Geo-Congress 2022: Deep Foundations, Earth Retention, and Underground Construction

Piles in weak soils are sensitive to adjacent excavation and construction loads. Excessive lateral displacement of piles during construction adds shear and bending moment to the piles. This is of great concern because these additional lateral loads are sometimes not accounted for when designing piles, and the horizontal bearing capacity as well as shear strength of the piles is low. This paper presents a case study of a recent project with severe pile damage and large soil displacement during construction due to the installation of adjacent piles and excavating foundation pits. With improper construction procedures, the piles were severely damaged and have almost no bearing capacity before the construction of the superstructure even begins. On-site measured data were used as inputs for numerical simulations, and analyses were carried out to study the behaviors of pile groups affected by construction activities. The simulation was for each construction activity/stage as well as the whole construction process to determine the effects of excavation and deep foundation construction equipment loadings used for pile installation. The effects on the lateral displacement of piles and surrounding soil were quantified. The results from finite element analysis on the lateral deformation of piles are in good agreement with the measured data, with differences of about 5%–10%. Several construction procedures are recommended, with recommendations made to mitigate pile damages and large soil displacement due to construction. With the recommended procedures, the lateral deformation of pile groups can be greatly reduced by at least 90% and pile damages can be eliminated.

There has been a growing concern in recent years about the effects of anthropogenic noise due to pile driving on underwater wildlife. Current guidelines for mitigating hydroacoustic effects associated with these geotechnical events are based upon a relatively simple transmission loss formulation known as the Practical Spreading Loss Model (PSLM). This model is easy to implement, but it may produce overly conservative results. During a previous study, sound data during pile drives from two sites in Florida showed much higher sound attenuation than predicted by the PSLM at one of the sites. This study focused on explaining this discrepancy using computational fluid dynamics. Specifically, synthetic pile drives were simulated using Siemens’ Star-CCM+. These models tracked sound decay from a single hammer blow that was imposed at a modeled pile using site-specific bathymetry data. Results showed that discrepancies between measured transmission loss coefficients and the practical spreading loss model could not be explained due to local bathymetry alone. However, if different sound absorption criteria were used at the sites’ mudlines, the model was able to replicate results. The data therefore suggest that geotechnical conditions may play a significant role in determining anthropogenic sound loss due to pile driving.

Auger cast-in-place (ACIP) piles that support highrise buildings are the only structural elements of those structures that are not reinforced considering the load combinations defined in the Building Code or ASCE 7. Current design practice only considers the first mode forces that are determined by the loads transferred by the pile cap to the piles. Three-dimensional numerical models of highrise foundations indicate a second mode of forces defined by moments and shear forces’ actions caused by the soil’s lateral displacements under the building loads. These second mode forces have been confirmed by instrumentation installed along production piles that have been monitored during the building’s construction. The second mode forces act at depths that are not conventionally reinforced, leading to potential pile structural failure. The impact of considering or not the second mode forces in pile design is illustrated with the study of the performance of two highrise buildings of known settlement behavior. The effect of the second mode forces, combined with the local limestone’s sensitive behavior, is included in the foundation analysis of the two highrise buildings. The structures have similar heights and similar geotechnical conditions. The numerical modeling predicts deformations under the two buildings and the forces acting on the piles. The forces acting on the piles are compared with the pile’s structural capacity. The limestone strength and deformation modulus, based on mobilized deformation, is considered. The analysis results indicate that the pile loads, combined with the limestone-sensitive behavior, generate settlements that correlate with the length of reinforcement used to construct the piles. A three-dimensional numerical method to define the second mode forces acting on the piles is recommended to analyze highrise foundations under the required load combinations. An example of the results of such an analysis performed for a 47-story building foundation is presented here. A reduction factor for the calculated forces is recommended based on stratigraphic characteristics and numerical modeling assumptions. The resulting forces are used to define the pile’s shear and longitudinal reinforcement.

Majority of Thai oil and gas facilities are located along the eastern coastal provinces. In a recent oil and gas project, over 10,000 precast piles were installed by conventional driving and jack-in methods. The jack-in piling method was adopted in areas where new facilities were built near existing plants with stringent noise and vibration criteria. Extensive soil investigation program was executed, providing the necessary geotechnical data for subsoil characterization and foundation design. Preliminary pile load tests were conducted to establish the pile driving criteria for driven piles and the termination criteria for jack-in piles. For driven piles, soil-setup was investigated showing a linear relationship with pile length. As for jack-in piles, the termination criteria were determined by considering appropriate jacking force with sufficient holding time, meeting acceptable efficiency and performance. From the load test results, shorter jack-in piles had achieved required ultimate capacities with greater skin friction during installation.

