A semi-probabilistic procedure for developing societal risk function

Seismic risk is typically quantified probabilistically for a single asset or evaluated through regional loss assessment for selected earthquake events. Ideally, a recurrence relationship for a loss quantity, economic loss or casualty, can be obtained for risk-informed decision-making. This can be achieved by a fully stochastic approach, for which a large amount of input information is required, whilst there is usually a lack of transparency that might hinder repeatability of the outputs. Hence, the objective of this paper is to introduce a simple and unambiguous procedure for developing parametric societal risk function based on rigorous loss modelling of response-specific probabilistic scenarios. This is then illustrated for the Greater Melbourne Region with fatality as the loss quantity. The proposed semi-probabilistic procedure can be extended to other loss quantities, as well as evaluating societal risk of other natural hazards or multiple hazards.