Reliability Assessment of FRP-Strengthened Concrete Bridge Girders in Shear

Abstract

This paper presents the results from an investigation into the reliability of reinforced concrete bridge girders strengthened in shear using fiber-reinforced polymers (FRP). Two expressions for the shear design of FRP-strengthened girders were developed as part of a National Cooperative Highway Research Program project. The expressions were developed for two distinct cases, namely, bonded and anchored FRP reinforcement. Uncertainties inherent in the new design models were first assessed using an extensive database of hundreds of experimentally tested specimens. Variabilities in material, fabrication tolerances, dead and live loads, and distribution factors obtained from the literature were then included in a limit state function for the shear strength mode of failure. The reliability of a representative design space comprising 18 bridge girders that cover different span lengths, shear deficiency levels, and girder location (exterior versus interior) was calculated for both bonded and anchored strengthening techniques. The use of approximate expressions for reliability index calculations was deemed unacceptable because of the high scatter in the results. This paper discusses the difference between two reliability methods [Monte Carlo simulations and first-order reliability method (FORM)] that were first used to ensure the validity of FORM in analyzing the developed limit state function. The results showed that the new design expressions result in reliability index values in the range targeted by most design codes (3.00–3.50), albeit lower than the target value used in the calibration of AASHTO load and resistance factor design (LRFD) (