This study investigated the effect of applying a one heating–cooling cycle on the interface strength parameters of saturated clay soil-concrete, and the potential use of heating to improve the pile capacity. A large direct shear test device with inner dimensions of 300 mm, 300 mm, and 200 mm for width, length, and height, respectively, was modified to perform the soil-concrete interface tests. The experimental tests were conducted on low-plasticity soil with PI = 12, medium-plasticity soil with PI = 29, and high-plasticity soil with PI = 60. Two specimens at different testing temperatures (~20°C and 70°C) were tested for each normal stress and clay type under different normal stresses ranging from 30 to 150 kPa. The specimens’ temperature was increased gradually during the heating process from room temperature (~20°C) to ~70°C. The specimens were then cooled back to room temperature. The test results showed significant increase in both the peak and at large displacement interface friction angles by a percent ranges from 13.6% to 35.5%.

Development of offshore wind in deeper waters that are beyond the feasible range of fixed towers requires a cost-effective anchorage for floating offshore wind turbines (FOWTs). The multiline ring anchor (MRA) system has been devised as a cost-effective anchor for FOWTs, by virtue of its high efficiency, its capability for securing multiple mooring lines, and its adaptability to a wide range of seabed conditions. Taut mooring systems become increasingly attractive as water depth increases, leading to combined horizontal–vertical loading on the anchor. Since the MRA is deeply embedded in the seabed, it has capabilities for resisting the vertical component of the forces imposed by taut moorings. Since the MRA lacks the reverse end bearing resistance of a conventional suction anchor, its design requires careful attention to ensure that it can resist the vertical load demand from a taut mooring. Previous preliminary studies show that the uplift resistance of the MRA in soft clay can be improved by attaching wing plates, increasing anchor size, or installing in deeper depth, the latter being limited by the constraints of suction installation. Wing plates turn out to be a very promising option, but more reliable studies of their performance are needed to support an optimally designed, cost-effective anchor. Thus, rigorous three-dimensional finite element analyses were conducted to understand how wing plates improve the uplift resistance and provide reliable evaluations of the vertical load capacity of the MRA. The results show that the soil-anchor adhesion factor, the total number of wing plates, and the width of wing plates are important factors contributing to anchor uplift resistance. To investigate an optimal design of the wing plates, a comparative study was carried out to compare the effects of wing plates on minimizing capital costs. The studies show that attaching wing plates can be an economical solution for improving axial capacity.

The influence of loading rate on the ultimate axial uplift capacity of single-helix round-shaft helical anchors in clay was investigated. Field load tests were performed on a set of 5 full-scale helical anchors using different loading rates in both stiff and soft clay. An initial set of tests was performed 10 days after installation in the stiff clay. The anchors were then left to age and a second set of tests was performed 110 days after installation. The anchors were then advanced into the lower soft clay. Tests were performed 10 days after installation and then left to age for 280 days. Axial uplift load tests were performed using loading rates that varied over three orders of magnitude, in terms of load/min. The results showed that in the soft clay the ultimate capacity of single-plate helical anchors was related to the loading rate in much the same way as previously published results for load rate effects on driven piles and undrained shear strength in clays, first showing a decrease in capacity with slower loading rates and then an increase in capacity as the time to failure increased. In the stiff clay, the results showed a decrease in load capacity as time to failure increased. Aging had little effect on the capacity of the anchors in the soft clay but showed a substantial increase in capacity in the stiff clay, especially at slow loading rates. The results demonstrate that the loading rate can have an influence on ultimate uplift capacity of helical anchors in some soils and may be an especially important consideration when comparing field load test results with predictions of anchor behavior and may also influence empirical torque-to-capacity correlations.

In general, the pile tip shear, pile tip moment, and pile shaft resistance are neglected when predicting the resistance of piles subjected to monotonic lateral force. However, these resistance may contribute significantly to a short or rigid pile. In this case, when the lateral force is applied at the top of the pile, the entire body rotates without much bending deflection. The contributions of shear and moment resistance at the bottom of the pile as well as the side friction-induced resistance can be significant. A total of twenty lateral load tests with aspect ratio (pile length over pile diameter, L/B) smaller than ten in sand and clay was collected for this study. The tip shear resistance and the vertical side friction at the perimeter of the pile primarily engage in the mobilization of contact friction between piles and soils, which can be modelled with similar bilinear curves. The first part of the bilinear curve represents the soil resistance as a linear function of pile displacement, while the second part of the curve is a constant, which reaches the fully-mobilized maximum side friction. The tip moment resistance curve is a non-linear function of rotation with a limit of 0.5 radians in this study. With these additional resistance, the load-deflection curves are predicted using Matlock (1970) and Reese et al. (1975) in clay and Reese et al. (1974) which evolved into the American Petroleum Institute (API) (2010) Recommended Practice in sand, and the program LPILE is used. The predicted loads considering additional resistance (tip shear, tip moment, and side friction) are compared to the ones without additional resistance at selected deflection levels. Finally, one can conclude that the contributions of the tip shear and tip moment resistance are significant on piles with aspect ratios smaller than or equal to 5.0 (L/B ≤ 5.0). With additional resistance including tip shear and tip moment, the predictions seem to be closer to the measurements in Richmond Test. However, no relevance was found between the ratio of predicted load considering moment resistance from side friction over predicted load and aspect ratio.

