Geotechnics, Volume 4, Issue 2 (June 2024) – 8 articles

The study explores the capabilities of large language models (LLMs), particularly GPT-4, in understanding and solving geotechnical problems, a specialised area that has not been extensively examined in previous research. Employing a question bank obtained from a commonly used textbook in geotechnical engineering, the research assesses GPT-4’s performance across various topics and cognitive complexity levels, utilising different prompting strategies like zero-shot learning, chain-of-thought (CoT) prompting, and custom instructional prompting. The study reveals that while GPT-4 demonstrates significant potential in addressing fundamental geotechnical concepts and problems, its effectiveness varies with specific topics, the complexity of the task, and the prompting strategies employed. The paper categorises errors encountered by GPT-4 into conceptual, grounding, calculation, and model inherent deficiencies related to the interpretation of visual information. Custom instructional prompts, specifically tailored to address GPT-4’s shortcomings, significantly enhance its performance. The study reveals that GPT-4 achieved an overall problem-solving accuracy of 67% with custom instructional prompting, significantly higher than the 28.9% with zero-shot learning and 34% with CoT. However, the study underscores the importance of human oversight in interpreting and verifying GPT-4’s outputs, especially in complex, higher-order cognitive tasks. The findings contribute to understanding the potential and limitations of current LLMs in specialised educational fields, providing insights for educators and researchers in integrating AI tools like GPT-4 into their teaching and problem-solving approaches. The study advocates for a balanced integration of AI in education to enrich educational delivery and experience while emphasising the indispensable role of human expertise alongside technological advancements. Full article

Understanding the role that settlement can have on the base resistance of piles is a crucial matter in the design and safety control of deep foundations under various buildings and infrastructure, especially for long to super-long piles (60–90 m length) in soft soil. This paper presents a novel assessment of this issue by applying explainable machine learning (ML) techniques to a robust database (1131 datapoints) of fully instrumented pile tests across 37 real-life projects in the Mekong Delta. The analysis of data based on conventional methods shows distinct responses of long piles to rising settlement, as compared to short piles. The base resistance can rapidly develop at a small settlement threshold (0.015–0.03% of pile’s length) and contribute up to 50–55% of the total bearing capacity in short piles, but it slowly rises over a wide range of settlement to only 20–25% in long piles due to considerable loss of settlement impact over the depth. Furthermore, by leveraging the advantages of ML methods, the results significantly enhance our understanding of the settlement–base resistance relationship through explainable computations. The ML-based prediction method is compared with popular practice codes for pile foundations, further attesting to the high accuracy and reliability of the newly established model. Full article

This paper examines the effects of collapsible soil structure on shear wave velocity. The study attempts to simulate hydraulic fill sand deposits, which represent a natural soil deposition process that can result in a collapsible soil structure. A series of resonant column tests and bender element tests on Ottawa sand was conducted on sand specimens and prepared by dry pluviation and simulated hydraulic fill methods subjected to various confining pressures. Shear wave velocities measured from both methods of deposition are compared and discussed. Results from this study show that for soil specimens with the same void ratio, samples prepared by simulated hydraulic fill have a lower shear modulus and shear wave velocity than the specimens prepared by dry pluviation, and the differences are more pronounced at higher confining pressures. The resonant column test results performed in this study were consistent with results from the discrete element analysis, full-scale testing, and centrifuge testing. The discrete element analysis suggests that soil fabric and number of particle contacts are the key factors affecting the shear wave velocity. These factors are dependent on the methods of deposition. Results from this study examining hydraulic fill collapsible structure shear wave velocity provide a step forward toward a better correlation between soil dynamic properties measured in field and laboratory tests. Full article

This study investigated the effect of compaction effort and soaking time on the shear strength properties of fine-grained gypsum-containing soils. The objective was to demonstrate that increasing compaction effort increases soil strength, specifically cohesion and the angle of shear strength, when subjected to soaking in freshwater. Unconsolidated undrained triaxial tests were carried out on CBR soil samples with different soaking times. The results showed a transition from brittle to ductile failure behaviour as the soaking time increased. Mohr–Coulomb failure envelopes showed reduced cohesion and angle of shear strength with increasing soak time. Regression models were developed to establish correlations between soaked and unsoaked strength parameters. Strong relationships were found between soil strength properties, compaction effort and soaking time. Empirical equations were proposed to estimate the cohesion and angle of shear strength from compaction effort and soaking time. This study highlighted the importance of considering gypsum-rich soils in civil engineering design. Gypsum dissolution during wetting significantly affected soil strength parameters. The regression models and empirical equations provide engineers with tools to assess the influence of compaction effort and soaking time on soil strength, thus aiding decision making when designing structures on gypsum-rich soils. Full article

