Automatic classification of near‐fault pulse‐like ground motions

This study presents an automated, quantitative classification method for near-fault pulse-like ground motions, distinguishing between forward-directivity and fling-step (FS) motions. The method introduces two novel parameters—the pulse velocity ratio and pulse area ratio—which transform the classification standard from a qualitative to a quantitative framework. Combined with an enhanced pulse extraction technique that captures permanent displacement characteristics, these parameters significantly improve classification efficiency and repeatability. This automated approach overcomes the limitations of manual classification, providing reproducible results. The identified FS ground motions can be applied to the dynamic analysis of cross-fault structures, enhancing the reliability of seismic hazard assessments.