Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Seismic risk analysis, at large scale and local scale, is fundamental support for mitigating the investment costs and recovery works, impacting society's economy. The presented article's main purpose is to present a mechanicalbased model to evaluate the fragility curves through nonlinear static analysis. The results achieved from statistical analysis allows defining a representative building stock for two height classes. Material properties for typical masonry typologies that characterise the unreinforced masonry building heritage for the specific area of investigation are assigned to the prototype buildings. The paper proposes a procedure for fragility function evaluation to reduce computation time and limit the results' loss of reliability. A technique based on macro-element formulation and threedimensional equivalent frame model is proposed to evaluate the assessed buildings' seismic response. The proposed model includes implementing the macro-element formulation in open-source software, Opensees, that allows higher versatility and the possibility to include in the analysis sources for better reliability and representation regarding the real building. Equivalent frame modelling allows an efficient way to represent the seismic response and a compromise between computation costs and high accuracy. Modelling technique validation is proposed by comparing results derived on specified studies concerning, ambient vibration and seismic response. A limited set of capacity curves is generated from the aleatory variation on the material properties. Then, the number of capacity curves is expanded by a correlation between the seismic response parameters. The increase in capacity curves aims to cover the variability on the structural capacity. Fragility curves are then derived from a stochastic selection of response spectra, by using GMPE allowing to estimate the probability distribution of each damage state, as a function of the reached displacement. Reliability of the mechanical fragility curves is proved by comparing recent studies, involving both empirical and mechanical based fragility functions. The assessment's feasibility is demonstrated by applying the procedure over most representative prototype building configurations for the Switzerland city of Basel.