This study utilizes a combined approach of Finite Element Method (FEM) simulation and Artificial Neural Network (ANN) modeling to analyze and predict the load-displacement relationship of bored piles in clayey sand. FEM is applied to simulate the nonlinear relationship between load and vertical displacement, with input parameters including load and the mechanical properties of the soil. The results obtained from FEM are used as input data for the ANN model, enabling accurate predictions of vertical displacement based on these parameters. The findings of this study show that the predicted ultimate bearing capacity of the bored piles is highly accurate, with negligible error when compared to field experiments. The ANN model achieved a high level of accuracy, as reflected by an R2 value of 0.9992, demonstrating the feasibility of applying machine learning in pile load analysis. This research provides a novel, efficient, and feasible approach for analyzing and predicting the bearing capacity of bored piles, while also paving the way for the application of machine learning in geotechnical engineering and foundation design. The integration of FEM and ANN not only minimizes errors compared to traditional methods but also significantly reduces time and costs when compared to field experiments.

This direction paper explores the evolving landscape of physics-informed machine learning (PIML) methodologies in the field of geotechnical engineering, aiming to provide a comprehensive overview of current advancements and propose future research directions. Recognising the intrinsic connection between geophysical phenomena and geotechnical processes, we delve into the inter of physics-based models and machine learning techniques. The paper begins by elucidating the significance of incorporating physics-informed approaches, emphasising their potential to enhance the interpretability, accuracy and reliability of predictive models in geotechnical applications. We review recent applications of PIML in soil mechanics, hydrology, geotechnical site investigation, slope stability analysis and foundation engineering, showcasing successes and challenges. Furthermore, we identify promising avenues for future research in geotechnical engineering, including the integration of domain knowledge, model explainability, multiphysics and multiscale problems, complex constitutive models, as well as digital twins and large AI models within PIML frameworks. As geotechnical engineering embraces the paradigm shift towards data-driven methodologies, this direction paper offers valuable insights for researchers and practitioners, guiding the trajectory of PIML for sustainable and resilient infrastructure development.

Biopolymer-based soil treatment (BPST) enhances soil strength through biofilm matrix formation within soil voids. This study investigates the effects of biopolymer concentration, porosity, and soil packing conditions on biopolymer distribution and connectivity after dehydration. Laboratory experiments assessed the degree of biopolymer filling (DoBF), final condensed biopolymer concentration, and biopolymer film connectivity under simple cubic and rhombohedral packing conditions. The results show that higher initial biopolymer concentrations increase final biopolymer volume, though not proportionally due to threshold effects. Rhombohedral packing results in higher final condensed biopolymer concentrations than simple cubic packing, despite having lower DoBF values, while biopolymer connectivity peaks at an optimal porosity (n approximate to 0.35). Further analysis revealed a strong correlation between biopolymer matrix formation and soil mechanical properties, including uniaxial compressive strength (UCS), cohesion, and friction angle. UCS was found to decrease with increasing porosity, and a predictive model was developed using experimental data. The rhombohedral and simple cubic packing conditions respectively define the upper and lower bounds of the shear parameters. A back-calculation approach confirmed that DoBF provides the most accurate estimation of friction angle and UCS, reinforcing its importance as a key parameter in soil stabilization. These findings emphasize the need for optimized biopolymer concentration and soil structure adjustments to enhance reinforcement efficiency. The study offers valuable guidance for geotechnical applications, enabling the development of optimized biopolymer injection strategies that enhance mechanical performance and promote efficient material utilization.

This study focuses on the behaviour of buried gas pipelines subjected to surface loading. The study is oriented towards an experimental campaign carried out on small-scale pipelines, with three different wall thicknesses, both in monotonic and cyclic conditions. Pipes have been instrumented with strain gauges and inner displacement sensors, allowing to record deformations, stresses and ovalisation of the pipe, in addition to the load-settlement relationship at the soil surface. Results show that the presence of the pipe affects the global soil response (stiffness and bearing capacity). Analysis of the strain distribution and pipe deformed shape indicate that the pipe response is complex, with no symmetry along the horizontal axis, and a heart-shaped deformation pattern. The pipe rigidity affects the local behaviour at the pipe level (displacement pattern, evolution of stresses during cyclic loading and increasing lateral support). Classical pipeline design theory has been assessed based on the experimental observations, invalidating several underlying hypotheses.

While many employ a hyperbolic stress-strain relationship for soils, it is known that such a relationship is accurate over either the small strain range as encountered in earthquake and soil dynamics problems or a relationship with different input parameters that are needed over large strains as is required for finite element analyses of large deformation behavior. The two characterizations do not become one. A proposed power relationship is presented that was developed to characterize the triaxial test stress-strain behavior of cohesionless material from lubricated or frictionless cap and base tests (some 144 tests) covering a range in the natural variation in particle size, particle shape and surface roughness, over low to high confining pressure. This relationship covers the range in strain from 10(-6) to soil failure. It has been used successfully to date in laterally loaded pile response characterization (the Strain Wedge Model) and shallow foundation load-settlement-bearing capacity response. Most recently, it has been extended to assess the behavior of rock-like material (caliche). The relationship and its comparison with the hyperbolic relationship for large strain and the shear modulus reduction curve for seismic behavior are presented here.

