Development of Fragility Curves of Concrete Hollow Block (CHB) Masonry Walls in the Philippines Using the Rigid Block Analysis

This study focuses on the seismic vulnerability of concrete hollow block (CHB) walls, a common construction material in the Philippines, by developing fragility curves based on the rigid block analysis. The rigid block analysis assumes that the wall behaves as a rigid, non-deformable body, which simplifies the modeling of overturning mechanisms in masonry walls. Nonlinear dynamic analysis was performed using a MATLAB framework, with 100 accelerograms applied in both directions. Fragility curves quantify the probability of a structure reaching or exceeding specific damage states (DS) under varying intensity measures. The probabilistic contributions include the probability of rocking activation, the probability of DS exceedance (DS1 to DS5), and the probability of collapse, evaluated using Peak Ground Acceleration (PGA) and Peak Ground Velocity (PGV). Fragility curves were developed for 16 wall configurations, categorized based on load-bearing function, block thickness, presence of gables, and transverse wall connection levels. These curves were then combined to reveal overall performance trends. The results indicate that non-load-bearing walls and walls without gables exhibit greater stability while increasing block thickness marginally improves robustness. Medium levels of transverse wall connections provide a significant stabilizing effect. The parametric study revealed that wall height presents complex behavior, with increased height amplifying vulnerability in poorly connected configurations, but contributing to stability in well-connected walls. The wedge angle emerges as the most significant parameter for stability. The effect of reinforcement was also studied, showing that steel reinforcement delays the initiation of rocking. Validation of the model using damage data from the 2013 Bohol earthquake reveals that the fragility estimates are more conservative than observed damage, indicating that the model provides higher probability estimates for failure.