Seismic correlation coefficient evaluation for the application of separation of variables approach in fragility assessment

Seismic correlation can affect seismic probabilistic risk assessment results when components within the system have similar dynamic characteristics. While the fragility assessment methodology is well established, researches to evaluate and apply seismic correlation in fragility are still in progress. In this study, the seismic correlation coefficient (SCC) was quantitatively analyzed based on fragility assessment methods. In order to evaluate the SCC of several random variables associated with structure response in fragility assessment, a numerical analysis model of the structure was constructed and statistical analysis of the numerical results was performed. The SCC of the floor responses of the structural model considering random variables was evaluated from two perspectives. The first perspective involved applying variables individually or simultaneously when performing probabilistic structural analysis. The second perspective aimed to assess whether simplification of the structural analysis model affects the SCC. For this purpose, SCCs were calculated for each method and applied to a simplified failure sequence model to compare the combined fragility curves. The results of this study demonstrate the applicability of separation of variables method to the evaluation and generalization of the SCC for random variables of the fragility curve.