This paper focuses on enhancing the prediction of vertical soil displacement during deep excavation using Artificial Neural Networks (ANN). Precise prediction of soil movement is essential to ensure the stability of the construction site and surrounding infrastructure. Traditional methods, such as the Finite Element Method (FEM), while accurate, are time-consuming and computationally intensive. In contrast, ANN offers fast and reliable predictions, making it a valuable tool for real-time decision-making. This study integrates FEM-based data to train the ANN model, ensuring the ANN captures complex, non-linear interactions between input variables like depth, pore water pressure, and coordinates. The model is trained and evaluated using performance metrics such as MAE, MSE, RMSE, and R2. With a high correlation coefficient R2 = 0.969238, the ANN model provides predictions with minimal error, demonstrating its effectiveness in replicating real-world measurements. The combined approach of ANN and FEM leverages the strengths of both methods, with FEM offering detailed physical insights and ANN optimizing computational efficiency. The results indicate that ANN-based models can serve as an efficient predictive tool in large-scale construction projects, improving safety by anticipating potential soil displacement issues. Future research will focus on expanding the model's applicability across different soil conditions and enhancing prediction capabilities with other machine learning algorithms.

The recession of a sandy bluff was investigated in a controlled laboratory wave flume, replicating the complex interactions between hydrodynamic forcing, sediment transport processes, and bluff slope stability. A comprehensive monitoring approach measured water levels, pore water pressures, moisture content, and detailed bathymetric-topographic data, providing a thorough understanding of the governing mechanisms and their interrelationships within the beach-bluff system. Bluff recession occurred through notch formation at the bluff toe, followed by a series of minor and major episodic bluff failures. Pore-water pressure variations within the bluff were closely linked to morphological changes on the beach and the bluff's instability. The final beach profile exhibited distinct characteristics: near the shoreline, it was steeper than the equilibrium beach profile due to the sediment supplied by bluff recession. Cross-spectral analysis between water level fluctuations and pore water pressure signals revealed a strong coupling between incident wave energy and pore water pressure responses within the beach-bluff system. The rapid rise in saturation, along with the formation and expansion of the notch, contributed to bluff instability and episodic failure events.

This study demonstrates the feasibility of utilizing machine learning (ML) for routine identification of sand particles. Identifying different types of sand is necessary for various geotechnical exploration projects because understanding the specific sand type plays an important role in estimating the physical and mechanical properties of the soil. To accomplish this, dynamic image analysis was employed to generate a substantial volume of sand particle images. Individual size and shape descriptors were automatically extracted from each particle image. The analysis involved use of 40,000 binary particle images representing 20 different sand types, and a corresponding six size and four shape descriptors for each particle (400,000 parameters). Six ML models were trained and tested. The work demonstrates that using size and shape features the models efficiently identified up to 49% of individual sand particles. However, when clusters of particles were considered in conjunction with a voting algorithm, classification accuracy significantly improved to 90%. Among the ML models studied, neural networks performed the best, while decision tree exhibited the lowest accuracy. Finally, the use of size consistently outperformed shape as a classification parameter but combining size and shape parameters yielded superior results across all sands and classifiers. These findings suggest that ML holds much promise for automating sand classification using ordinary images.

There are a large number of microorganisms such as bacteria and fungi in the soil, which affect the physical and mechanical properties of the rock and soil. Microbial solidification technology is the use of microbial metabolism to induce mineral precipitation, thereby changing the soil structure and improving the physical and mechanical properties of the soil. This article uses microbial activated magnesium oxide solidification technology to treat red clay samples, and explores the effects of magnesium oxide content, bacterial solution concentration, and initial moisture content on the shear strength and disintegration of red clay. The experimental results are explained reasonably through scanning electron microscopy experiments and ImageJ quantitative analysis software. The experimental results show that the shear strength of red clay is positively correlated with the content of magnesium oxide and bacterial solution concentration, but negatively correlated with the initial moisture content; The hydrated magnesium carbonate generated in the experiment is the key reason for the improvement of shear strength. Hydrated magnesium carbonate can play a role in bonding red clay particles and filling the pores of red clay; Significant reduction in disintegration of microbial magnesium oxide solidified red clay.

This article explores the role of artificial intelligence (AI) in predicting nanomaterial properties, particularly its significance within geotechnical engineering. By analyzing multiple AI-based studies, the review concentrates on the forecasting of nanomaterial-altered soil characteristics and behaviors. Encouraging findings from these studies underscore AI's ability to accurately predict the geotechnical properties of nanomaterials, though challenges remain, particularly in quantifying nanomaterial percentages and their implications across various applications. Future research should address these challenges to enhance the accuracy of AI-based prediction models in geotechnical engineering. Nonetheless, the growing adoption of AI for predicting nanomaterial properties demonstrates its potential to revolutionize geotechnical engineering. AI's capacity to uncover intricate patterns and relationships beyond human capabilities enables more precise soil behavior predictions, fostering innovative solutions to geotechnical challenges. Its ability to process vast datasets, adapt to various scenarios, and continuously learn from new information makes AI an indispensable tool for understanding nanomaterial properties and their impact on soil behavior. In summary, the integration of AI and geotechnical engineering represents a pivotal advancement in comprehending nanomaterial properties and their practical applications. As research advances and AI technologies evolve, transformative progress in geotechnical engineering is expected. By harnessing AI's capabilities, researchers can unlock groundbreaking insights, drive innovation, and shape a more resilient and sustainable future for the geotechnical engineering industry.