Engineers often estimate the amount of liquefaction-induced building settlements (LIBS) as a performance proxy to assess the potential of earthquake-induced damage to buildings. The first robust LIBS models were initially developed in 2017 and 2018 using traditional statistical approaches. More recently, machine learning techniques have started to be used in developing LIBS models. These recent efforts are a step forward in realizing the potential of machine learning in liquefaction engineering; however, they have often considered only one ML technique for a given dataset and typically used only held-out test sets for model assessment. In this study, five ML-based LIBS models with varying flexibility (i.e., ridge regression, partial least square regression - PLSR, random forest, gradient boosting decision tree - GBDT, and support vector regression) are developed using a LIBS database generated by soil-structure numerical simulations of different buildings and soil profiles shaken by ground motions with varying intensity measures. The motivation for considering models with different flexibility is to include different bias-variance trade-offs. Feature selection with different ML techniques indicates that cumulative absolute velocity, spectral acceleration at one second, contact pressure, foundation width, the thickness of the liquefiable layer, and a shearing liquefaction index are important features in estimating LIBS. The developed ML-based models are assessed considering prediction accuracy in test sets, performance against centrifuge tests and case histories, and trends. The assessment indicates that the random forest, GBDT, and SVR models perform best, providing standard deviation reductions up to 40% relative to a multi-linear regression. Specifically, the random forest and GBDT models exhibit a root mean square error (RMSE) of 0.29 and a coefficient of determination (R2) of 0.93 on test sets, demonstrating a notable improvement compared to a traditional multi-linear regression model, which yields an RMSE of 0.47 and an R2 of 0.82. Moreover, random forest and GBDT, alongside SVR, show a good performance in centrifuge tests and case histories. Finally, given the scarcity of LIBS models, this study also contributes to treating epistemic uncertainties in estimating LIBS, which is ultimately beneficial for performance-based assessments.

A database of detailed liquefaction ejecta case histories for the 2010-2011 Canterbury earthquakes is interrogated. More than 50 mm of ejecta-induced settlement occurred at thick, clean sand sites shaken by PGA6.1 = 0.35-0.70 g (wherein PGA6.1 is the peak ground acceleration for a Mw 6.1 earthquake), whereas ejecta-induced settlement at highly stratified silty soil sites did not exceed 10 mm even when PGA6.1 exceeded 0.45 g. Cone penetration test-based liquefaction-induced damage indices that do not consider soil-system response effects, such as post-shaking hydraulic mechanisms, overestimate the severity of ejecta at stratified silty soil sites. Considering post-shaking hydraulic mechanisms captures the lack of ejecta at stratified silty soil sites. It also improves the estimation of ejecta severity at clean sand sites with severe-to-extreme ejecta. Strongly shaken clean sand sites that did not produce ejecta typically had thick strata with high tip resistances, thick non-liquefiable crusts, or deeper non-liquefiable strata overlying liquefiable strata. Ejecta-induced fissures formed in the nonliquefiable crust during the Feb 2011 earthquake which liquefied soil at depth could exploit to produce ejecta during the Jun 2011 earthquake. When significant ejecta formed on the roads, elevated adjacent ground with houses typically had negligible ejecta.