Geo-Congress 2024: Foundations, Retaining Structures, Geosynthetics, and Underground Engineering

This paper presents a novel approach for predicting the lateral capacity of large-diameter monopiles in multi-layered soil using deep learning and transfer learning techniques. With the increasing interest in offshore wind energy, the cost-effective design of offshore wind turbine foundations is crucial. Traditional methods such as the p-y method have limitations in analyzing larger diameter monopiles. As an alternative to high-fidelity numerical models (e.g., finite element analysis and finite volume analysis), deep learning models have gained popularity for load response predictions and design purposes. Yet, one major challenge of deep learning models is that insufficient data can result in poor model performance. In this study, a deep learning model incorporating convolutional and fully connected layers was developed to capture the complex interactions between pile geometry parameters and soil conditions. To address the challenge of limited dataset size, transfer learning was employed, leveraging a pretrained model on a large dataset. The results demonstrate that the proposed model can provide accurate predictions, even with a small dataset. Transfer learning significantly reduces the required data size while preserving high prediction accuracy.

Pile resistance during and immediately after driving in cohesive soils is often significantly lower than the eventual resistance that develops in the long term. This time-dependent increase in resistance, commonly called pile setup, can lead to large differences between dynamic estimates of geotechnical pile resistance at the end of pile driving (EOD) and the actual pile resistance after setup has occurred. In this study, data from 28 dynamic pile load tests performed at nine bridge sites in northern Missouri were compiled and analyzed to characterize and better understand the setup characteristics. Setup was observed in 27 of the 28 piles. The average setup factor at times greater than 60 h for piles driven in clay profiles was 1.75 with a standard deviation of 0.3, while setup factors for piles driven in mixed soil conditions were more variable with a standard deviation of 0.45 and a slightly higher average setup of 1.86. The setup in clay profiles is shown to correlate well with average pocket penetrometer strength values.

This paper investigates the effect of the pile modeling approach on the factor of safety of a clay slope stabilized with deep foundations using three-dimensional numerical analysis. The short-term stability was considered due to the surcharge load on the crest. In the analysis, the piles were modeled using both linear-elastic and elasto-plastic approaches for comparison purposes. The factor of safety, pile behavior, and pile internal forces were evaluated for each case. The results show that the linear elastic pile model does not allow the structural failure of the piles in factor of safety analysis that results in higher factor of safety than the critical failure mode of the pile-stabilized slope. Modeling elasto-plastic reinforced concrete piles without considering reinforcement resulted in early development of multiple plastic hinges, and hence it is a conservative approach. The results indicate that the approaches that use elasto-plastic reinforced concrete pile models provide a better analysis of the pile-soil interaction which allows the model to evaluate the factor of safety for the more critical failure mechanism of the coupled slope-pile system.

When soils surrounding a deep foundation settle, a reversal of the direction of interface skin friction occurs over some upper section of the piles, thereby increasing both the axial stresses in the piles and the pile settlement, also known as downdrag. Limit analysis is generally ineffective at predicting pile settlements as it does not account for the dependency of mobilized friction on the relative displacement between the soil and pile surface. In earlier works, a hyperbolic equation was used to model the relationship between mobilized friction and the relative displacement between pile and soil. In this paper, the hyperbolic equation is extended to model the friction reversal. The hyperbolic cyclic equations are experimentally verified then incorporated into a spreadsheet program “t-zMich” with a new downdrag module. The iterative analysis accounts for both rigid translation and elastic compression of the pile to predict the mobilized side friction along the length of the pile and the overall downdrag.

The article addresses the axial load responses of helical piles subjected to pseudo-static monotonic and incremental cycles of loads through centrifuge tests in the sand. Also, a relative comparison was made with a similar test on a conventional smooth pile with no helices. Strain gauges as instrumentations placed over pile depth provided the axial load response profiles. All tests are conducted in flight at a 20 g scale. It was observed that both smooth and helix piles had significant loss of strength due to repeated cycles. Moreover, the theoretical equations overestimate the capacity of the piles through tests in the post-cyclic monotonic loading stage. Last, the internal ultimate load and unit shaft resistance distribution are shown along the pile depth. The loss of pile resistance was consistent along the depth for the post-cyclic loading stage.

A continuum-based analytical framework for a rigid circular pile embedded in an elastic, multi-layered soil continuum under static lateral load and moment is developed using the principle of virtual work. Soil displacement is chosen based on the geometry and kinematics of the pile, and the vertical displacement in the soil caused by the large-diameter pile is explicitly considered in the analysis. An iterative algorithm is used to obtain the solutions of the equations for the pile and soil displacement. The method is computationally efficient and provides accurate results for laterally loaded large-diameter rigid piles without adjusting the soil modulus. The equivalent lateral soil spring stiffnesses at the side of the pile are also derived analytically from the solution, and curve-fitted equations for the spring stiffnesses are provided for piles in homogeneous soil under different loading conditions.

Pre- and post-pile installation soil properties surrounding newly installed drilled displacement piles (DDPs) in sands were evaluated by interrogating cone penetration tests (CPT) performed before (CPT_PRE) and after (CPT_POST) pile construction. CPT raw measurements, that is, tip resistance (q_c), sleeve friction (f_s), and excess pore water pressure (u2), were used to quantify the change in soil resistance. The main parameters influencing the soil resistance improvement were identified as follows: the degree of confinement (depth), the soil resistance prior to the DDP installation (q_c.pre), and the pile arrangement and distance between the DDP and the location where the improvement is being assessed. Furthermore, the change in soil behavior was evaluated via Robertson’s normalized CPT soil behavior type (SBTn) chart along with the fines content (FC). The results showed that clean sands, with low FC improved slightly more than sands with noticeable FC (30%−40%) when the soil is in a dilative state prior to the pile installation.

