Assessing post-seismic damage on an urban/regional scale remains relatively difficult owing to the significant amount of time and resources required acquire information conduct a building-by-building seismic assessment. However, application new methods based artificial intelligence, combined with increasingly systematic availability field surveys damage, has provided perspectives for This study analyzes effectiveness relevance number machine learning techniques analyzing spatially...

In this work, we explored the feasibility of predicting structural drift from first seconds P-wave signals for On-site Earthquake Early Warning (EEW) applications. To purpose, investigated performance both linear least square regression (LSR) and four non-linear machine learning (ML) models: Random Forest, Gradient Boosting, Support Vector Machines K-Nearest Neighbors. Furthermore, also explore applicability models calibrated a region to another one. The LSR ML are validated using dataset...

Building-specific loss assessment methodologies use component fragility curves to compute expected losses in the aftermath of earthquakes. Such are not available for steel columns assuming they remain elastic because capacity design considerations. Nonetheless, first-story moment-resisting frames (MRFs) experience damage through flexural yielding and formation geometric instabilities. This paper uses an experimental database that was recently assembled develop two sets univariate drift-based...

In this paper, a number of spectral and ordinary ground motion intensity measures (IMs) are tested for use in structural performance assessment. Real strong values recorded at the top bottom US, Japanese Romanian buildings analyzed order to identify source uncertainties prediction engineering demand parameters (i.e. drift) given IMs σEDP|IM). The efficiency sufficiency each IM from large set building earthquake data different criteria characterizing seismic (magnitude source-to-site...

Abstract Assessing or predicting seismic damage in buildings is an essential and challenging component of risk studies. Machine learning methods offer new perspectives for characterization, taking advantage available data on the characteristics built environments. In this study, we aim (1) to characterize using a classification model trained tested survey from earthquakes Nepal, Haiti, Serbia Italy (2) test how well given region (host) can predict another (target). The strategy adopted...

Abstract. Assessing or forecasting seismic damage to buildings is an essential issue for earthquake disaster management. In this study, we explore the efficacy of several machine learning models characterization, trained and tested on database observed after Italian earthquakes (DaDO). Six regression- classification-based were considered: random forest, gradient boosting extreme boosting. The structural features considered divided into two groups: all provided by DaDO only those be most...

Assessing or forecasting seismic damage to buildings is crucial for earthquake disaster management. Several classical assessment methods are available by combining hazard, exposure, and vulnerability. However, during emergencies, collecting all the necessary data may not be feasible due time resource constraints, as this information readily available. In context, machine learning can offer a paradigm shift reasonably assessing relying on cost-effectively. study, we aim study prediction...

Abstract In this study, accelerometric data from seven Japanese buildings under long-term monitoring were analysed to explore the variability of buildings’ co-seismic response over time and its within- between-building components, using capacity curves developed in acceleration-displacement-response-spectrum format. The include 2011 Tohoku Mw9.1 earthquake, which caused building damage different levels severity, time-varying actual considering earthquakes before after 2011. Result showed...

Abstract In this study, accelerometric data from seven Japanese buildings under long-term monitoring were analysed to explore the variability of buildings’ co-seismic response over time and its within- between-building components, using capacity curves developed in acceleration-displacement-response-spectrum format. The include 2011 Tohoku Mw9.1 earthquake, which caused building damage different levels severity, time-varying actual considering earthquakes before after 2011. We observed that...

Abstract. Assessing or forecasting seismic damage to buildings is an essential issue for earthquake disaster management. In this study, we explore the efficacy of several machine learning models characterization, trained and tested on database observed after Italian earthquakes (the Database Observed Damage – DaDO). Six were considered: regression- classification-based models, each using random forest, gradient boosting, extreme boosting. The structural features considered divided into two...

<p>Over the last two decades, seismic ground motion prediction has been significantly improved thanks to development of shared, open, worldwide databases (waveform and parametric values). Unlike motion, earthquake data recorded in buildings are rarely shared. However, their contribution could be essential for evaluating performance structures. Increasing interest deploying instrumentation gives hope new observations, leading better understanding behavior. This manuscript...