A Bayesian approach for estimating the post-earthquake recovery trajectories of electric power systems in Japan

Abstract

ABSTRACT Post-disaster recovery modelling of engineering systems has become an important facet of catastrophe risk modelling and management for natural hazards. The post-disaster recovery trajectory of a civil infrastructure system can be quantified using (a) the initial post-disaster functionality level, Q o ; (b) rapidity, h (i.e., the rate of functionality restoration); and (c) recovery time, R t . This study uses a Bayesian estimation approach to derive a set of probabilistic models to estimate Q o , R t , and h of electric power networks (EPNs) using post-earthquake recovery data from 16 large earthquakes in Japan between 2003 and 2022. The considered predictor (explanatory) variables include earthquake magnitude, year of occurrence, seismic intensity, and exposed population (PEX). Apart from being a simple and efficient stand-alone tool, the proposed data-driven models can be a useful benchmarking tool for simulation-based approaches for EPN recovery modelling.