This paper presents an explainable ensemble learning framework using Bootstrap Aggregating to predict structural damage in masonry buildings during seismic events. It estimates the peak ground acceleration (PGA) leading to the damage control limit state (significant damage) based on structural parameters. The model achieves high accuracy (R2=0.9536, MAE=0.0057) and interpretability through SHAP, aligning with structural engineering principles. Compared to finite element analyses, it offers faster computations (milliseconds) and scalability, enabling rapid intervention planning after earthquakes. Developed under the MEDEAproject(EUGrantn. 10101236), it supports disaster response and enhances seismic resilience.
Explainable Ensemble Learning for Structural Damage Prediction under Seismic Events
Baldassini, Michele;Foglia, Pierfrancesco;Lazzerini, Beatrice;Pistolesi, Francesco
;Prete, Cosimo Antonio
2025-01-01
Abstract
This paper presents an explainable ensemble learning framework using Bootstrap Aggregating to predict structural damage in masonry buildings during seismic events. It estimates the peak ground acceleration (PGA) leading to the damage control limit state (significant damage) based on structural parameters. The model achieves high accuracy (R2=0.9536, MAE=0.0057) and interpretability through SHAP, aligning with structural engineering principles. Compared to finite element analyses, it offers faster computations (milliseconds) and scalability, enabling rapid intervention planning after earthquakes. Developed under the MEDEAproject(EUGrantn. 10101236), it supports disaster response and enhances seismic resilience.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