This paper presents a case study detailing a field monitoring program conducted in Central Florida to analyze the effects of impact pile driving in sandy soils in terms of ground deformations and vibrations. The project consisted of a 48.8-m-long precast prestressed concrete pile driven by an impact hammer. A description of the project, the testing program, and pile installation for two consecutive days of driving are presented. The subsurface conditions mainly consisted of the presence of medium-dense poorly graded sands and silty sands with some interbedded clay layers. The instrumentation included geophones and survey stations to measure particle velocities and ground deformations, respectively. Records of force and energy time histories transmitted to the top of the pile using embedded data collectors are also reported. Conclusions are drawn regarding the peak particle velocity values recorded away from the pile as well as the ground deformations in terms of settlement and heave. The location of the tip of the pile and the pile penetration rate were the main triggering factors of ground deformations even when vibration levels were measured below local peak particle velocity limits.

Estimating pile length is critical when reusing unknown foundations. In this study, we provide a novel pile length estimate approach based on guided wave theory that uses only one sensor for data collection, addressing the inaccessibility of the pile top or lateral side for sensor installation as well as the need for accurate pile length calculation. The data acquisition transducer can be mounted on the pile’s lateral or top surface. The normalized magnitude of the recorded frequency domain responses is defined. The dispersion analysis diagram of the wavenumber as a function of the normalized magnitude is then determined using the dispersion relation provided by the guided wave model. The pile length can be computed at low frequencies using the periodic analysis of the dispersion analysis diagram and the period determined from the diagram. Numerical analysis can be used to validate the suggested method’s usefulness and correctness.

The design of deep foundation systems often warrants a simplified approach for analysis by structural engineers. The point-of-fixity method is widely used as a common design practice where the complex interaction between soil and deep foundation under lateral load is sought to be replaced by an equivalent, fixed-based cantilever model. Recent studies attempted to propose more robust procedures to calculate depth-to-fixity, depending on various factors affecting the nonlinear lateral response of the soil-foundation system, as opposed to using empirical equations from design codes. This study proposes a simple method to determine depth-to-fixity, based on an analytical solution to the equivalent cantilever model. Following the recommendations from past research, the new method utilizes numerical lateral response obtained from the finite-difference program, NVShaft, to effectively translate the complex nonlinear soil model into a simplified equivalent cantilever analysis. Case studies of three lateral load test programs were considered to evaluate the proposed method and preceding point-of-fixity models based on design codes and recent literature.

The disconnected piled raft (DPR) foundation system is a structural system where the raft and piles are not directly connected. Instead, the space between them is filled with gravel and geogrid layers to create a more even pressure distribution and reduce differential settlement. This research paper investigates the interaction between the raft, piles, and soil in saturated fat clay using 3D finite element analyses. The study explores various aspects, such as the effects of different pile configurations, raft-soil gaps, and stiffness on load distribution and settlement. The analysis also focuses on the structural response of the piles, including axial forces, bending moments, and internal stress. The study models the piles as embedded beam elements and the concrete slab as solid elements. The soil behavior is simulated using the hardening soil small-strain (HSss) model, considering the Mohr-Coulomb failure criterion, stiffness dependence, small-strain behavior, and nonlinear soil stiffness. By comparing the finite element analysis results with those from a simplified method, the study provides insights into the accuracy and effectiveness of the proposed modeling approach for evaluating the DPR system.

This paper presents a study on the impact of rigid awns and their deployment on interface friction. Awns are appendages attached to the exterior surface of a geo-system and bio inspired by grass seeds. Awns provide frictional anisotropy and assist the seed in self-embedding into the soil or clinging to animal hair. In geo-systems, like piles, deployable awns can provide frictional anisotropy reducing installation effort and increasing global capacity. In addition, flexible awns can be folded up to enable space saving for transportation. This paper presents the results from a set of interface shear tests in a modified direct shear device. Single rigid awns were tested at various angles, from horizontal, as a pseudo-static simulation of deployment, in loose and dense sand, in both the cranial (towards the head) and caudal (towards the tail direction). It is shown that awns opened at larger angles provide higher interface friction and that shearing in the cranial direction provided more resistance than in the caudal direction. This demonstrates that deployable awns could be used in geo-systems to provide friction anisotropy and increase capacity.

Lateral large-scale testing of two pile specimens with a diameter of 0.46 m and a total length of 4.57 m was performed up to complete structural failure. The piles were embedded in 1.83 m of loose backfill sand with a relative density of 20%. At the tip, the piles were embedded in 1.22 m of simulated rock. Pile yield occurred at a lateral pile head displacement of 6.35 cm. The piles reached an ultimate capacity (Pult) near 72 kN at a corresponding pile head displacement of 20 cm. Post-test soil excavation around the piles was used to document pile damage and crack propagation. The originally predicted shear failure did not occur; rather, a flexure-triggered failure through the formation of a plastic hinge above the rock-socket was observed. Upon failure investigation, soil material was placed back into the excavation, and additional soil was placed with the same technique to raise the fill height above the rock socket from 1.83 to 3.05 m. The unique test setup allowed for re-testing of the damaged pile specimen, which is rarely possible in practice. Important understanding can be drawn from the test repetition: (1) magnitude of residual capacity of the damaged pile, (2) assessment of soil resistance independent of the pile as the pile was previously loaded to failure, and (3) evaluation of potential retrofit or repair strategies upon partial or full loss of structural pile integrity. Retesting with 64% more soil compared to the original soil thickness provided an additional 80% capacity compared to Pult of the damaged specimen.

