Estimating the critical chloride threshold of reinforcing steel in concrete using a hierarchical Bayesian model

Significant research has been conducted to assess the amount of chlorides that results in corrosion initiation, C _T, and propagation of the embedded reinforcing steel in concrete. However, the lack of a standard definition for corrosion initiation and the lack of a standard test and measurement criterion for determining corrosion initiation, all contribute to wide scatter in reported C _T values in the literature. In this study, a hierarchical Bayesian approach is used to develop a predictive model for C _T based on the reported C _T values in the literature. The developed model accounts for the influence of cement type, supplementary cementitious materials, water-to-binder ratio, and the surface condition of the reinforcing steel on C _T. The developed model can be used to streamline and standardize the process of selecting C _T for service life modeling but to reduce the uncertainty in service-life predictions, it is essential to develop a standard test for determining C _T.