Updating of service-life prediction of reinforced concrete structures with potential mapping

Fully probabilistic models are available for predicting the service life of new reinforced concrete structures and for condition assessment of existing structures. Frequently, the decisive mechanism limiting the service life of reinforced concrete structures is chloride-induced corrosion, for which these models predict probabilistically the time to corrosion initiation. Once the corrosion process is initiated, corroding areas can be detected nondestructively through potential mapping. The spatial information gained from potential mapping can then be used for updating the service-life prediction, taking into consideration the spatial variability of the corrosion process. This paper introduces the spatial updating of the probabilistic model with potential mapping and concrete cover measurements by means of Bayesian analysis. A case study is presented, where potential mapping is applied prior to a destructive assessment, which serves to verify the approach. It is found that the potential mapping can provide significant information on the condition state. With the presented methods, this information can be consistently included in the probabilistic service-life prediction.