Crowchild Trail is a major arterial road in Calgary. Originally constructed in 1967, the crossing of the Bow River had become a cause of major traffic congestion. As part of the short-term improvements to Crowchild Trail, a road and bridge widening program was proposed to add additional northbound and southbound through lanes and reconfigure the on-ramps to the northbound lanes to relieve congestion. This presented three main geotechnical challenges: (1) insufficient factor-of-safety on the eastern slope of the south approach fill, (2) the potential of differential settlement between the existing abutment and the connected abutment extension, and (3) the ability to reuse the in-river bridge piers. The original construction of the south abutment used a spread footing on a large earth embankment. To facilitate the addition of one new northbound lane and two on-ramps, an abutment extension was required. This necessitated the construction of additional fill on the side of the existing steep abutment slopes. Due to the proximity of existing roads, the new lanes would need to be supported by a multi-tiered retaining wall. The foundation soils were not suitable to allow the construction of a multi-tiered MSE wall; therefore the bottom tier was designed as a spaced anchored pile wall. An additional complication was the need to connect the new abutment to the existing abutment. The abutment connection needed to be made before the addition of the majority of the load, which consisted of bridge girders and decking. This presented the potential to crack the abutment if too much differential settlement occurred. It was initially proposed to place the abutment extension on piles; however, this had the potential of creating a stiffness imbalance under cyclic traffic loading, which could also result in cracks developing along the abutment connection. Ultimately, a load transfer platform was designed to support the abutment extension and match stiffness between the new and existing fills. This paper describes the geotechnical considerations that went into developing the design and evaluating the ability to re-use the in-river bridge piers. The paper also presents the design and performance of the anchored pile wall and abutment extension, which used 2D and 3D numerical modeling techniques. The performance of the anchored pile wall has been monitored during service, which shows the seasonal effects on anchor loading that are also presented within the paper.

As part of a project to replace an existing power plant within a crude oil refinery in Grangemouth, Scotland, with a new fuel-efficient, hydrogen-ready energy plant, modules weighing over 11 MN had to be delivered and installed in an operating facility that is underlain by soft clay. For the facility to remain in operation during the construction of the new energy plant, the modules had to cross operating pipelines. A pile-supported bridge was constructed over the pipelines. However, the challenge proved to be the design of the bridge approach embankments. Existing underground utilities prevented the use of piles or ground-improvement to support these embankments. Ultimately, expanded polystyrene (EPS) geofoam blocks were adopted as the solution. Given that there was no British precedent for the use of geofoam to support such a large moving load, extensive analyses had to be performed to convince the project of the viability of this geofoam solution. Through refinement of the undrained shear strength profile of the foundation soils, application of numerical analysis to evaluate both stability and deformation of the embankment, and use of full-scale proof-testing, a geofoam embankment design was developed, constructed, tested, and successfully utilized for the installation of the 11-MN modules.

Pile capacity is an important issue in geotechnical engineering, with substantial practical implications. The practice currently relies on a few traditional mechanics-based design methods; however, with the recent advancement in machine learning (ML), many studies started investigating its diverse applications such as predicting pile capacity. The accuracy of the predicted capacity is associated with the quality of the information used as input features. This emphasizes the significance of the feature selection process. In this study, the effect of seven feature selection techniques on the performance of nine machine learning models are investigated using a dataset of 481 piles. The outcome was then compared to the performance of traditional design methods. It was concluded that using the support vector regression ML model combined with sequential feature selection technique resulted in a better performance in terms of precision and the mean absolute percentage error over available methods.

A recent study in Japan presented lateral load tests on a large pile group of seven rows by nine columns. Results from this study show current design methods overestimate the p-multipliers of trailing piles and hence, overestimate the pile capacity in the tested pile group. In this study, 3D finite element models are developed to study the interaction between two piles at various pile spacings and loading directions. Results from the finite element studies are utilized to modify the p-multiplier model. The GROUP computer program is used to calculate the pile cap load-deflection curves with the modified p-multiplier model. The resulting load-deflection curves compare well with experimental curves, especially for pile groups with a large number of piles (≥18 piles) or pile groups with more than five rows of piles.

The growing offshore wind energy sector has led to renewed interest in the design performance of pile foundations under cyclic lateral loading. Under service conditions, the lateral pile-soil reaction at small deflections and its evolution under many load cycles is of significance. Loading from occasional extreme storm events adds further complexity. In the long term, cyclic loading could cause changes in stiffness and lead to progressive tilting of the pile. To improve our understanding of the underlying mechanisms that may occur in the pile-soil system, an apparatus for model pile-segment testing has been developed at the University of Oxford. The apparatus provides a controlled environment for pile-segment loading tests and is capable of simulating field-level stresses, strains, and drainage conditions. The system can apply large numbers of cycles, at varying frequencies and amplitudes, and permits measurement of soil reaction at the soil-pile interface. The development of the new apparatus and preliminary test results are presented herein.

Excessive use of fossil fuels and environmental pollution caused by the construction industry is a global sustainability problem that needs continuous attention. It is important to consider environmental impacts of geo-structures as early as in the design phase. Life-cycle assessment (LCA) is a standardized methodology for quantifying environmental impacts of a system. In this study, LCA is used to quantify global warming impact of single drilled shafts constructed in sandy soil profiles. Parametric study is conducted to study the effects of (1) soil properties (i.e., critical-state friction angle, relative density, coefficient of earth pressure at rest), (2) design parameters (i.e., applied load and volume of steel reinforcement), and (3) site-specific parameters like the hauling distances of concrete mix, steel reinforcement cage, and construction equipment on the global warming impact of drilled shafts. The results of the LCA parametric study can help designers optimize geotechnical designs with environmental considerations.

This paper presents the results of two load tests on instrumented drilled shafts in very weak limestone performed to investigate the ultimate side friction that can be developed in this type of rock. The test site is located in NW Puerto Rico and involved a very weak porous limestone with a representative unconfined compressive strength of 2.7 MPa, and an average rock quality designation (RQD) of 34%, corresponding to a highly weathered state. The paper also evaluated the predictive performance of 10 commonly used empirical correlations of the maximum unit side friction based on the unconfined compressive strength. The range of the predictions of the ultimate side friction was found to be very large, highlighting the importance of improving our capability of predicting this parameter for design purposes. Furthermore, the measured ultimate side friction values were found to be above the upper bound of the range of predicted values using the selected static methods. The higher measured values are attributed to the medium to high surface roughness of the rock socket walls. The authors recommend the development of improved correlations that incorporate important design aspects such as rock socket roughness and stiffness ratios of the surrounding rock and drilled shaft.

