Seismic Fragility Assessment of Highway Bridges Using D-vine Copulas

Tong Zhou,Ai-Qun Li
DOI: https://doi.org/10.1007/s10518-018-0474-x
IF: 4.556
2018-01-01
Bulletin of Earthquake Engineering
Abstract:Seismic vulnerability of highway bridges is crucial to the seismic risk assessment of highway transportation networks. A new framework for developing the fragility function at the system level utilizing the D-vine copula theory is proposed. In this methodology, first, the joint probabilistic seismic demand model (JPSDM), which represents the conditional joint distribution of various engineering demand parameters (EDPs) under each intensity measure (IM) level, is constructed using the D-vine approach; then, sample realizations of both the established JPSDM and capacity models of these considered components are compared to derive the system-level fragility curves. This methodology is illustrated in a typical two-span RC continuous girder-box bridge case, where pier columns, elastomeric pad bearings, the sliding bearings, and the abutments in both passive and active directions are considered as the major components. The results demonstrate that the performance of the optimal D-vine for characterizing the dependence structure among various EDPs is relatively superior in comparison with commonly-utilized multivariate standard copulas. The combination of the quantitatively-identified optimal D-vine copula and their conditional lognormal marginal distributions is adequate to construct the JPSDM of these five EDPs. Furthermore, the effects of different copula selections on the overall system vulnerability are well captured, which highlight the necessity for quantitative identification of the optimal D-vine copula for modeling the correlations among various EDPs.
What problem does this paper attempt to address?