Seismic actions are usually considered for their inertial effects on the built environment. However, additional effects may be caused by the volumetric-distortional coupling of soil behaviour: the fast cyclic shaking on saturated soils caused by earthquakes generates temporary undrained or quasi-undrained conditions and subsequent pore pressure variations that, if positive, reduce the effective stresses, eventually leading loose granular soils to liquefaction. Whatever the amount of seismically induced pore pressure build up, buildings on shallow foundations suffer settlements and tilts that may be extremely large when soils approach liquefaction, as demonstrated by several recent case histories. The paper proposes an equivalent elastic approach in effective stresses to predict the co-seismic (undrained) component of the seismically induced settlement of shallow foundations, which usually is the most relevant one, by considering the decrease of soil stiffness during the seismic event. The total settlement can be then estimated by adding the post-seismic (drained) component, also evaluated in this paper via a quite simple approach. Even though the equivalent elastic model is stretched into a highly non-linear soil behaviour range, especially when the soil is approaching liquefaction, the model considers the relevant capacity and demand factors and proved effective in simulating some centrifuge tests published in the literature. In the paper, the simplifying assumptions of the approach are clearly indicated, and their relevance discussed. It is argued that notwithstanding some limitations the model is physically based and therefore it allows for understanding and checking the relative relevance of all the parameters related to soil, foundation, and seismic action. Thus, it is a tool of possible interest in the design of shallow foundations in liquefaction-prone seismic areas.

In this study, a novel data-driven approach is carried out to predict the pore pressure generation of liquefiable clean sands during cyclic loading. An extensive and comprehensive database of actual stress-controlled cyclic simple shear test results in terms of pore pressure time histories is gathered from a large number of experiments. While the classical machine learning (ML) algorithms help predict the number of liquefaction cycles in a few models, the desired level of accuracy in predicting the actual trend and robustness in pore pressure build-up is only achieved in deep learning (DL) methods. Results indicate that the Long-Short Term Memory (LSTM) working model, employed with Stacked LSTM and the Windowing data processing method, is necessary for making fairly good cyclic pore pressure build-up predictions. This study proposes a model that can ultimately be utilised to predict the pore pressure response of in-situ liquefiable sandy soil layers without resorting to plasticity-based complex theoretical models, which has been the current practice. The robustness achieved in the model reassures the reliability of the study, raising confidence in developing data-driven constitutive models for soils that have the potential to replace conventional plasticity-based theories.