The 2017 Puebla, Mexico, earthquake event led to significant damage in many buildings in Mexico City. In the months following the earthquake, civil engineering students conducted detailed building assessments throughout the city. They collected building damage information and structural characteristics for 340 buildings in the Mexico City urban area, with an emphasis on the Roma and Condesa neighborhoods where they assessed 237 buildings. These neighborhoods are of particular interest due to the availability of seismic records captured by nearby recording stations, and preexisting information from when the neighborhoods were affected by the 1985 Michoacán earthquake. This article presents a case study on developing a damage prediction model using machine learning. It details a framework suitable for working with future post-earthquake observation data. Four algorithms able to perform classification tasks were trialed. Random forest, the best performing algorithm, achieves more than 65% prediction accuracy. The study of the feature importance for the random forest shows that the building location, seismic demand, and building height are the parameters that influence the model output the most.
Understanding seismic risk enables efficient resource allocation in the effort to increase the resilience of our cities and communities. Field reconnaissance and data collection following disasters document the damaging effects of earthquakes to enable lessons and wisdom to accumulate from past events. An important aim of such field data analysis is establishing a better understanding of building performance such as causes of building failures. These lessons provide essential basis to improve our design codes, develop regulations and policies, to increase building resilience by addressing the infrastructure vulnerability. Currently, to make use of the datasets from around the world, significant effort is required to decode the data which often have unique local and regional context and bias. The struggle begins at data collection where there is a lack of consistent methodology and definitions that can adequately cover the regional nuance. This manuscript proposes a new paper-based tool which addresses the need for a global yet detailed universal methodology for building damage assessment post-earthquakes. The new form is based on the GEM taxonomy v2.0 and the European Macroseismic Scale EMS-98. The recent Mexican earthquake from the 19 September 2017 led to significant building damage in the capital Mexico City and the state of Morelos. A team from New Zealand assessed damage throughout the capital and tested the new paper form in Calle La Morena. The street case study presents a novel visualization of the damage data and buildings characteristics which highlights the correlation between the damage and the building features. It is hoped that this kind of illustration will lead to better comprehension of the damage drivers.
This report presents the observations and findings following the 2017 Puebla earthquake that occurred inMexico on September 19th, 2017. The reconnaissance mission was a collaboration between the New ZealandSociety of Earthquake Engineering (NZSEE), the Universidad Autónoma Metropolitana (UAM) Azcapotzalco,the American Concrete Institute (ACI) Disaster Reconnaissance team, and the Colegio de Ingenieros Civilesde Mexico (CICM). During the earthquake, 77 buildings suffered partial or total collapse and more than8,000 buildings experienced damage ranging from slight damage to significant structural damage necessitatingdemolition. As observed in previous earthquakes, the unique soil conditions of Mexico City resulted inextensive damage to the city’s infrastructure, primarily due to local site effects. The earthquake causedrelatively more damage to buildings built on transition and soft soil zones (i.e. between hard and deep softsoils) than those on hard soils.The NZSEE and UAM team focussed on areas with widespread and extensive damage. They also assessedthe performance of repaired and retrofitted buildings after the 1985 Michoacán earthquake. It was found thatthe lessons learnt from the 1985 Michoacán earthquake led to some risk mitigation measures which benefitedseveral buildings in the 2017 earthquake. Retrofitted buildings were found to have performed very well withlittle or no damage when compared to other buildings.
Abstract. This paper presents a new framework for the seismic loss prediction of residential buildings in Ōtautahi / Christchurch, New Zealand. It employs data science techniques, geospatial tools, and machine learning (ML) trained on insurance claims data from the Earthquake Commission (EQC) collected following the 2010–2011 Canterbury earthquake sequence (CES). The seismic loss prediction obtained from the ML model is shown to outperform the output from existing risk analysis tools for New Zealand for each of the main earthquakes of the CES. In addition to the prediction capabilities, the ML model delivered useful insights into the most important features contributing to losses during the CES. ML correctly highlighted that liquefaction significantly influenced building losses for the 22 February 2011 earthquake. The results are consistent with observations, engineering knowledge, and previous studies, confirming the potential of data science and ML in the analysis of insurance claims data and the development of seismic loss prediction models using empirical loss data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.