Optimising portfolio loss reduction using a first-order reliability method sensitivity analysis

Advancements in loss estimation methodologies for building portfolios provide an opportunity for pre-disaster risk mitigation, particularly through structural retrofit. Budget constraints, however, limit which vulnerable buildings within a region are to be retrofitted to desirable levels of seismic resistance. This paper presents a new reliability-based approach to prioritise retrofit strategies to support regional earthquake risk mitigation efforts. The proposed framework uses the first-order reliability method (FORM) to evaluate a probability distribution of repair costs for a suite of spatially distributed buildings. Sensitivity measures computed within the FORM analysis are paired with retrofit costs. The proposed method is applied to a San Francisco neighbourhood building inventory, and results suggest which building categories should be upgraded to minimise estimated post-earthquake repair costs. The importance of spatial correlation as well as the level of loss to be minimised is investigated relative to cost-effective retrofit spending.