Probabilistic resilience assessment of civil systems: Analysis and validity of the PEER framework

The PEER framework formula, promoted and used by researchers from the Pacific Earthquake Engineering Research (PEER) Center, has been proved to be a significant step in the realization of a performance-based earthquake engineering approach. The key objective of this short study isto investigate the use of the classical PEER framework in assessing the resilience of facilities or systems against natural hazards, with a particular focus on seismic loads. The resilience is a metric that accounts for the performance of a facility or a system after a shock and during its recovery phase. It is therefore of interest to have accurate, yet simple, tools to compute statics of this metric. In this study, the PEER framework formula is first revisited by imposing the resilience of a facility as a final decision variable, after which the limitations and the range of applicability are closely analyzed. The resilience assessment framework thusly developed sets the foundation for the implementation of resilient management strategies which can address optimal targets for a given hazard. A simple example with fully ergodic uncertainties approximation is introduced to practically show the computation of resilience statistics, such as the mean annual rate, the complementary cumulative distribution function, the mean annual cumulative value, and the approximate probability of extremes. Finally, we introduce the concept of a “resilience breaking point”, which represents scenarios that lead to the point where the system stops being resilient The resilience breaking points, as well as the potential events that could lead to such points, are of particular importance, as they define the critical scenarios that should be taken into meticulous account by policy makers. It is shown that the framework validity bounds can be used to define the range in which some of these critical points may occur.