A series of four-point bending tests on micropile specimens composed of a steel casing with a threaded joint and filled with grout had been performed. The objective of the testing program is to assess the predictability of failure loads and failure modes of the threaded joint when subjected to bending. This paper summarizes partial results on casings with outer diameter of 7″, 7-5/8″, and 9-5/8″ connected by threaded joints of variable length. It was found that casing diameter and threaded length play an important role on the overall connection behavior, controlling both failure mode and flexural strength. Moreover, the observed displacements suggest that the Euler-Bernoulli beam theory, often assumed on bending related problems, is not satisfied at the threaded joint, primarily due to the discontinuity introduced by the threads.

There has been a growing concern in recent years about the effects of anthropogenic noise due to pile driving on underwater wildlife. In recent years, investigators have collected several datasets of noise levels during pile driving from sites throughout the state of Florida using a unique buoy-mounted instrumentation system first presented at Geo-Congress 2020. At Geo-Congress 2023, an empirical model was presented that appears to be a solution for predicting underwater noise during pile driving where pile driving is presented as a function of drive-type (i.e., impact vs. vibrational drive) and pile-type (i.e., steel vs. concrete). This new model implied that sound attenuation is more dependent upon source level and geometric conditions than environmental energy spreading. However, the model had not yet been validated, it was unclear if the model could be applied universally or was specific to Florida geotechnical conditions, and it was unclear how universal the source-level effects were on attenuation. During this study, a blow-by-blow analysis was conducted using computational fluid mechanics and field data to help answer these questions.

The paper presents the results of a numerical investigation to study the time-dependent settlement and structural behaviour of piled raft foundations due to consolidation and subsidence. It is well understood that structures constructed on soft soils undergo large deformations with time. Moreover, few researchers have concluded that due to consolidation and pore pressure drawdown (dewatering and pumping), there is a possibility of separation of the raft from the soil, which can lead the piles to take higher loads than the design loads. A 3D numerical model of a piled raft foundation using finite element approach was incorporated in this paper. The validation was done with the reported experimental results. A parametric study was conducted to study the variation in the distribution of axial load along the pile undergoing consolidation and drawdown, the increase in settlement with time, and the load transfer from the raft to the pile head due to the process of consolidation and drawdown with variation in the pile slenderness ratio. From the finite element simulations, it was observed that soil continues to settle as a result of pore pressure drawdown. Therefore, large drag force occurs in the piles. As the soil continues to settle, the contact pressure between raft and the soil reduces, thereby changing the working conditions of the piled raft. Pile takes additional load due to the loss in shearing strength of the soil at the raft-pile interface, which may result in the failure of the piles. The maximum axial force transmitted to the piles after drawdown is 1.02 to 1.77 times higher than that transmitted during structural service load. Therefore, the need to study the effect of consolidation and drawdown on the serviceability and structural aspects of piled raft is of utmost importance.

Extreme rainfall events in November 2021 caused significant damage to several bridges on the Coquihalla Highway in British Columbia, Canada. As a result of the devastation, the Juliet Bridge needed to be replaced within a compressed timeframe in order to re-open to its full capacity. Due to the emergency nature of the work, site investigations were carried out immediately, including mud rotary boreholes and instrumented Becker penetrations Tests. Problematic low plastic glacial lacustrine soils were encountered in the foundation soils. The bridge was designed with deep piled foundations that extended into the glacial lacustrine soils. During initial pile driving for the first bridge (northbound), the shaft resistance was determined to be much less than assumed in the design. Piles were re-tested between three days and nine days after the initial drive with minimal increase in shaft resistance. Due to schedule constraints, the piles were driven to a denser glacial till layer at depth. The schedule for the second bridge (southbound) allowed for a 142-day re-strike test to be carried out on one of the piles for the new bridge. This allowed for a significantly longer setup period to be tested. This paper focuses on the key challenges faced in the design and construction of the piles and presents the pile driving data obtained during this project.

The deployment of marine hydrokinetic (MHK) energy devices and offshore wind turbines (OWT) on a shared anchoring configuration has been a topic of interest in past years. This research aims to address several challenges related to the shared anchoring system of the co-located system. The focus has been on monopiles as the foundation support system and point wave energy converters as MHK. Wind and wave loading scenarios compatible with North Carolina offshore wind lease areas have been considered. The effect of wave and wind loading on the mooring system of the floating MHK is simulated via ANSYS AQWA. Then, numerical analyses are conducted using PLAXIS 3D to simulate a shared anchoring system by considering environmental and mooring system loading on OWT. Using the critical loading scenarios, a systematic assessment of the excess capacity of OWT monopiles as a function of the embedded length over diameter (L/D) range of 5−10 is performed. It is concluded that according to the available criteria, current monopiles have extra capacity to carry additional load from MHK devices while meeting ultimate limit state (ULS) and serviceability limit state (SLS) criteria.

Recently, load transfer criteria for large-capacity helical piles have been proposed based on the 5% helix diameter (D_h) displacement criterion. However, helical piles used as foundation for solar farms in Saemangeum were typically small. The topmost helix is located at shallower depths to provide lateral resistance against wind loading, and the spacing ratio of the helices was less than 3.0D_h. To investigate the factors that affect axial capacity, the load transfer method is used to analyze the helical piles in the compression and uplift tests. It was found that the individual bearing method did not well represent the measured capacity of the helical piles based on the 0.05D_h criterion. In the calculation of the uplift capacity, the lead helix is usually assumed to provide most of the resistance; however, it was necessary to neglect the top helix in this study due to the unique configuration of the helical piles.

In this study, finite element modelling was performed in Plaxis 3D to investigate the time-dependent behavior of a single pile and pile groups embedded in a layered soil under combined lateral and compressive loads. The proposed numerical model was validated with the available literature and showed good agreement. The total pile head displacements were observed to be 7%−15% higher for piles under combined loads than the piles under pure compressive loads. Beyond a critical depth of 6−10 m, the effect of lateral loads on the pile was insignificant, irrespective of its length, spacing, and load ratio. The differential settlement ratio was observed to be maximum at 0.35−0.4 times the normalized depth, and predominant up to a spacing of 5d.