This paper presents new static analysis (SA) methods to predict both unit shaft resistance (q_s) and unit end bearing (q_b) of driven piles in siltstone using 15 pile load test results from 11 bridge projects in Idaho, Montana, and Wyoming. The driven piles included steel H-piles and open-ended pipe piles. The q_s values determined from dynamic load tests were related to the unconfined compressive stress (q_u), corrected standard penetration test (SPT) N-values (N₁)₆₀, and calibrated adhesion factor of siltstone. A combined term involving pile penetration and SPT N-values was considered in the prediction of q_b. The proposed SA methods were validated using a separate pile dataset consistent of nine observations from static and dynamic load tests in Wyoming. The proposed SA method in terms of q_u (M−1) is recommended for q_s prediction because it has the mean relative bias close to one and the lowest coefficient of variation. The validation study reveals that the q_s values were 179% and 385%, on average, overpredicted by the existing α-method λ-method. The proposed SA method provided good estimation of q_b; however, the uncertainty in q_b prediction was relatively high. Higher Load and Resistance Factor Design (LRFD) resistance and efficiency factors were determined for the proposed SA methods using three probability-based methods.

The transitional behaviors of intermediate geomaterials (IGMs) between soil and hard rock create relatively high uncertainties in designing and constructing driven piles in IGMs. The performance and acceptance of driven piles in IGMs are often evaluated using dynamic load test (DLT) methods. Although dynamic methods offer considerable technical and economic benefits, a static pile load test (SLT) is warranted to understand the static behavior of driven piles in IGMs. This paper presents three static and dynamic load tests of steel H-piles in IGMs. Test pile 1 (TP1) was driven into siltstone in Wyoming, and the other two test piles (TP2 and TP3) were driven into shale in Iowa. Vibrating wire strain gauges were installed along each pile length to determine the shaft resistances and end bearing. Each pile was axially loaded, and pile top displacements were measured. Two different pile load systems (independent and integrated) were designed for the SLT. Four failure criteria were used to determine the ultimate pile resistances. It was observed that DLT underpredicted the total pile resistances of TP1 by about 33%, 20%, and 24% when compared to SLT results based on Davisson, 10% B, and De Beers YL criteria, respectively. Compared with the De Beers’ YL criterion for TP2, DLT underpredicted the total pile resistance by 11% and shaft resistance by 60%, but overestimated the end bearing by 15%. For TP3, DLT underestimated total pile resistance by 9%, 11%, and 11% compared with the three criteria, respectively.

The geotechnical conditions in Central Florida consist predominantly of medium-dense silty-sands that allow the installation of deep foundations using pile driving methods to carry large infrastructure loadings to the limestone bedrock. This process generates vibrations in the surrounding soils that could trigger ground deformations and possible damage to nearby structures. The process is a complex dynamic soil-pile interaction problem that involves many variables including the transmitted energy of the driving hammer, the length, material, and cross-sectional properties of the pile, and the dynamic properties of the surrounding and supporting soils, among others. This paper presents a numerical model developed in PLAXIS 2D for the continuous driving process of a prestressed concrete pile. The constitutive soil model used was the critical-state based hypoplasticity model for sands enhanced with the intergranular strain concept. A parametric study was performed to elucidate the effect of the above-mentioned variables on the vibrations and ground deformations in the vicinity of the pile. A wide range of relative void ratios varying from 25% to 70% is presented. The effects of type of hammer on soil response at the ground surface are also examined. Conclusions are drawn regarding the geomechanical characteristics of these soils to trigger large pile-driving induced ground deformations.

Inspired by the natural anchoring mechanism of the tree root system, the authors have investigated the possibility of developing a new foundation configuration in this paper. For this study, a foundation to support an onshore wind turbine, a structure that resembles a coconut or palmyra tree, was considered. First, simplified configurations of the new foundation were developed with a different number of primary piles (primary root) and secondary piles (secondary root) resembling a tree root system. These configurations consisted of six, twelve, and eighteen primary piles arranged and inclined symmetrically about the vertical axis. Each configuration consisted of a vertical pile at the center of the root bulb/ball representing the taproot. Two secondary piles were added to each primary pile in each configuration resulting in a total of six configurations. Additional three configurations were developed by increasing the length of the primary pile. Each model was analyzed in ABAQUS by applying three different vertical compression and tension loads. The results showed a reduction in vertical settlement of the foundation when the number of the primary pile and secondary pile was increased. However, the contribution of the second pile was not as significant as primary pile. The highest and the lowest Mises stresses were observed at the top and bottom of the pile, respectively. The configurations with increased length of primary pile resulted in a lower vertical settlement for the same load. However, for the configuration with six primary piles, the longer length did not affect the vertical settlement when subjected to higher two loads. Further investigation is necessary to understand the effective and efficient configurations thoroughly.

