Seismic Vulnerability Assessment of RC Buildings via Bayesian Updating: Evidence from the 2017 Puebla Earthquake

OverviewThis dataset contains the structural models, seismic records, and statistical analysis results supporting the research article: "Seismic Vulnerability Assessment of RC Buildings via Bayesian Updating: Evidence from the 2017 Puebla Earthquake". The data enables the full reproduction of a forensic engineering assessment of mid-rise RC buildings with unreinforced masonry (URM) infills, typical of Mexico City's transition zone.MethodologyThe research integrates two primary sources of information:Numerical Evidence: High-fidelity non-linear static (pushover) and reliability analyses performed in OpenSeesPy. A suite of 73 unscaled ground motion records from the September 19, 2017 earthquake (PGA max = 1.705 m/s²) was used to characterize seismic demand.Empirical Evidence: Field reconnaissance data from 62 buildings (19 observed collapses) following the earthquake.The synthesis is performed through a Beta-Binomial Bayesian Updating framework to refine the prior analytical collapse probability ($P_f$ = 46.9%) into a posterior estimate ($P_f$ = 32.9%).Dataset StructureThe repository is organized into three main components:/scripts: Contains Jupyter Notebooks (.ipynb) for the OpenSeesPy structural modeling, fragility curve generation, and Bayesian inference algorithms./data: Includes the compressed ground motion suite (.zip) consisting of 73 seismic records from the Mexico City transition zone./results: Raw data in CSV format including the Pushover capacity curves, fragility function coordinates, and probability density matrices (Prior vs. Posterior distributions). High-resolution versions of the manuscript figures are also provided.Usage NotesThe Python scripts require the openseespy, numpy, scipy, and matplotlib libraries. The CSV files can be imported into any statistical software for comparative vulnerability studies.