Cairo Monorail is a high-speed driverless transit system under construction in Cairo, Egypt. The typical structure consists of four span guideway beams with expansion joints at approximately 120 m in length. This structural system makes the structure sensitive to differential settlement as well as rotation angles of the pier caps. The viaduct will be founded mainly on large diameter bored piles with diameters ranging from 1.8 to 2.4 m. The general geology of the New Administrative Capital city can be described within the geology of Cairo-Suez District (Saharan deserts). Thus, long-term changing of subsurface water conditions is expected. It has also been witnessed in similar projects within the same area due to the geological nature. Therefore, California Bearing Ratio (CBR) tests were performed on silty sand samples collected from different locations along the monorail path in dry and wet conditions. Results show that there is a loss in the soil stiffness ranging from 15% to 40% due to wetting. To consider the wetting effect on piles, pile load tests were conducted in dry and wet conditions. Pile load test results show that pile capacity reduction due to wetting was about 20%.

Lightweight foam glass aggregate (FGA) has a low unit weight and high friction angle that, when used as a backfill behind walls and abutments, is expected to produce significantly lower earth pressures compared to conventional cohesionless fill soils. This paper describes a study performed to evaluate the minimum extent of FGA backfill required behind a wall or abutment to keep the theoretical earth pressure within 15% of the theoretical earth pressure from FGA fill alone. Three analysis methods were applied to determine the minimum dimensions of a trapezoidal prism of FGA backfill located between the wall and conventional soil or aggregate: (1) Coulomb planar surface analysis, (2) noncircular limit equilibrium (LE) analysis using Spencer’s method, and (3) plane strain strength reduction factor (SRF) analysis. These methods were applied in a parametric study that considered a range of values for the unit weight and friction angle for a cohesionless conventional backfill soil and FGA backfill located between the wall and the in situ conventional soil. Based on the analysis results, an empirical equation was developed for the maximum cut slope angle as a function of material properties and wall height.

Current AASHTO load and resistance factor design (LRFD) specifications in the US have adopted the stiffness method for the internal stability design of geosynthetic mechanically stabilized earth (MSE) walls. The same method will appear in a future edition of the corresponding CSA LRFD code in Canada. During the development of the stiffness method, the problem of a bridge abutment footing seated on the top of a MSE wall structure was not investigated. This paper presents an initial attempt for a method to include the additional footing loads in the calculation of reinforcement loads and strains using the AASHTO stiffness method and correction factors applied to the 1H:2V footing load dispersion method. The example uses the soil properties from the well-documented SR-18 walls in the US and considers walls of three different heights, four different reinforcement stiffness values, and three different reinforcement spacings.

This paper summarizes the inverse analyses that explicitly considered hydrodynamic time effects and the results of such optimizations for the supported excavation for the Simpson Querry Biomedical Research Center (SQBRC) in Chicago. To evaluate hydrodynamic effects arising from pore pressure dissipation, consolidation stages were included in finite element simulations of the excavation to explicitly model the time effects. Both stiffness parameters and hydraulic conductivity were computed based on optimization using the lateral wall movements during excavation. The resulting optimized parameters considering the time effect are compared with those using only undrained construction stages to assess the changes in parameter values that arise from explicitly considering the time effects. Conclusions are drawn with respect to the applicability of the conventional undrained assumption for excavations through clay.

This paper presents a case history on the design, construction, and performance of soil nail walls for the addition of express lanes along the I-405 freeway corridor in southern California. The walls were constructed to support vertical cuts into existing “old” embankment fills. Federal Highway Administration GEC 007 recommends ultimate bond strength values of 21−41 kPa for augured nails in silty sand fill. Ultimate bond strength values of 93−145 kPa were selected for design based on local experience, literature review, and engineering judgement. Back-calculated ultimate bond strength values from field pull-out tests were about 30 to greater than 145 kPa. The objective of the paper is to validate the original design with actual pull-out tests. It is concluded that the GEC 007 ultimate bond strength values for fill may be conservative, and higher values can be justified in old embankment highway fills in southern California.

This paper presents the design and construction of a new pump station and sanitary sewer overflow tank in Maple Ridge, British Columbia. The site is underlain by soft to firm silts and clays, overlying compact to very dense sand, which, in turn, are overlying stiff to very stiff marine clay. The natural groundwater level was as shallow as about 0.5 m depth below the ground surface. Excavation between 4.5 and 11 m depth below the ground surface was required. Vertical shoring walls were necessary to limit the excavation footprint and to cut off groundwater. Sheet pile walls with internal bracing were assessed as the most practical and economical temporary shoring option. The contractor’s alternative design, which was constructed, included sheet pile walls with external bracing using soil anchors. This paper presents the geotechnical design of temporary shoring, the construction details, and settlement monitoring adjacent to the excavation. The measured settlement data are compared against settlement magnitudes and patterns estimated using published correlations. In general, the measured settlement adjacent to the excavation was two to three times the settlement predictions using published correlations. The discrepancy between the predicted and measured settlements is believed to be the result of a combination of construction method and staging, including the time delay between the excavation and soil anchor installation, insufficient soil anchor capacity, the time delay in the installation of new anchors, and lateral deflection of cantilevered sheet piles.

Geosynthetic-reinforced retaining (GRR) walls have been widely used in the construction of bridge abutments and approach slabs. Traditionally, deep foundations carry the bridge loads, rather than the abutment walls. Bridge loads can be supported directly on footings constructed on the reinforced soil mass. Settlement of footings on top of GRR abutment walls is a key parameter that should be considered for design of these structures. The objective of this study was to develop a method to estimate the vertical settlement of strip footings on top of GRR abutment walls. To achieve the above objective, the effect of strip footing loading on stability of several GRR abutment walls with flexible facing was investigated. Limit equilibrium analyses were conducted to determine the factors of safety of these abutments. The calculated and measured settlement results indicate that the GRR abutment walls with lower factors of safety had larger vertical settlement of footings than those with higher factors of safety. Based on this finding, a data analysis was conducted to establish an exponential relationship between the calculated factor of safety and the maximum vertical settlement of footings on top of GRR abutment walls.

