Comparative analysis of seismic fragility estimation methods for a buried pipeline

Abstract

Structural damage caused by earthquakes has various uncertainties, and seismic fragility estimation is considered an essential element in assessing seismic risk by considering these uncertainties. The types of seismic fragility curves, which are defined to be the probability that a structure will have beyond a certain level of damage due to earthquake, are classified as empirical, judgmental, analytical, and hybrid. Among these, analytical curves have gained much attention due to their convenience and wide applicability. Since their first development, analytical methods of seismic curve derivation have evolved through various techniques, including the capacity spectrum method (CSM), incremental dynamic analysis (IDA), probabilistic seismic demand model (PSDM), finite element reliability analysis (FERA), and Monte Carlo simulation (MCS). Each method is based on different assumptions and mathematical techniques, leading to various shapes of fragility curves depending on the target structure and analysis conditions. This study aims to apply several analytical methods to a structure and compare the results. For this comparative purpose, seismic fragility curves were derived for a buried pipeline with a diameter of 0.762 m and a length of 1,200 m using CSM, PSDM, FERA, and MCS. Furthermore, the probability of failure for three damage states—Minor, Moderate, and Major—was expressed as a log-linear relationship with peak ground acceleration (PGA) and compared across the different methods. As a result, it was found that the CSM produced the most conservative results, while FERA exhibited trends most similar to those obtained through MCS. On the other hand, CSM showed its practical applicability based on the computational efficiency and simplicity of the analysis.