A quantitative assessment framework for the physical vulnerability of building to subaerial landslide-induced impulse wave

Evaluating the physical vulnerability of building is a crucial part of risk management for subaerial landslide-induced impulse wave (SLIIW). However, quantitative assessment is currently a challenge because uncertainties involved in every stage propagate sequentially, aggravating uncertainty effects and computation costs. Herein, we present a scenario-based probabilistic assessment framework for the SLIIW impacting building. Specifically, this framework evaluates the physical vulnerability of building based on two factors: hazard loading and building resistance, by systematically incorporating fluid dynamics simulation (capturing wave properties), formulation of wave impact intensities, pushover-based finite element analysis of structure (providing resistance capacities of building for each damage state), and efficient reliability computation. To this end, we propose a method, AK(TEAD)-DPIM which integrates the adaptive Kriging (AK) surrogate model with Taylor expansion-based adaptive design (TEAD) strategy and direct probability integration method (DPIM), to compute probabilities of SLIIW intensities exceeding resistance capacities. The building vulnerability is co-expressed via exceedance probabilities of damage states and the ratio of repair costs. In the complex fluid/structural dynamics context, the primary contribution of the method lies in integrating metamodel theory and DPIM, reducing computational costs triggered by probability-space partition strategies. Results of AK(TEAD)-DPIM with lower computation cost agree well with those of the traditional methods: probability density evolution method, and Monte Carlo simulation. This framework offers application prospects in the quantitative assessment of the cascading hazard and addresses the challenge of high data dependency inherent in data-driven approaches.