Slope Models in Finite Element Mesh Subjected to Numerous Earthquake Records

This DesignSafe published dataset is associated with research aimed at developing simulation-based predictive models of earthquake-induced slope displacements through classical regression and artificial neural network (ANN) approaches. Towards this end, finite element modeling was used to build a displacement database associated with different slope geometries, soil properties, and ground motion characteristics. A total of 49 slope geometries were analyzed, with each subjected to 1,051 earthquake motions. The maximum resultant displacements on the slope surface are collected for the 51,499 analyses performed. These displacements were used to develop statistical models that predict displacement as a function of the ground motion intensity and slope characteristics. These statistical models were developed by the least-squares method and by the artificial neural network (ANN) approach.