ABSTRACT
Design guidelines for fiber-reinforced polymers (FRP) that are used in the construction of civil applications must take into account the loss of stiffness and strength these materials undergo over time. American Associatio n of State Highway Transportation Officials (AASHTO) load resistance factor design (LRFD) uses statistical methods to develop a strength reduction factor for a particular combination of loads. LRFD uses statistical representations of mechanical properties at a specific time coupled with a statistical representation of the loading situation to determine the appropriate resistance factors. LRFD alone is an inadequate way to determine the future reliability of a composite structure due to the lack of knowledge about the future properties of the composite structure. A new method that combines LRFD and a material properties evolution scheme is proposed to predict the future reliability of a FRP structure. The Reifsnider residual strength model is used to track the remaining strength of the structure through time. The residual strength model is capable of taking into the account the variability of the loading conditions, mechanical and environmental. Inputs that are unique to this method are the statistical representations of a materials S-N curve and stiffness loss. The strength evolution model is conducted in a Monte Carlo style to determine a statistical representation of the remaining strength at some future time. Using this distribution of remaining strength as input, the LRFD methods can be used to determine the reliability and the resistance factors for the structure at a specific time in the future.