Development of brownfields, including former landfills, into commercial, industrial, or mixed-use properties is increasing. However, design and construction of such redevelopment projects is complex and must consider a variety of geotechnical, structural, environmental, and regulatory hurdles. This paper describes the design and construction challenges of installing deep foundations through refuse and soft clay strata to a variable bearing layer. Design challenges included developing axial capacity in soft soils and providing sufficient lateral capacity for seismic loading after settlement of soils around the piles. Construction challenges included early refusal; low blow counts at the target tip elevation; repeated misfires when driving through the soft Young Bay Mud; maintaining pile verticality through the MSW; and broken or unusable piles.

The Top-Loaded Bi-Directional Test (“TLBT”) is a new method to apply bi-directional loads to a deep foundation element with the loading source located above the foundation head. In the TLBT reusable load assembly, loads are applied to the foundation using the R-System which consists of two stacked steel plates located at the geotechnical resistance balance point or at the foundation base connected to the load assembly via vertical elements. The top plate or the Shaft Bearing Plate (“SBP”) will transfer loads to the foundation upper portion, and the bottom plate or the Base Bearing Plate (“BBP”) will transfer loads to the foundation lower portion as well as the foundation base. At the surface, above the foundation head, a hydraulic jack is located between a Top (“TLA”) and Bottom (“BLA”) load assembly. The TLA and BLA are connected to vertical elements which are consequently connected to the R-system. As the jack is pressurized and expanded at the surface, the R-System plates are separated, and the foundation is bi-directionally loaded. Like the conventional Bi-Directional Load Tests (“BDLT”), from instrumentations, strains within the foundation element, the R-System upward and downward movement, and the jack pressures are measured. From the measured values, loads and displacements are calculated. Due to the test method practical constructability, the TLBT provides a foundation testing system with lower risk of creating challenges during the foundation construction and installation. This paper presents the description of the TLBT method including benefits and challenges during the foundation testing. In addition, comparison results from full-scale load tests performed by both bi-directional load testing methods on adjacent test shafts are included.

Load tests were conducted on vibrated and driven open-end steel pipe piles at a site consisting of alluvial silty sand. The water table at the site was below the level of the piles, giving unsaturated conditions over the embedded length of the piles. Axial uplift tests to failure were conducted on the piles for design of a foundation system to support elevated PV solar panel arrays. Repeat load tests were performed on each of the piles periodically over an aging period of about 2,500 days to evaluate changes in load capacity over time. Laboratory and field tests were performed to characterize the site conditions. In general, the piles showed an initial increase in load capacity between the first loading and the first repeat test but thereafter additional tests showed either a decrease of load capacity or a constant capacity with no gain. Vibrated piles showed higher capacity than comparable size driven piles but also showed similar relaxation behavior. Following the last load test, the piles were extracted to examine the surface of the piles and the extent of any corrosion. A layer of soil about 3 mm–6 mm was strongly adhered to the surface of the piles suggesting that failure occurred in the adjacent soil and not at the soil-pile interface.

Many deep foundation applications require large shaft resistances for tensile loading, such as reaction piles and caissons. A suitable analog for designing surfaces whose load transfer response depends on the direction of loading is the skin along the underbody of snakes. Snakes have developed scales that minimize frictional resistances when they move forward (caudal direction, along the scales) and maximize frictional resistances when they move backward (cranial direction, against the scales). Previous studies show that surfaces designed based on such analogues exhibit directional dependence in terms of both interface friction angle and dilation. In order to assess the shaft and base capacity generated by piles with such surfaces, a series of centrifuge load tests were conducted on model piles with internal strain gages to evaluate the mobilization of shaft and base resistances. It was found that piles installed in the caudal direction mobilized smaller installation forces and similar pullout capacities compared to the rough pile, whereas piles installed in the cranial direction generated greater installation forces and a similar pullout capacity compared to the smooth pile. In addition, spatial variability in the model was assessed using CPT soundings and it was found that the shaft texturing had a negligible effect on the unit base resistance during installation. The results presented highlight that the surface of piles can be designed to reduce skin friction mobilized during installation relative to the skin friction generated during tensile loading resistance.

Recent experience with dynamic pile testing results of Large Diameter Open Ended Pipe piles (LDOEP) has provided some insight to consider when such piles need to penetrate very dense soil or rock layers. In soil conditions where these piles need to penetrate such layers to achieve sufficient geotechnical resistance or fulfill lateral stability requirements, the potential for pile damage needs to be assessed. In addition to potential uniform compression stress at or near the pile toe, consideration should also be given to possible eccentric compression stress. The uniform pile toe compression stresses can be evaluated by wave equation results during the design phase of the foundations. Then, the design team can consider alternate pile diameters and wall thickness to prevent over stressing of the pile toe. During pile installations, dynamic pile monitoring can calculate the pile toe stress based upon the Case Method equations. CAPWAP analysis can further refine the calculation and location of the maximum compression stress. However, none of these methods can directly assess eccentric stresses which might occur where the pile toe encounters the dense soil or rock layer at an angle due to pile batter or sloping soil/rock layer. The paper will present examples where such stresses were encountered, and pile damage was apparent. Based upon the pile monitoring results and the evident pile damage encountered a method to calculate the apparent eccentric stresses is provided. The method proposed would be used in combination with wave equation or dynamic monitoring results to predict the combination of uniform dynamic and eccentric stress for piles driven at an angle or to soil layers with an expected slope.

