System-level evidence and fragility-based quantification of connecting columns as alternative load paths in prefabricated reinforced masonry shear wall structures

General precast shear wall systems rely on wall-panel connections to maintain force-transfer continuity and structural integrity between prefabricated units. In prefabricated reinforced masonry shear wall (PCMSW) systems, connecting columns primarily function as splice connectors between adjacent precast wall panels. Owing to their higher compressive strength, greater reinforcement ratio, and closely spaced transverse reinforcement, connecting columns possess the mechanical potential to serve as alternative load paths during severe seismic events; however, systematic quantification of this mechanism at the system level remains limited. In this study, a three-dimensional explicit finite element model developed in LS-DYNA and benchmarked at two complementary levels—a system-level dynamic shaking-table benchmark and a component-level cyclic benchmark for the connecting-column / vertical-joint region—is employed to investigate the collapse behavior of a 10-story PCMSW structure. Incremental dynamic analyses are conducted using 22 pairs of ground-motion records for six connecting-column layout schemes, supplemented by a representative near-median ten-record parametric sensitivity study on key physical variables. The results indicate that once the masonry walls lose effective load-carrying capacity, the median vertical load-sharing ratio of connecting columns increases from 0.06 before the system-level onset of sustained wall degradation to 0.42, while the median lateral shear participation ratio rises from 0.10 to 0.35. Within the adopted continuum-modeling framework, the fitted, model-conditioned fragility results show that the median collapse intensity increases from 0.42 g to 0.68 g, corresponding to a median collapse-intensity ratio ηs of 1.62, with the fitted point estimate of the logarithmic standard deviation decreasing from 0.38 to 0.30 and the conditional collapse probability at 0.40 g decreasing from 44.9% to 3.8%; these absolute fragility quantities are interpreted as comparison-internal measures rather than direct empirical predictions of physical collapse capacity. Layout analysis indicates that positioning connecting columns near the exterior longitudinal walls yields the highest gain (ηs=1.67[jls-end-space/]). Among the physical parameters examined, the axial load ratio is the primary adverse factor, whereas the volumetric transverse reinforcement ratio is the most effective positive measure.