Assessing the safety of existing structures: Reliability based assessment framework, examples and application

Abstract

Safety, reliability and risk are key issues in the preservation of our built, cultural heritage. Several structural collapses make us aware of the vulnerability of our technical and natural environment and demand an adequate engineering response. In the analysis phase, an objective way to assess the safety of the structure is essential. The present raises the need for a reliability based assessment framework for existing masonry structures. Although this field of research is relatively young, different techniques have been proposed and optimised. These permit to calculate the global probability of failure of complex structures, relying on deterministic techniques able to calculate the stability state for a prescribed set of parameters. This paper illustrates how these techniques can be a valid tool to evaluate the bearing capacity of existing structures. Focus is on reliability methods based on simulation procedures (Monte Carlo, Directional Sampling), combined with an adaptive meta-model (Response Surface, Splines, Neural Networks). Several benchmark examples demonstrate the applicability of the methodology. The mutual efficiency of the different reliability algorithms is discussed. The application focuses on the assessment of an existing masonry structure. The overall stability of a Romanesque city wall of Leuven (B) is studied in detail. The analysis treats the present safety of the city wall, regarding the uncertainties in load, geometry and resistance. Because of the low degree of safety of several parts of the wall, consolidation measures and strengthening techniques are proposed.