COMPDYN 2015 5 th ECCOMAS Thematic Conference on Computational Methods in Structural Dynamics and Earthquake Engineering M. Papadrakakis, V. Papadopoulos, V. Plevris (eds.) Crete Island, Greece, 25–27 May 2015 USING DIFFERENT UNCERTAINTY MODELS FOR SEISMIC ASSESSMENT OF RC BRIDGES R. Monteiro 1,2 , R. Delgado 2 , and R. Pinho 3 1 Institute for Advanced Study of Pavia Palazzo del Broletto, Piazza della Vittoria n. 15, 27100 Pavia (Italy) e-mail: ricardo.monteiro@iusspavia.it 2 Faculty of Engineering, University of Porto Rua Dr. Roberto Frias, s/n Porto (Portugal) e-mail: rdelgado@fe.up.pt mailto:rui.pinho@unipv.it 3 University of Pavia, Department of Civil Engineering and Architecture Via Ferrata n. 1, 27100 Pavia (Italy) e-mail: rui.pinho@unipv.it Keywords: Seismic Assessment, RC Bridges, Failure Probability, Latin Hypercube Sampling, Uncertainty Models, Nonlinear Demand Prediction Abstract. The seismic assessment of structures depends on a large number of aleatory and epis- temic uncertainties, which are majorly associated to the estimation of the structural demand and capacity, both usually featuring considerable dispersion levels, particularly when reinforced con- crete structures are being assessed. When focusing on bridges, additional complexity may be in- troduced by the irregular behaviour in the transverse direction. Several procedures may be used for the assessment of the seismic safety of bridges, deterministic or probabilistic, and all rely on an accurate prediction of the demand, obtained via linear or nonlinear static or dynamic analysis. This work employs both static and dynamic analysis methods for demand estimation within a rela- tively straightforward framework to compute the failure probability of existing bridges. Different variables typically considered in a seismic assessment procedure (geometry, material properties, earthquake records, intensity level) are statistically characterized, catering for a global simula- tion process, where each iteration step is associated to an independent structural nonlinear static or dynamic analysis. Failure probability is then obtained through different uncertainty models, corresponding to the convolution of capacity and demand distributions or the probabilistic analy- sis of a safety indicator, defined as the difference between capacity and demand at each random simulation realisation. A case study of seven bridge configurations, with different (ir)regularity levels, is considered together with a relatively large set of real earthquake records. The simula- tion process is carried out using Latin Hypercube sampling, expected to considerably reduce the number of realizations with no reliability loss. Conclusions have allowed the identification of vul- nerable configurations and shown the differences in considering different uncertainty models. 887