Soft soil, due to its high compressibility and low strength, poses several threats to the overlying structures. The design of infrastructure founded inevitably on such incompetent foundation soils demands due diligence to accommodate for unintended adverse effects raised by excessive subsidence in such soils. Geosynthetic-reinforced soil (GRS) abutments are among the important structures constructed for different important purposes. Slope stabilization and bridge abutments are two examples of such applications. These composite structures contain different load-carrying elements, which come into play when superimposed by dead and live loads. Geosynthetics reinforcing elements, facing panels, backfill quality, underlying foundation competence, and the toe reaction condition are among the most important factors impacting the overall behavior of such structures. The existing literature in geotechnical engineering mainly focuses on the normal practice of using such walls where the foundation soil is assumed competent enough before embarking on the construction of the overlying GRS abutment. However, there are situations in civil engineering that are imposed by the harsh local and environmental conditions, and proper maintenance of the erected structure is necessary to minimize potential issues. Soft soils, found abundantly in Canada, provoke the GRS abutment system detrimentally by mobilizing extra load values at the reinforcement-facing connection points and along the length of different layers. This article explores the resilience of the GRS abutment system in handling the adverse effects of underlying soft soil. It delves into the assessment of different retrofitting measures and demonstrates their respective effectiveness.

The present experimental study evaluates the liquefaction resistance of pond ash reinforced with different types of fibers (polypropylene, jute cellulose, and coir fibers). A series of stress-controlled cyclic simple shear (CSS) tests was performed to study the liquefaction resistance of fiber-reinforced pond ash. Cyclic loading was applied at 1 Hz frequency under 100 kPa of overburden pressure. All specimens were prepared at 95% of maximum dry density (MDD) and optimum moisture content (OMC) of pond ash. For each fiber type, a minimum of three CSS tests were conducted at different cyclic stress ratios (CSR) to obtain the cyclic resistance ratio (CRR). CRR was determined as the CSR required to liquefy the specimen in 15 loading cycles. This experimental research reported that including jute cellulose fibers in pond ash increased the liquefaction resistance of pond ash. The inclusion of other fibers showed a minimal increase in liquefaction resistance.

Geosynthetic clay liners (GCLs) comprising sodium bentonite (NaB) have been shown to exhibit significant membrane behavior, whereby contaminants are restricted from passing through the clay due to electrostatic repulsion, potentially enhancing long-term containment performance. Recent membrane behavior research has focused on evaluating membrane efficiency (ω) of enhanced bentonites (e.g., bentonite polymer composites) and testing under more complex conditions (e.g., elevated temperatures and unsaturated conditions), often drawing comparisons with earlier published data on NaB-GCLs. However, the validity of comparing NaB-GCL results across studies and from earlier work remains unclear. No prior study has evaluated the variability in reported ω values for the same type of NaB-GCL manufactured in different years. In this study, multistage membrane behavior tests were performed on specimens of the same NaB-GCL type manufactured at different dates. One specimen (GCL1) was taken from a NaB-GCL roll manufactured in the last five years. Another specimen (GCL2) was taken from a NaB-GCL roll that had been stored in a laboratory since 2005. Values of ω were measured for both specimens and also compared to results reported in 2002 for the same type of NaB-GCL (GCL3). To support interpretation of the results, cation exchange capacity, mass per unit area, and swell index tests also were performed. For the same salt solution concentrations, GCL2 exhibited significantly lower ω than GCL1 and GCL3. Lower values of ω corresponded to lower bentonite mass per unit area. The results have important implications regarding limitations in comparing results of current membrane behavior research on enhanced materials and testing conditions to commonly cited literature data.

The characterization of long-term tensile stiffness of geogrids is required for various geotechnical applications, such as the design of geosynthetic-reinforced load transfer platforms within column-supported embankments. While ultimate creep strain information is generally included in manufacturer-provided product data sheets, current recommendations for the estimation of the long-term tensile stiffness of geogrids at the serviceability limit of strain require interpretation of isochronous curves not readily available to practitioners. Therefore, reduction factors generated for ultimate creep strain failure are often used for serviceability loading conditions, a method which has not been validated. This paper presents a method for appropriately scaling available ultimate creep strain information to obtain reduction factors for the serviceability limit creep strain. For a variety of geogrids, reduction factors estimated using the proposed method and direct use of the reduction factor at rupture are compared to reduction factors generated with interpreted values from isochronous curves. Results of the proposed scaling framework show good agreement with factors from the interpretation of isochronous curves; additionally, results suggest that the direct use of the reduction factor at rupture may be acceptable.

This paper describes the first known static plate load (SPL) testing program performed on compacted foamed glass aggregate (FGA) fill produced in the US. The testing programs consisted of 14 SPL tests performed on approximately 610 mm (24 in.) thick compacted test fills prepared under controlled conditions. A portion of the tests included either a 1.9 kPa (40 psf) or 7.2 kPa (150 psf) pressure applied to the top of the fill to simulate the confinement provided by soil or crushed stone typically placed between the FGA and the pavement or floor slab. The results show that the modulus of FGA is increased by the influence of the surface pressure and that the diameter of the plate can have an effect on the results. This paper provides the details of the equipment, procedures applied to SPL testing of FGA, and interpretation of the results.

This research investigates the behavior of geocell-reinforced capping layers in ballasted tracks subjected to cyclic loading. Large-scale laboratory testing and numerical modeling techniques were employed. The cyclic tests applied a 25-t axle load under frequencies ranging from 10 to 30 Hz. The geocell-reinforced capping layers were modeled using the discrete element method. The geocell structure was simulated by bonding small balls to build a realistic geometry and shape. Model parameters were calibrated based on tensile and bending tests performed on the geocell material. The DEM simulations accurately represented the irregular shape of the capping aggregates using bonded circular particles. The findings indicated that the geocell effectively reduced both vertical and lateral displacements of the capping layer. The DEM analysis provided valuable insights into the contact force chain distributions within the capping assembly.

