Seismic assessment of non-engineered incrementally constructed unreinforced masonry infilled RC frames considering material variability

In many low: and middle-income countries, over half of the buildings are non-engineered structures, often including unreinforced masonry (URM) infilled with reinforced concrete (RC) frames, which often have poor seismic detailing. These buildings are typically built incrementally, leading to variations in structural systems and material quality, which result in poor seismic behavior. This study investigates the seismic performance of these buildings through a case study of a building incrementally constructed in two stages in Colombia's Sabana Centro region. The structure's system and building materials were identified through field observations. Based on these observations, representative models were developed for each construction stage. Material variability was accounted for through Latin Hypercube Sampling, using local material data to develop 150 models for each construction stage. Nonlinear dynamic analysis was performed using OpenSeesPy with FEMA P695 seismic records. Fragility curves detailing four damage states (slight, moderate, extensive, and collapse) were analyzed for damage probabilities and expected losses. Fragility curves, including uncertainties in material properties, were also evaluated, and the impact on seismic fragility using mortar and chicken wire mesh as a retrofitting technique was analyzed. Results showed that the collapse probability at the Maximum Considered Earthquake (MCE) is 42 % for the first stage and increases to 56 % for the second. Material properties strongly influence these probabilities, ranging from 32 % to 50 % for the first stage and 55 %–68 % for the second. Expected losses exceeded 50 % for both stages. Retrofitting significantly reduced the collapse probability, from 47 % to 5 % at the MCE level.