In May 2017, a monopile (40.5 m in length, 4.5 m in diameter, and a weight of approximately 360 metric tons) was installed in the Dutch sector of the North Sea in a water depth of approximately 28 m to serve as the foundation of an offshore platform. The installation of the pile was somewhat unique: with the use of so-called vibro lifting tool the pile was picked up from a barge, upended to the vertical position and then driven into the seabed without any gripper frame or seabed installation frame. Another interesting aspect was that the plan called for the pile to be vibro driven to the final penetration (approximately 31.5 m below the seabed), if possible. An impact hammer was kept available to drive the pile to final penetration if needed. The feasibility of this approach was confirmed by a vibro driving prediction (VDP) using the results of a cone penetration test (CPT) as input for the soil model (mainly dense to very dense sand with a 2 m thick medium to stiff clay/silt layer near the surface). During installation the vibro driving was monitored with vibro driving analysis (VDA) equipment. Vibro installation hit refusal at a depth of 25 m (due to issues with the hydraulic control units) and thus the final 6 m had to be driven with an impact hammer. Upon completion the installation process was analyzed. In this paper all steps are being described, starting with the original soil model based on the CPT results, the assumptions made regarding the soil behavior during pile driving, the output of the original VDP, the analysis of the VDA data, and the output of the revised VDP based on actual installation parameters. When comparing the latter with the actual installation it clearly demonstrated that in granular soils the original plan to vibro drive a monopile to its actual target depth is not only viable, but can be predicted such that it matches the actual observed results very well.

Pile foundations for bridges are sometimes constructed by placing a pile in a pilot hole which is pre-drilled into rock. In such cases, dynamic testing is typically not performed, and the geotechnical capacity of the pile should be estimated based on static analysis methods. In this case study, the load capacity of an H-pile installed in a pilot hole in a weathered rock was analyzed. The pile was a production pile for a bridge and was tested by a static load test. A numerical model was used to simulate the general load-displacement behavior of the pile. Four static analysis methods were used to determine the geotechnical capacity of the rock. In general, these static analysis methods provided a wide range of the calculated load capacity. Both the numerical model and the static analysis showed that the generalized Hoek-Brown failure criterion and its parameters produced a reasonable estimation of the load capacity for this pile.

Open-ended pipe piles are often used for both land and offshore structure foundation because of their relatively low-driving resistance. During driving, soil enters into the open-ended displacement piles, which cause soil plugging. This paper investigates the plugging effect of open-ended piles jacked into sand. A series of tests were conducted in the laboratory on small scale model pipe piles installed in sand with different soil conditions. The effect of pile diameter and degree of saturation on plugging behavior was evaluated. The laboratory test results show that plug length generally increases as pile diameter increases. Furthermore, the plug length decreases with the increase in the value of matric suction. The tests were also conducted with three different types of sands: coarse, medium, and fine sand. The plug length ratio (PLR) was found higher in coarse sand and lower in fine sand. The jacking force for pile driving was evaluated from the model tests, which significantly depends on the plugged length as well as soil properties. The load settlement curve for the piles indicated that the ultimate load-carrying capacity for pipe piles under dry conditions is greater than saturated conditions.

Despite the widely recognized flexural weakness at threaded connections between steel casing segments of micropiles, relatively simple recommendations are often applied to address the reduced bending capacity at these locations and their degree of conservatism is not well-understood. A methodology to predict the bending capacity and governing mode of failure (jump-out or rupture) at the threaded connections is applied and compared with observations from several independent studies. The results suggest that the 50% reduction in wall thickness and bending capacity often adopted in practice is not always conservative and highly dependent on the thread connection details and geometry. It was illustrated that jump-out is the weak mode of failure. Thread details—including shape, thread angle, and thread length—are examined to illustrate the sensitivity of the failure mode to thread geometry.

The wind turbine foundation design is complex due to the large cyclic loading demands. Thus, investigation on the monotonic and cyclic behavior of wind turbine foundation systems is necessary to ensure safe and cost effective design. This paper provides insight into the behavior of different wind turbine foundation systems under complex loading conditions. The gravity foundation system (GFS) and the tensionless pier system with reinforcing collar (TPC) and without reinforcing concrete collar (TP) are examined. The 3D FEM modeling was conducted. An advanced constitutive model capable of capturing the cyclic and the monotonic behavior of cohesionless soils was adopted. The results from this study demonstrates that the TPC has a superior performance in maintaining the serviceability requirements under large load demands.

