Most natural granular deposits are spatially variable due to heterogeneities in soil hydraulic conductivity, layer thickness, relative density, and continuity. However, existing simplified liquefaction evaluation procedures treat each susceptible layer as homogeneous and in isolation, neglecting water flow patterns and displacement mechanisms that result from interactions among soil layers, the groundwater table, foundation, and structure. In this paper, three-dimensional, fully coupled, nonlinear, dynamic finite-element analyses, validated with centrifuge experimental results, are used to evaluate the influence of stratigraphic layering, depth to the groundwater table, and foundation-structure properties on system performance. The ejecta potential index (EPI) serves as a proxy for surface ejecta severity within each soil profile. The results reveal that among all the engineering demand parameters (EDPs) and geotechnical liquefaction indices considered, only EPI predicted a substantial change in the surface manifestation of liquefaction due to changes in the location of the groundwater table and soil stratigraphy. This trend better follows the patterns from case history observations, indicating the value of EPI. Profiles with multiple critical liquefiable layers at greater depths resulted in base isolation and reduced permanent foundation settlement. Ground motion characteristics have the highest influence on EDPs, among the properties considered. The outcropping rock motion intensity measures with the best combination of efficiency, sufficiency, and predictability were identified as cumulative absolute velocity (for predicting foundation's permanent settlement and free-field EPI) and peak ground velocity (for peak excess porepressure ratio). These results underscore the importance of careful field characterization of stratigraphic layering in relation to the foundation and structural properties to evaluate the potential liquefaction deformation and damage mechanisms. The results also indicate that incorporating EPI alongside traditional EDPs shows promise. [ABSTRACT FROM AUTHOR]