In assessing the seismic performance of geotechnical systems, engineers often use analytical models to estimate the amount of seismically-induced displacements. These models consider system properties, earthquake parameters, and ground motion intensity measures (IMs) as inputs and have been typically formulated using “traditional” statistical techniques. This study discusses a new set of machine learning (ML) based models to estimate seismically-induced slope displacements in subduction and shallow crustal earthquake zones and liquefaction-induced building settlements in shallow crustal earthquake zones. Machine learning is used to select efficient features that explain seismically-induced displacements and settlements. The feature selection suggests no significant gain in accuracy beyond a small subset of features. Based on the selected features, a set models is developed considering several ML-based techniques with varying flexibility, interpretability, and bias-variance trade-offs. The developed models are assessed by evaluating test errors, their scaling compared to existing models, and their performance in case histories. The developed ML-based models contribute to performance-based assessments and also enhance the treatment of epistemic uncertainties in estimating seismically-induced displacements. Lastly, as discussed in the paper, caution should be exercised when assessing ML models as they could be perceived as a “black box” without proper context and extrapolate inappropriately.

Earthquake-induced soil liquefaction can cause settlement around piles, which can translate to negative skin friction and the development of drag load and settlement of the piles. The overall phenomenon and the interaction of the individual mechanisms in time are still not fully understood, leading to either conservative or unconservative designs in practice. A series of centrifuge model tests were performed to assess liquefaction-induced downdrag and understand the interplay and effects of (i) pile embedment and pile-head load, (ii) excess pore pressure generation and dissipation; and (iii) reconsolidation and ground settlement on pile response during and post shaking. It was found that most of the pile settlements were co-seismic while full liquefaction (excess pore pressure ratio of 100%) is not a prerequisite for the development of significant drag loads. Furthermore, the patterns of pore water diffusion elucidated how they can either exacerbate or alleviate excess pore pressures in various soil layers. Informed by the results of the centrifuge testing program, follow on investigations focused on developing a validated TzQzLiq numerical model that essentially accounting for the change in the pile’s shaft and the tip capacity as free-field excess pore pressures develop/dissipate in soil. The numerical model was later leveraged to develop a simplified displacement-based design procedure for axially loaded piles. Ongoing advanced fully coupled nonlinear numerical models investigate the state-of-the-art in accounting for the aforementioned mechanisms and their potential for translation to practice. The paper presents an overview of these efforts.