The T-shaped strip footing is a good choice for building foundations because it effectively resists overturning forces and accommodates eccentric loading. The footing's embedded within the soil enhances its capacity to counteract the forces generated by eccentric loading, providing additional stability and support. This study presents a couple of finite element limit analysis (FELA) and regression machine learning models in predicting the bearing capacity of eccentrically loaded T-shaped footings on anisotropic clays. A numerical simulation of T-shaped strip footings in anisotropic clay under eccentric loading is performed using a FELA software, namely OptumG2. At the same time, the regression soft computing models employed four techniques, including the genetic programming (GP), age-layered population structure-genetic algorithm (ALPS-GA), offspring selection genetic algorithm (OSGA), and grammatical evolution (GE) models. The AUS yield criterion is utilized to govern the soil properties, while the footing is modeled as a rigid material. By emphasizing the stability of the surrounding soil, this study neglects the failure of the footing itself since the footing is assumed to be very rigid. Parametric analyses are conducted using a dimensionless approach. The influences of eccentricity (e/B), the insertion length ratio (D/B), the anisotropic strength ratio (re), and the adhesion factor (alpha) on the bearing capacity factor (N) are investigated. The impact of these dimensionless parameters on the shear dissipation of the model to monitor the failure pattern is discussed. The current results are compared with prior solutions, showing consistency. Moreover, predictive regression machine learning techniques (GP, ALPS-GA, OSGA, and GE models) are applied to develop empirical equations for N estimation, with the proposed OSGA model demonstrating superior performance, achieving coefficients of determination (R2) of 0.985 and 0.984 for the training and testing sets, respectively.

Braced excavation is a common practice in underground construction for buildings and tunneling in urban areas. To determine the stability and failure behavior of the surrounding soil around the excavation, ensuring that it does not collapse and maintains safety, this study focuses on investigating the undrained stability of a fully supported 3D rectangular excavation with an embedded wall in anisotropic clay. The analysis is conducted through three-dimensional finite element limit analysis (3D FELA). To model the failure behavior of soils, the anisotropic undrained shear strength (AUS) failure criterion is employed in FELA. The factors influencing excavation failure are categorized into four dimensionless variables: the anisotropic strength ratio (re), the embedded rigid wall ratio (D/H), the aspect ratio (A/B), and the depth ratio (H/B). The solutions of the stability number (N), as well as the failure mechanism of 3D excavations, are derived from both the lower bound (LB) and upper bound (UB) approaches. This paper also introduces a novel machine learning approach for predicting the stability of supported rectangular excavations in anisotropic clays. Moreover, this study investigates the potential of using an advanced machine learning model which is XGBoost technique. The gathered evidence indicates that XGBoost demonstrates remarkable precision in forecasting the stability of supported excavation in anisotropic clays.

This study aims to explore the significant impact of soil fabric anisotropy on the ultimate bearing capacity of eccentrically and obliquely loaded shallow foundations overlying a geosynthetic-reinforced granular deposit. For this purpose, the well-established lower bound theorems of limit analysis (LA) in conjunction with the finite elements (FE) formulations and second-order cone programming (SOCP) are exploited to perform the bearing capacity estimations. The consideration of the soil mass's inherently anisotropic response in the granular layer involves the utilization of distinct internal friction angles in various directions. The lower bound FELA framework adopted in this study incorporates both the pull-out and tensile mechanisms of failure in the reinforcement layer. The marked contribution of soil inherent anisotropy to the impacts of ultimate tensile strength (Tu) T u ) and embedment depth (u) u ) of the geosynthetic reinforcement on the failure mechanism, bearing capacity ratio (BCR), BCR ), and failure envelope of the overlying obliquely/eccentrically strip footing is rigorously examined and discussed. It is generally concluded that for a given embedment depth, failure envelopes of the surface footing in both V-H H and V-M M planes shrink appreciably with the increase in the soil anisotropy ratio as well as the decrease in the geosynthetic ultimate tensile strength. Moreover, the influence of soil inherent anisotropy on the overall bearing capacity of shallow foundations is more evident in the case of using strong reinforcement compared to the weak geosynthetic. The findings of this investigation demonstrate that overlooking the soil inherently anisotropic behaviour in the numerical analysis of shallow foundations would give rise to undesirable non-conservative and precarious designs.

A finite element limit analysis (FELA) combined with a machine learning technique is adopted to determine failure envelopes of a strip footing resting on anisotropic soil under combined loadings (V, H, M). Based on numerical results from 2D FELA, the parametric studies on the failure envelopes in the 2D space (H, M) and 3D space (V, H, M) considering the effect of anisotropic behaviors of clays and interface interaction between soil and structure, are investigated. Additionally, the failure mechanisms of the investigated strip footings are also illustrated. These results can enhance the understating of practical engineering for designing a strip footing resting on anisotropic soil under combined loadings (V, H, M). In the later part, an artificial neural network (ANN) is adopted to propose the correlation equation between input parameters and their corresponding outputs using the data from FELA. Based on the developed ANN model, building the failure envelopes of strip footing subjected to general load (V, H, M) becomes quicker than the traditional methods. For more detail, the careful explanation for applying ANN results is presented through the design example. It can be a valuable procedure for practical engineers to establish the failure envelope of a footing on anisotropic clay under combined loading (V, H, M).

The uplift capacity of pipeline systems in geotechnical engineering is influenced by internal loading and external factors, making it a significant consideration in pipeline design problems. Previous research has conducted experimental tests and numerical solutions to investigate the relationship between force and displacement or the resistance of pipelines in numerous soil media. This paper proposes a machine-learning regression technique to predict the uplift capacity of buried pipelines in anisotropic clays with parametric analysis. Specifically, the Multivariate Adaptive Regression Spline (MARS) is employed to establish the relationship between input parameters, namely the depth ratio (H/D), anisotropic strength ratio (re), load inclination (beta), overburden factor (gamma H/Suc), adhesion factor (alpha), and the output uplift resistance (N) obtained from the finite element limit analysis (FELA), utilizing the AUS material model integrated with the OptumG2 commercial program. Furthermore, the sensitivity analysis outcome shows the embedded depth ratio is the most critical parameter, followed by the anisotropic strength ratio, overburden factor, load inclination, and adhesion factor. Additionally, the shear velocity field contours show that when the depth ratio and the load inclination increase, the dissipation of shear changes. Design data visualizations, tables, graph contours, and empirical equations are created and can be utilized to aid in the development of practical designs.