As alternative backfills become more commonplace in the construction of mechanically stabilized earth (MSE) walls, the selection of reduction factors in calculating the long-term design strength (LTDS) should be evaluated. Alternative backfill materials have the potential to approach the FHWA default reduction factors for durability. The FHWA default reduction factors for durability are based on the minimum molecular weight (Mw) and maximum carboxyl end group (CEG) concentration of the high tenacity polyester fibers used in geogrid production in conjunction with limiting the pH range of the soil backfill for polyester (PET) geogrids. However, these default reduction factors are conservative, and the default pH range is based on limited testing of a particular coated PET geogrid in environments with a pH over 9. This paper outlines the importance of product-specific testing in elevated pH backfill environments. A detailed multi-year investigation of the reduction in tensile strength of two different manufacturers’ coated PET geogrid exposed to pH values ranging between 10 and 11.4 was performed. The comparison between the two different coated PET geogrids shows the reduction in tensile strength due to submersion in elevated pH aqueous solutions can vary, even with similar Mw and CEG products. This paper illustrates the importance of product-specific testing and the conservatism built into the FHWA default durability reduction factors.

The current study utilizes 3D printing technology to manufacture geocells on the lab scale. 3D printed polypropylene sheets were ultrasonically welded to make geocells of different pocket sizes. Geocells of three different pocket sizes were manufactured, keeping the height of geocells constant, with the aspect ratio varying between 0.6 and 1.2. These geocells were characterized for their tensile strength through wide width tension tests and peel strength tests. Modified direct shear tests were carried out on 3D printed geocells to obtain and compare the interface shear response. Results showed that including geocells improved the interface adhesion of sands, thus enhancing their interface efficiency. The sand-geocell interface with a 100 mm pocket size of cells showed an interface efficiency of 1.16. The effect of aspect ratio on the interface shear response of 3D printed geocells was also studied and found that geocells with an aspect ratio close to unity showed the highest interface strength.

This paper presents a comparison of conventional approaches along with a composite material approach to estimate vertical deformation of a multi-axial stabilized working platform for a ringer crane. The major challenge in design was to satisfy very stringent criteria for allowable total and differential settlement. Using a conventional analysis approach, the geogrid stabilized ringer crane platform was designed and constructed in 2017. The platform was 1.8 m thick and composed of five layers of multi-axial geogrid with triangular apertures and crushed angular graded aggregate material. Back-analyses were carried out using actual deformation measurements from the ringer crane platform during operation and an adaptation of a composite material model that considers aggregate and geogrid interactions, which enhance the composite material strength for a mechanically stabilized soil platform. Although post-analyses were not done using finite element methods, an adaptation of the composite material model was incorporated into a conventional approach and provided results closer to measurements. Comparison of settlement analyses post-operation to those of initial design provides important insight into material numerical modeling techniques.

The expansive behavior of clayey soil in response to climate-induced changes in the soil water characteristic curve (SWCC) is a significant issue for many types of earth infrastructure. The application of geosynthetic material has been common to reduce the climate-induced changes in SWCC. Engineered turf, which is a composite geosynthetic material, has gained popularity for different earth systems to increase overall infrastructure resiliency. This paper’s objective was to investigate engineered turf’s effect on the climate-induced changes in SWCC at shallow depths in the field conditions using the statistical non-parametric measure: Spearman rank correlation coefficient (ρ_s). This research hypothesized that since the changes in soil moisture and suction would relatively be simultaneous for exposed ground under variable climate, thereby exhibiting a reasonable negative correlation between water content and suction, whereas the degree of simultaneity in the changes between water content and suction of the soil under the engineered turf would display arbitrary correlation. To test the hypothesis, two test beds, (1) a compacted clay bed (CCB) and (2) a compacted clay bed overlain by engineered turf (ETB), were constructed with expansive soil and instrumented with collocated moisture sensors and tensiometers identically to collect concurrent water content and suction data continuously. The analysis revealed that the estimated ρ_s values for CCB were almost −1.0 during different drying conditions indicating a very strong correlation. On the contrary, the estimated ρ_s values for ETB were +0.79 to −0.32 indicating an irrational to a weak correlation between ρ_s. The results indicated the engineered turf to be an effective barrier to climate-induced changes in SWCC.

This study presents a numerical investigation of the hydraulic (drainage) and mechanical (reinforcement) characteristics of nonwoven geotextile and its efficacy in the application of geosynthetic-reinforced soil wall comprising fine-grained marginal soil as backfill material. The numerical simulations of rainwater infiltration in unsaturated soil revealed a capillary barrier phenomenon of geotextile, which prevents the progression of saturated wetting front in the marginal soil backfill through the geotextile layers until the attainment of breakthrough suction at each soil-geotextile interface. Each geotextile layer successively functions as a capillary barrier and a drainage medium as the saturated front progresses within the backfill, thus aiding the backfill to preserve greater suction for a longer duration of rainwater infiltration. Local instability in the soil wall is observed due to the accumulation of pore water above the geotextile layers undergoing the capillary barrier action. Sand cushions sandwiching the geotextile layers substantially improved the local and global stability by minimizing the capillary barrier action.

The bearing capacity and failure mechanism of encased stone columns are affected by many factors, such as relative density and shear strength parameters of aggregates, tensile strength, stiffness of geosynthetics, encasement length, diameter of the column. Many studies have investigated the behavior of encased stone columns using triaxial testing, small-scale laboratory tests, full-scale field tests, and numerical analysis. Most of the studies used geotextile as encasement, and very limited studies are done on geogrid. The present study focused on the performance of geogrid as encasing material. Using a large-scale triaxial apparatus with dimensions of 150 mm in diameter and 300 mm in height, 10 encased stone column (ESC) aggregate element tests and four ordinary stone column aggregate element tests (OSC) were conducted under confining pressures of 0, 50, 100, 150, and 200 kPa. The results concluded that the inclusion of geogrid encasement could contribute to the significant improvement in load-carrying capacity and apparent cohesion or confining pressure of OSC. The results showed that the load-carrying capacity of ESC is less dependent on surrounding confinement at lower confining pressures and more on the tensile strength of the geogrid.

