) tool was applied to “open the black box” model to obtain the coupled effect of the various parameters, including displacement-related intensity measures such as RMSD and PGD on the maximum IDR during significant structure deformations. The method developed within this research has the potential to be a powerful tool for the prediction of the seismic responses. The method can be used in many areas, including probabilistic seismic demand, fragility assessment, and rapid evaluation of earthquake damage.
Han-Yu Feng, Aifu Sun, Hanwei Wang et al.· Buildings· 0 citations