This paper presents a quasi-static empirical approach for estimating hysteretic damping of axially loaded piles in an offshore wind jacket structure. Damping is important because it can reduce fatigue in these structures, which may experience many millions of loading cycles. The proposed approach is developed within a ‘t-z’ analysis framework and based off modulus reduction and damping curves derived from dynamic element tests. A procedure and equations are provided to estimate the damping ratio through integration of the elastic strain energy density and dissipated energy density over the deformed soil mass. The relationship between damping and pile displacement, called a ‘D-z’ curve, is proposed for both discrete and lumped foundation models. Validation of the approach is still needed and research is ongoing.

In the design of full-scale piles in the field, base geometry effects are usually not considered on the base resistance calculations due mainly to the limited research and lack of reliable data. In this paper, the effect of two base geometries (flat and conical) on the base resistance of model piles in sand is studied by performing a series of tests in a half-cylindrical calibration chamber with image analysis capabilities. Detailed displacement and strain fields obtained with the Digital Image Correlation (DIC) technique and base resistance measurements obtained with miniature load cells installed in the base of model piles demonstrate that the addition of a conical base to a pile significantly change the deformation pattern of the sand around the base during loading and the magnitude of the unit base resistance. The DIC analyses revealed that soil elements located at corresponding positions in the sand domain undergo up to 30% more displacement below the model pile with a flat base than when the model pile has a conical base. Moreover, the magnitude of the volumetric strains at the soil-pile interface is between 30% and 40% greater at the corner of the flat-base than next to the shoulder of the conical-base model pile. The ratio of the unit base resistance q_b of a pile with a conical base to the q_b of a model pile with a flat base ranges from 0.75 to 0.84 at relative displacement levels equal to 10% of the diameter B of the model pile and from 0.70 to 0.84 at greater relative displacements. The results of this research also indicate that the ratio of the ultimate unit base resistance to cone resistance (q_b,10%/q_c) depends significantly on the base geometry and ranges from 0.95 (in medium-dense sand) to 1.27 (in dense sand) for displacement piles, and from 0.09 (in medium-dense sand) to 0.15 (in dense sand) for nondisplacement piles.

A comparative analysis of the results of field tests to determine the bearing capacity of a pile at the facility in Petropavlovsk is presented. The aim of the study was to carry out a comparative analysis of the results of dynamic and static tests in conditions of swelling soils at one facility in order to identify the difference in bearing capacity indicators. In conditions of swelling soils, the results of static and dynamic tests of the pile have a significant difference in determining the bearing capacity. Dynamic tests show only a preliminary estimate of the dynamic bearing capacity and the possibility of piling in various parts of the construction site. To correctly determine the bearing capacity of a pile in difficult soil conditions, it is necessary to carry out tests with a static indentation load. The results of static and dynamic tests of piles are graphically interpreted and theoretically substantiated.

This project on the southeastern coast of the United States consisted of several structures supported on driven precast pre-stressed concrete (PPC) piles. The driven pile foundation was initially designed utilizing guidance from American Association of State Highway and Transportation Officials’ LRFD Bridge Design Specifications. Using AASHTO LRFD Bridge Design Specifications, the pile lengths and corresponding nominal resistance increased significantly to satisfy design requirements related to downdrag. A result of the longer piles was excessive pile drive time and a risk of pile damage. The driven pile foundation was subsequently redesigned utilizing methods that consider pile behavior under the service load, including drag loads associated with the presence of a compressible layer loaded by several feet of fill. The Neutral Plane method, in general accordance with the Federal Highway Administration (FHWA) Publication No. FHWA-NHI-16-009 guidance document, was incorporated to model pile behavior. The results of the model allowed utilization of shorter piles, which resulted in easier installation and significant time and materials savings.

In the present study, the potential use of energy piles to support the foundation of bridge abutments, high-rise buildings, earth retaining structures, seafront structures, or buildings located on sloping ground has been investigated. In such structures, energy piles are subjected not only to combined axial/horizontal mechanical loads but also to thermal loads. A small-scale laboratory test was conducted using an instrumented model energy pile embedded in dry sand, in which model pile was loaded at its head to the combined allowable axial/horizontal mechanical loading prior to ten heating/cooling cycles with temperature variation of ±1°C. The results evidenced that imposing heating–cooling cycles to the model pile resulted in irreversible settlement and horizontal displacement of the pile head, which gradually accumulated over time and reached 0.6%D and 1.1%D after ten heating/cooling cycles, respectively. The influence of heating/cooling cycles on the mechanical response of the model energy pile was more predominant in the horizontal direction compared to the axial direction. The highest increment of irreversible pile head settlement and horizontal displacement was induced by the first heating–cooling cycle; however, subsequent increments decreased and exhibited a tendency towards stability as the number of thermal cycles increased.

