Time-Variant Reliability Analysis and Flexural Design of GFRP-Reinforced Bridge Decks

Abstract

Glass-fiber reinforced polymer (GFRP) reinforcement is being used in bridge decks as a replacement for steel reinforcement. It is thought that since the GFRP reinforcement does not corrode, it could be a more sustainable material for reinforced concrete structures. However, it is widely reported that GFRP bars do deteriorate when embedded in concrete or when immersed in concrete pore solutions. The American Concrete Institute (ACI) and AASHTO use environmental exposure factors to reduce the design strength due to this loss of capacity. However, these exposure factors have not been calibrated. In addition, limited research has been performed to quantify the time-variant flexural moment capacity of GFRP-reinforced flexural members. Recently, a Bayesian approach was used to develop a time-variant probabilistic capacity model based on capacity data of GFRP reinforcement embedded in concrete for a period of up to seven years. This model is used herein to assess the time-variant flexural moment capacity and the time-variant structural reliability of a bridge deck considering different types and sizes of GFRP bars. Even though GFRP-reinforced bridge decks are designed to be overreinforced and the designs are governed by the serviceability limit, the analysis results indicate that the probability of failure of the decks containing both 13M (#4) and 19M (#6) GFRP bars at a reference temperature T23 of 23°C (73°F) is higher than the failure probability generally accepted in the AASHTO load and resistance factor rating (LRFR) specifications. A lower exposure temperature reduces the probability of failure over the 75 year period. The probability of failure is less than values accepted by AASHTO.