Data analytic engineering and its application in earthquake engineering: An Overview

Loi, D.W., Raghunandan, M.E., Shanmugavel, M., and Swamy, V. (2014), Proceedings of the IEEE International Conference on Computational Intelligence and Computing Research, 1334-1337. ISBN: 978-1-4799-3972-5. (SCOPUS), 2014

This paper deliberates the challenges of using regression models for earthquake data analysis and compares them with the field measurements. Regression analyses to model the peak ground acceleration (PGA) data are discussed with magnitude and distance as variables. Suitability of the models are further compared with the ground motion (PGA) field records obtained from the seismic stations within the peninsular Malaysia. Far field (distance above 300km from the epicenter) and local earthquakes within 50-300km with a wide range of moment magnitude (1.0 - 9.1) are considered in this study. Result from the regression models showed significant error between the predicted and field data. Further discussion highlights that the ground motion prediction equation(GMPE) is a function of multiple variables developed from the specific site properties. The paper concludes with a note showing the significance of statistical input and analysis in the GMPE’s to achieve a more realistic earthquake data prediction model.

Click here for more information.