To investigate the effects of surface patterns and roughness on the static interface shear response, the Cyclic Interface Shear Test (CIST) device, which was developed by the research team, was utilized. The CIST system allows for directly measuring the shear stress vs. vertical displacement response of the soil-structure interface. Furthermore, the CIST setup allows for investigating the effects of surface characteristics (i.e., texture, pattern, spacing, and asperity height or roughness) on the soil-structure interface properties. In this study, 3D printed smooth plates simulating conventional steel piles and three rough plates of different surface roughness [i.e., asperity height (h) = 0.34, 0.65, 1.75 mm] were used and then the soil-plate interface was subjected to monotonic shearing at a constant rate. The interface shear responses of the rough surfaces are compared to the soil-smooth and soil-soil shear response. The soil used in the experimental study is a normally consolidated kaolinite clay mixed with sand to reduce consolidation time and to allow for larger number of tests. This paper summarizes the tests setup, sample preparation, and presents the results of a series of the static interface shear tests. The interface shear strength of rough surfaces increased with surface roughness level (i.e., asperity height or h). The preliminary tests presented in this paper show that a contractive behavior was observed at the soil-smooth interface, while the soil-rough plates interface exhibited a dilative behavior. Compared to the soil-soil shear strength, the interface shear strength of the smooth plate (no surface elements) was 45% of soil-soil strength and rough plates were 71%, 74%, and 94%, respectively. Similarly, the interface shear resistance of rough surfaces increased by ~60, 65, and 111% when compared to soil-smooth interface resistance (no surface elements).

This article describes a case history in which temporary segmental casings became trapped in a deep glacially overconsolidated soil. The segmental casings were used to construct 2.5 m diameter drilled shafts to depths up to 30–35 m along the alignment of an elevated transportation structure. Although the advancement of the casings and excavation went smoothly, in several circumstances the casings became stuck during extraction. This situation resulted in abandonment of the concrete placement operation and extensive subsequent work to recover the casings. A recognition of the ground conditions causing this situation led to the identification of locations that could have similar problems and the development of a workable solution. Analytical and numerical analyses were conducted to support the qualitative understanding of the problem and are also described.

Large pile groups (≥=100 piles) are often used to support liquefied natural gas (LNG) tanks to resist not only large vertical structural loads but also significant lateral loads from earthquakes or hurricanes. In general, the design of a large pile group follows standards or prevalent methods that are established based on field or centrifuge test results for a small pile group (≤=25 piles). This does not recognize that lateral group effect of a large pile group is more pronounced than that of a small pile group, and the use of the recommended group reduction factors for a small pile group can be unsafe for the design of a large pile group, in particular for a pile foundation with ≥= 1,000 piles (such as an LNG tank foundation). This study employed a series of 3D continuum finite-element analyses to investigate lateral responses (bending moment, shear force, and lateral displacement) of piles in square pile groups from 3 × 3 to 40 × 40 as well as in circular pile groups supporting LNG tanks. The results were used to examine the difference of lateral responses of piles between single isolated piles, small pile groups, and large pile groups, which would allow for a better understanding of lateral behavior of a large-scale pile foundation supporting an LNG tank.

Suction caissons have been widely considered as foundations in offshore engineering. Installation of a suction caisson foundation in seabed soil is affected by seepage conditions. This study aims to investigate the seepage behavior during caisson installation in sand and sand overlaying inclined clay. Small-scale 1g model tests and numerical studies using finite-element analysis were used to determine the variation and distribution of excess pore water pressure generated by suction in a homogeneous sand. A series of three-dimensional (3D) finite-element models were also undertaken in which installation is considered through sand into an inclined clay layer to simulate the behavior of seepage. The results of this study, which may not be valid for larger scale models, indicate that the inclined impermeable layer (clay) below the sand is affecting the flow in sand, and the influence is dependent on the distance of the impermeable layer to the caisson tip.

Predicting lateral wall displacement due to excavation is essential to the successful design and construction of deep excavations, particularly in dense urban environments where adjacent structures or utilities may be affected by excavation-induced movements. Wall movements are a function of many factors, including but not limited to soil stratigraphy, groundwater elevation, wall stiffness/type, depth of excavation, bracing, and excavation method (e.g., up-down vs. conventional). During construction, surcharges, construction sequence, and means and methods can also impact wall movement. This paper compares theoretical design estimates of lateral slurry wall movement to lateral wall movements measured by inclinometer for deep excavations supported by concrete diaphragm (slurry) walls in the Greater Boston area. While access to sophisticated finite element models has increased in the last several decades, the accuracy of predictions has not. The majority of overestimations occur when performance (i.e., maximum lateral wall movement) is better than average; average performance of excavation support systems has remained consistent over time.

