The presence of random fissures significantly alters the mechanical properties and failure mechanisms of rocks. To systematically investigate the impact of fissures on the failure behavior of sandstone, a multivariable random fissure numerical model was developed based on the Weibull distribution probability density function, in combination with a random fissure generation algorithm and cohesive element embedding method. This study primarily focuses on analyzing the influence of fissure ratio (R), fissure dip angle interval (A), fissure length interval (L), and fissure width interval (W) on the sandstone failure process. The results show that the failure modes change with variations in R, A, L, and W, specifically manifested as the formation of “X”-shaped, “Y”-shaped, or inverted “Y”-shaped primary cracks; the increase in fissure ratio significantly reduces both peak stress and total damage dissipated energy (ALLDMD), and promotes the propagation of tensile cracks; the increase in L leads to more complex failure patterns, but its effect on peak stress and peak strain fluctuates non-linearly, the ALLDMD remains insensitive to this change, while the number of tensile cracks decreases as L increases; conversely, an increase in W results in a failure mode characterized by a single crack path, the peak stress first increases and then decreases, and the ALLDMD exhibits an “N”-shaped fluctuation, though the overall variation is limited.

This study evaluates the seismic performance of the East Main Hall of Foguang Temple in Shanxi, focusing on the impact of wood property degradation on structural stability. A dynamic model of the hall is developed using the discrete element method (DEM) and Wallstat 5.1.3 software, simulating seismic responses under three conditions: intact wood properties, 0.85-fold reduction, and 0.75-fold reduction in wood properties. Peak ground acceleration (PGA) is used as the seismic intensity measure, and the maximum inter-story drift angle of the column frame is selected as the structural response parameter. Incremental dynamic analysis (IDA) is applied to generate seismic vulnerability curves to assess the influence of wood degradation on seismic performance. The results show that the DEM model’s natural frequency (2.40 Hz) is only 2.13% different from the code-estimated value (2.35 Hz), confirming the model’s reliability. As wood degradation increases, the maximum inter-story drift angle grows significantly, with the 0.75-fold reduction model exhibiting larger displacements than the intact and 0.85-fold reduction models. Seismic vulnerability curves indicate that wood degradation accelerates damage progression, with the 0.75-fold reduction model showing an 8.74% higher collapse probability under a PGA of 1 g.