The losses incurred by industrial facilities following catastrophic events can be broadly broken down into property damage and business interruption due to the ensuing downtime. This article describes a generalized probabilistic methodology for estimating facility downtime under multi-hazard scenarios. Since the vulnerability of each components of an industrial facility varies with the types of hazard, it is beneficial to adopt a system-of-systems approach for analyzing such complex facilities under multiple interdependent hazards. In this approach, the complex layout of the facility is first broken down into its constituent components. The component vulnerabilities to different hazards are combined using Boolean logic, assuming their repair time as a common basis for defining damage states of the component. This combination results in multi-hazard fragility functions for each component of the system, which give the probability of damage under combined occurrence of multiple perils. The time to repair a component is expressed probabilistically using restoration functions. Using fault tree analysis, the components’ fragility functions and restoration functions are propagated to calculate system-level downtime. We demonstrate the methodology on a case-study power plant to estimate downtime risk under combined earthquake and tsunami hazard.
All Science Journal Classification (ASJC) codes
- Water Science and Technology
- Atmospheric Science
- Earth and Planetary Sciences (miscellaneous)