Fragility analysis of slopes exposed to seismic hazard employing surrogate modeling techniques

Natural disasters may lead to failure of critical assets part of railway systems. Slope failure often results in major economic consequences. Due to the economic importance that railways systems have, it is pertinent to ensure good performance and long‐term. The objective of the present research is to propose tools that may allow obtaining the failure probability of soil slopes under a seismic event by employing surrogate models to ease the computational cost of considering the uncertainties associated with geotechnical and geometrical properties of a case study located in Portugal. The presented methodology and derived fragility curves, using peak ground acceleration (PGA) as the intensity measure (IM), can be used to assess slope performance under a seismic event for different safety levels, thus providing useful information for prioritizing assets and taking preventive actions to maintain the desired performance of the railway system.