What statistical method is commonly used to propagate uncertainties in seismic risk assessment?

Monte Carlo simulation is a statistical method commonly used to propagate uncertainties in seismic risk assessment. Seismic risk assessment involves many uncertain parameters, including ground motion intensity, soil properties, structural capacity, and damage-to-loss relationships. Monte Carlo simulation is a technique that uses random sampling to generate a large number of possible scenarios, each representing a different combination of these uncertain parameters. For each scenario, the seismic performance of the structure is evaluated, and the resulting damage and losses are estimated. By running a large number of simulations (e.g., thousands or millions), the method generates a distribution of possible outcomes, which can then be used to estimate the mean, variance, and other statistical measures of the seismic risk. The key advantage of Monte Carlo simulation is its ability to handle complex models and a wide range of uncertainties without requiring simplifying assumptions. It provides a comprehensive picture of the potential range of outcomes and their probabilities, enabling more informed decision-making regarding seismic risk mitigation. Other methods exist, but Monte Carlo is frequently employed due to its versatility and ability to accommodate complex dependencies between variables.