Zhaoyan Li

Earthquake ground-motion prediction is critical for seismic design, risk assessment, and infrastructure resilience. Traditional empirical models often struggle to capture the complex, non-linear physical mechanisms of wave propagation and site amplification, leading to high aleatory variability and limited predictive accuracy. To address these limitations, this study proposes a ground-motion prediction model for the Italian region based on a Mixture-of-Experts (MoE) framework, utilizing XGBoost regressors as specialized experts. To ensure a robust assessment of predictive generalization, the model is developed and verified using a strict event-wise chronological splitting strategy on the ITACAext 2.0 dataset, which effectively eliminates event-level information leakage. Compared with empirical ground-motion prediction equations (GMPEs) and baseline machine-learning models, the MoE-XGB framework achieves superior predictive accuracy, significantly reducing both inter-event and intra-event variability across a wide spectral period range. Beyond statistical performance, we employ SHAP (SHapley Additive exPlanations) analysis to open the “black box” of the model, demonstrating that the MoE-XGB framework successfully learns and synthesizes physically sound seismological relationships, including magnitude scaling, geometric attenuation, and site amplification. As a high-fidelity foundational predictive engine, the MoE-XGB model provides robust, physically consistent, and fully decomposed statistical inputs that serve as a reliable basis for future regional seismic hazard assessments and performance-based engineering. The proposed framework exhibits strong transferability and scalability, offering a valuable reference for data-driven ground-motion prediction in other seismically active regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.

Earthquake ground-motion prediction is crucial for seismic design, seismic hazard assessment, and the resilience of urban infrastructure. Although extensive research has been conducted for Italy, existing models cover only a limited range of earthquake types, exhibit insufficient accuracy and uncertainty control under complex scenarios – thus lowering reliability – and provide a restricted set of ground-motion intensity measures (IMs) that cannot meet the multi-indicator needs of engineering practice and risk assessment. To address these issues, this study proposes a ground-motion prediction model for Italy based on a Mixture of experts (MOE) framework, in which XGBoost is employed as the expert submodels to enhance predictive accuracy and stability across diverse scenarios. We conduct a systematic comparison between the proposed MOE-XGB and a baseline Gaussian process regression model with an exponential kernel (GPR, exponential). The results show stable and balanced improvements across multiple IMs – such as peak ground acceleration (PGA), peak ground velocity (PGV), and spectral acceleration (SA) at different periods – demonstrating advantages in both accuracy and robustness. Furthermore, using the larger and more diverse ITACA (Italian Accelerometric Archive) dataset, we retrain and evaluate MOE-XGB. The model achieves higher accuracy on all considered metrics and maintains stable performance in generalization tests based on independent earthquake events, highlighting strong generalization capability and robustness. In summary, the proposed MOE-XGB provides a high-accuracy and broadly applicable solution for ground-motion prediction in Italy; meanwhile, the framework exhibits good transferability and scalability, offering a useful reference for fusion-model-driven ground-motion prediction in Europe and other regions.