Human-induced vibration serviceability analysis of high-frequency floors under different layouts of obstacles based on random loading and random structural properties

s: Traditional analyses of human-induced floor vibrations often focus solely on random loading while relying on deterministic structural models and thereby overlooking the combined effects of random loading and structural variability on serviceability. This study investigates the influence of both random loading and structural randomness on the serviceability of high-frequency floors configured with different obstacle layouts. First, a random crowd loading model is established by combining the social force model and time-domain representation of human-induced loads, capturing the stochastic nature of pedestrian walking patterns. Next, a finite element model of the high-frequency floor is developed with semi-rigid boundary conditions, and its parameters are optimized to accurately reflect real vibration characteristics, providing a basis for introducing structural randomness. An improved biomechanical model is subsequently employed to construct a coupled crowd–floor interaction system. The probability density evolution and degree of reliability of the dynamic response are evaluated under different obstacle layouts, considering both random loading and structural uncertainty. The results indicate that accounting for both sources of randomness broadens the probability density distribution and significantly increases peak acceleration. Obstacle configurations influence pedestrian movement variability, thereby altering dynamic response and degree of reliability. These findings indicate that high-frequency floors may still exhibit vibration-induced serviceability issues, even when designed according to conventional frequency-based criteria.