Seismic fragility curve update based on empirical fragility estimates employing automated ERGO platform

Abstract

Earthquakes often affect a large area and cause enormous physical and socioeconomic losses. Thus, seismic risk assessment is essential to reduce such losses caused by earthquakes, and accurate seismic fragility curves are known to be the key to reliable seismic risk assessment. A number of studies have developed seismic fragility curves for various structures, and some researchers have attempted to improve seismic fragility curves based on empirical fragility estimates. However, there are still research challenges with the brevity and efficiency of the analysis. This study proposes a new method for efficiently updating the seismic fragility curves of structures based on actual fragility data. A seismic fragility curve is generally modeled as a lognormal distribution, which requires the determination of two probabilistic distribution parameters: logarithmic mean and logarithmic standard deviation. To simulate seismic losses with various distribution parameters and find the optimal values, this study employs ERGO, which is a GIS (Geographic Information System) based seismic risk assessment platform jointly developed by the Mid America Earthquake (MAE) Center and the National Center for Supercomputing Applications (NCSA) of the University of Illinois, Urbana-Champaign. Although ERGO is a widely-used platform to predict seismic losses given earthquake scenarios, its graphic user interface (GUI) makes it difficult to find the optimal distribution parameters through iterative analysis. To resolve this issue, the proposed method develops an automated ERGO platform which enables GUI-based ERGO analysis for various distribution parameters through macro controlled in MATLAB. Using the automated platform, optimization is performed to find the optimal values of the distribution parameters of a seismic fragility curve based on actual fragility estimates, and ERGO, which can deal with multiple structure types and damage levels at the same time, enables simultaneous updates of multiple fragility curves. To test the proposed method, it is applied to update the seismic fragility curves of building structures in the Republic of Korea. Based on actual 2km by 2km building-damage data caused by 2017 Pohang earthquake, seismic fragility curves for multiple building types and damage levels are successfully updated, and the updated fragility curves show a better agreement with the actual damage probabilities than the original fragility curves.