Electrical resistivity tomography (ERT) has turned out to be one of the most applied and user-friendly geophysical methods in geotechnical and geoenvironmental research. ERT is an emerging technology that is becoming popular nowadays for investigating subsurface conditions. Multiple attributes of the technology using various electrode configurations significantly reduce measurement time and are suitable for applications even in hardly accessible mountain areas. It is a noninvasive test for subsurface characterization and a very sensitive method used to determine geophysical properties, i.e., structural integrity, water content, fluid composition, etc. This paper aimed to elucidate the ERT technique’s main features and applications in geotechnical and geoenvironmental engineering through four case studies. The first case study investigated the possible flow paths and areas of moisture accumulation after leachate recirculation in a bioreactor landfill. The second case study attempted to determine the moisture variation along highway pavement. The third case study explored the slope failure investigation by ERT. The fourth case study demonstrated the efficiency of the ERT method in the landfill evapotranspiration (ET) cover to investigate moisture variation on a broader scale and performance monitoring. In all of the four cases, ERT exhibited promising performance. Full article

This paper assesses the effectiveness of different unsupervised Bayesian changepoint detection (BCPD) methods for identifying soil layers, using data from cone penetration tests (CPT). It compares four types of BCPD methods: a previously utilised offline univariate method for detecting clay layers through undrained shear strength data, a newly developed online univariate method, and an offline and an online multivariate method designed to simultaneously analyse multiple data series from CPT. The performance of these BCPD methods was tested using real CPT data from a study area with layers of sandy and clayey soil, and the results were verified against ground-truth data from adjacent borehole investigations. The findings suggest that some BCPD methods are more suitable than others in providing a robust, quick, and automated approach for the unsupervised detection of soil layering, which is critical for geotechnical engineering design. Full article

The Wave Equation Analysis of Pile Driving (WEAP) has been widely used to determine drivability, predict static resistance, and assure the integrity of piles in soils. Assigning static and dynamic properties of Soil-based Intermediate Geomaterials (S-IGMs) remains a challenge in WEAP, partly attributed to IGMs that act as transition geomaterials between soil and hard rock. Furthermore, reliable static analysis methods for unit resistance predictions are rarely available for driven piles in S-IGMs in the default WEAP method. To alleviate these challenges, this study presents improved WEAP methods for steel piles driven in S-IGMs, including proposed damping parameters and Load and Resistance Factor Design (LRFD) recommendations based on newly developed static analysis methods and the classification of S-IGMs. A back calculation approach is used to generate the appropriate damping parameters for S-IGMs for three distinct subsurface conditions utilizing a database of 34 steel H- and pipe piles. Newly developed WEAP and LRFD procedures are also recommended. Additional independent 22 test pile data are used to compare and evaluate the accuracy and efficiency of the proposed WEAP methods with the default WEAP method. Compared with the default WEAP, bearing graph analysis results revealed that the selected proposed WEAP method, on average, reduces the underprediction of pile resistances by 6% and improves the reliability with a 43% reduction in the coefficient of variation (COV). Calibrated resistance factors for the proposed WEAP method increase to as high as 0.75 compared to the current AASHTO recommendation of 0.50. An economic impact assessment reveals that the proposed WEAP method is more efficient than the default WEAP method as the average difference in steel weight for 32 test piles is 0.06 kg/kN, almost close to zero, reducing the construction challenges in the current engineering practice. Full article

The current scallop fishery sector allows many scallops to remain in specified fishing zones, and this process leads to heavy losses in the sector. Scallop fishermen aim to harvest the remaining scallops to reduce their losses. To achieve this, a fisherman must understand the scallop ecology on the seafloor. In our previous study, we proposed a method for measuring scallops using wheeled robots. However, a wheeled robot must be able to resist disturbance from the sea to achieve high measurement accuracy. Strong anchoring of wheels against the seafloor is necessary to resist disturbance. To better understand anchoring performance, we confirmed the wheel anchoring capacity in water-containing sand in an experiment. In this experiment, we towed fixed wheels on water-containing sand and measured the resistance force acting between the wheel and the sand. Afterward, we considered the resistance force as the wheel anchoring capacity on the water-containing sand. The experimental results capture the tendency for the anchoring capacity of sand with/without water to increase with sinkage. The results also demonstrate that the anchoring capacity of water-containing sand is lower than that of non-water-containing sand. However, the results indicate that when the wheels possess lugs, their presence tends to increase the wheels’ anchoring capacity in water. Full article