A move toward improved portfolio seismic risk assessment methods for the practicing engineer.

David J. DeBock, & Abbie B. Liel

Published 2015, SCEC Contribution #7141

The last decade has seen growing interest in probabilistic portfolio-level (regional) seismic risk assessment. State-of-the-art methods build on established performance-based earthquake engineering methods for individual buildings, in combination with robust treatment of spatial correlations of ground motion intensity, building response, and collapse, which complicate the analysis significantly. Consequently, these methods are practically out of reach for many potential users. Relationships between advanced portfolio risk assessments and a second set of risk assessments that are based on the summation of risks for individual buildings are examined. Results for a sample region show that a simple sum of the risks for each individual building within a portfolio under-estimates the values of small frequent losses and over-estimates the values of large rare losses. Since their differences are systematic, a simplified procedure to estimate portfolio-level seismic risks based on assessments of individual buildings (or building types) within the portfolio is feasible in the future.

Citation
DeBock, D. J., & Liel, A. B. (2015). A move toward improved portfolio seismic risk assessment methods for the practicing engineer.. Oral Presentation at 2nd ATC-SEI Conference on Improving Seismic Performance of Existing Buildings and Other Structures.