With the evolution of railway embankment design, the focus has been on the improvement of the strength and optimizing the thickness of the track bed to restrict stresses transferred to the subgrade from a train-induced dynamic load, within acceptable limits. Conventional practice is to design an appropriate thickness with the right choice of ballast and sub-ballast thickness with subgrade preparation. However, addressing issues with long-term operations is another facet that is yet to be standardized. With the advent of geocell, the spectrum of cost-effective long-term solutions has widened. This paper introduces a strategic design methodology for rehabilitation design to address operational issues, arising from loss of confinement, settlement, and embankment degradation, using high strength, modulus, and creep resistant polymeric alloy geocell. This methodology is supported by design execution and year-round observation indicating that this will be a promising cost-effective and environmentally friendly solution for the future.

A direct cone penetration test (CPT) method for assessing the load-displacement-capacity response of shallow foundations on sands that was previously developed at Georgia Tech (GT) is applied to four new case studies that have become available. The algorithm relates the ratio of applied foundation stress normalized to cone tip resistance to the square root of normalized displacement, defined as settlement divided by footing width. The case studies include footing load tests conducted on large shallow concrete foundations and steel plates situated on both natural sands and sands densified by ground modification and compaction methods. The case studies involve circular and square foundations from projects located in Oman, Kuwait, UAE, and Texas, with reasonably good predictive agreement shown for all cases. A means to extend the approach to include footings bearing on silty sands to sandy silts for drained behavior is also discussed.

The kinematic approach of limit analysis is employed to account for inertial effects caused by seismic loads. In addition to a footing’s capacity to carry vertical and horizontal loads, the method is adapted for calculating the resistance to possible moment loads that may be transferred to the foundation soil from the superstructure. A foundation soil collapse mechanism that is sensitive to moment loads is utilized in the analysis. This mechanism transitions to a more common translational collapse in the absence of moment loads. The Mohr-Coulomb strength criterion was used in the analysis. While the limit load on the strip footing is grouped into components dependent on the cohesion, overburden, and the weight of the foundation soil (in addition to the internal friction angle), this study was focused on the component dependent on the foundation soil weight. The outcome of calculations demonstrated that the mechanism used in the analysis yields more conservative results than other methods. Considering that the analysis employed yields the upper bound to the limit load, the results appear to be more accurate than those considered for comparison.

Study of ground-borne vibration has been developed since 1960, but little is known if wave propagation in the ground may directly affect buried utilities. In this research, the vibration from the installation of rockfill columns in a riverbank is investigated numerically. Finite element models were calibrated with field data from a site with a buried aqueduct nearby. The calibrated models were used to evaluate the vibratory response of the ground in terms of particle velocities and frequency spectra on the aqueduct walls. Results show that the softer soil layers keep the waves longer and particle velocities are higher in the long term. Particle velocities tend to increase around the aqueduct. The soil-structure interaction changes the shape of the wave and lowers its frequency to a range within the natural frequency of the soil. This change in frequency could impose movements on the aqueduct if it is allowed to move freely with the soil.

The fabrication of buried pipelines has been a significant part of human history because of its social and strategic importance. This paper presents a numerical study of the response of buried pipelines in soil to surface blast loading for varying cross-sectional shapes of pipelines. A 3D finite element model (FEM) is developed to simulate the dynamic response of the buried pipeline for blast loading to replicate the actual site conditions. The Coupled-Eulerian-Lagrangian (CEL) technique is adopted for the numerical modeling of this problem. Finally, a relative comparison between the blast responses of varying cross-sectional shapes is performed, and a corresponding discussion is made. The results showed that the varying cross-sectional shape significantly affected the pipeline’s dynamic response. The findings of this study provide valuable insight into the response of buried pipelines of different cross-sectional forms beneath soil for surface blast loading.

Conventional mechanical drilling methods have problems such as high levels of vibration, noise, and dust. To address these issues, rock drilling using abrasive waterjets has become widespread in geotechnical engineering. Waterjet drilling reduces vibration, noise, and stress during the non-contact drilling process and is safe for use in hazardous or populated areas. However, the cost of abrasive is a problem that needs to be addressed. This study suggests recycling the garnet used in waterjet drilling to improve cost efficiency. The drilling performance was analyzed for different waterjet energy parameters (water pressure and standoff distance; distance between the nozzle tip and surface of the specimen). The garnet abrasive was collected after each test case and tested for reusability by particle size standard: particles larger than 150 μm, using sieve tests analysis. The results revealed that the drilling performance improved with a higher water pressure and flow rate, whereas the reusability decreased by 26%. Operating costs were reduced by 34% when the abrasive was recycled. These findings are expected to be useful for optimizing the total cost and duration of rock drilling using abrasive waterjets.

Over the past two decades, the author along with his research group and industrial partner began to develop a novel Rayleigh-based distributed optical strain sensing technology capable of monitoring strain along a ground support scheme at sub-centimeter spatial resolution. This came out of the requirement to monitor and understand the geomechanics of composite ground support systems at the micro-scale. This line of research has been extended to monitor fully grouted rock bolts (FGRB) and rock mass deformation with an increased resolution of sub-millimeter spacing (0.65 mm). This distributed optical sensing technique has been tested in the laboratory and been implemented at multiple sites around the world. Within this context, this paper summarizes key lessons learned (i.e., spatial resolution, QA & QC, and industrial applications) over the past two decades regarding the use of fiber optics for monitoring the ground conditions and support elements in underground excavations. Through such monitoring, the goal is to improve current design and guidelines.

Errata identify incorrectly published items and technical errors. This file contains errata that shall be used from their effective dates; the most recent effective date is 02 24 2024.

Errata identify incorrectly published items and technical errors. This file contains errata that shall be used from their effective dates; the most recent effective date is 02 24 2024.

Errata identify incorrectly published items and technical errors. This file contains errata that shall be used from their effective dates; the most recent effective date is 04 19 2024.