Probabilistic seismic assessment of a high-rise URM building

According to the Sendai Framework for Disaster Risk Reduction (2015–2030), disasters have demonstrated that the recovery, rehabilitation and reconstruction phase, which needs to be prepared ahead of a disaster, is a critical opportunity to “Build Back Better”, integrating disaster risk reduction into development measures. In this respect, a significant number of structures, that constitute several European urban nuclei, belong to old constructive typologies, which were designed and built without any consideration for the seismic hazard. One of the most used typologies exhibiting this shortcoming is unreinforced masonry (URM). Therefore, an important step towards increasing resilience of European cities is to deeply understand the seismic behavior of this frequent typology. In order to do so properly, detailed probabilistic nonlinear building models should be developed. However, including the uncertainties associated with this typology is challenging due to the heterogeneity of the different manufacturing techniques, executed under primitive industrial standards, and to the construction techniques, which are dependent on regional uses and criteria in a pre-code scenario. The object of this research is twofold. First, a detailed quantification of the uncertainties related to the mechanical properties of this construction material is conducted. Then, the influence of this variability on the seismic performance of a representative building model of the Eixample district in Barcelona, Spain, is analysed. This building typology represents 72% of the building stock in this district with an average age of 90 years, which means that the construction practice, at that time, was only regulated by early council guidelines that are considered pre-code rules. Specifically, the probabilistic approach is illustrated with a case study performed on an existing seven-story (high-rise) URM. A detailed numerical model of this structure has been developed and randomized taking into account the variability of the material properties. Accordingly, 1000 models were generated and analysed by considering as input different sets of material random variables. The compressive strength, Young modulus, shear modulus and shear strength are chosen and modelled to encompass the material uncertainties. The seismic response of each variant (i.e. selected set of mechanical properties) is obtained through a simplified non-linear static procedure aiming to compare and categorize the influence of