Increasing the Robustness of Bayesian Analysis for the Reliability Assessment of Existing Structures

The reliability assessment of existing structures is usually carried out applying partial factor method. As in most cases the complexity of the problem is beyond traditional assessment procedures, probabilistic methods are often invoked in order to estimate the actual probability of failure. Nonetheless a factor that should be considered is the potential bias arising with the refinement of probabilistic models: in effect, the outcome of reliability assessment will be dependable only if the potential bias is negligible. In this paper a procedure that increases the robustness of probabilistic methods and Bayesian updating techniques for the reliability assessment of existing structures is presented. The procedure will be especially devoted to historical buildings.