Deep Soil Mixing (DSM) is a versatile ground improvement technology that is used for a variety of geotechnical and environmental applications such as foundations, groundwater cut-off walls, earth retention systems, and liquefaction mitigation. When used for earth retention structures, rows of DSM columns are arranged in cellular or grid patterns to provide a stable structure by improving a limited percentage of the total soil area within the system footprint. DSM earth retention systems are often designed using 2D numerical models, which require the assumption of plane strain conditions that do not explicitly account for the 3D cellular geometry. When 2D plane strain analyses are used, the 3D geometry of the DSM cells and the soil within the DSM cells is typically modeled by smearing (i.e., averaging) soil and DSM properties. When properly analyzed, 2D plane strain models can adequately predict deformations, structural demands, and global stability. However, this smearing approach does not account for the tendency of the relatively stiff DSM to attract significantly greater loads than the adjacent soil. This limitation of 2D modeling can lead to challenges in the interpretation of the stress distribution within these smeared zones, which can only be directly calculated using 3D numerical modeling. This article presents a comparison between the ability of 2D and 3D finite element analyses to calculate the soil/structure interaction and resulting stress distribution between DSM columns and the adjacent soil within the DSM cellular structure. Two types of improved soil consisting of loose and dense sands were considered to extend the applicability of these analyses. Using the results from the calculations performed for this study, design recommendations are presented for the rationale interpretation of DSM stresses in smeared zones from 2D plane strain analyses.

Seepage and stability performance of a 4-m high and 6-m long MSE wall reinforced with metal strips was evaluated under flooding and rapid drawdown conditions by numerical analysis and full-scale physical modeling. A full-scale wall was constructed with a poorly graded sand backfill at an indoor geo-structure testing facility located at the Royal Military College of Canada. Flooding and rapid drawdown conditions were applied by filling and emptying a water reservoir in front of the MSE wall. Variation of water levels with time in the MSE wall backfill was measured by standpipe piezometers and moisture and suction instrumentation installed in the backfill material. Numerical seepage and stability models were calibrated using data from the physical model tests. A 1-m flooding event (measured from the base of the wall) saturated the MSE backfill in 27 h. Backfill desaturation due to rapid drawdown of the reservoir took about two times as long compared to saturation time. Parametric studies were carried out using the calibrated numerical models to investigate variables including wall height, flood height, and backfill hydraulic properties. Simulations showed that limit equilibrium factor of safety can increase by up to 105% after flooding and decrease by 25% after rapid drawdown depending on water pressure head in front of the wall relative to the wall height.

The soil-structure interaction of moment-resisting frame buildings adjacent to a deep excavation in soft to medium clays is presented. An advanced numerical model is used to account for building-soil-excavation interactions, post-yielding behavior of the structural members, and solid-fluid coupled behavior of supporting soils. A traction interface is introduced to account for the gravity loads acting on the building and model the relative vertical movement between the soil and the foundation. The proximity of the building to the excavation, structural aspect ratio, and strength and stiffness of the building are evaluated. The excavation-induced response is presented in terms of ground settlements, angular distortions, angular drifts, and curvature ductilities. Curvature ductilities computed in beams closer to the excavation reached values up to five, developing plastic deformations as deep-seated movement occurred during the excavation of soft and medium clays.

Mechanically Stabilized Earth (MSE) wall technology has frequently been used for highway projects where wall heights up to about 9 m are common due to typical roadway geometry requirements. Only a few MSE structures have been constructed to heights approaching 30 m. One project where a single-tier MSE wall has been successfully constructed to 30 m tall is for the SoFi Stadium in Inglewood, CA. This paper will discuss aspects of MSE technology that were utilized for solving construction and design challenges on the project. These aspects include significant shortening of MSE soil reinforcing strip lengths at the bottom of the MSE wall to reduce excavation and creative use of alternative facing and backfill types. Unique wall geometries were utilized to achieve isolation of building columns for seismic events as well as required staged construction. Prefabricated wall components were staged on-site ahead of wall construction allowing for coinciding site preparation and allowing for rapid wall construction at the appropriate time.

Obstructions behind Mechanically Stabilized Earth (MSE) walls are a very common occurrence on Department of Transportation projects throughout the United States. The types of obstructions vary from drainage structures such as box culverts, inlets, vertical and lateral pipes, etc. to other types of obstructions including, drilled shafts, steel piling, temporary shoring for phase line construction, acute corners, existing foundations, and many more. Most of these types of structures are required on large highway projects. An MSE wall is structurally supported by soil reinforcements buried in compacted granular soil. Conflicts can occur when structures and pipes are also required behind the MSE wall, which can hinder the placement of the reinforcing strips. Typically, the contract plans developed for the Department of Transportation are by consultants, which only specifies the MSE walls. The actual design and detailing of the MSE wall are performed by the wall supplier. Sometimes, the consultants do not understand the intricate details in designing and detailing an MSE wall and how obstructions affect the design and construction of the MSE wall. Obstructions behind MSE walls cannot always be avoided and at times the obstruction can be easily accommodated. However, there are times in which the obstruction either hinders placement of the reinforcing strip or the reinforcing strip cannot physically be attached to the MSE wall. There are typical methods used to relocate reinforcements around obstructions such as a structural frame or angle irons. When possible, discrete reinforcing strips may be skewed to clear the obstruction, but other complicated situations can exist that require a more elaborate design. Depending on how the obstruction is relative to the MSE wall, it could make construction very difficult or impossible. The soil reinforcement needs to be installed perpendicular to the MSE wall and any hinderance could lead to internal stability and construction issues. Obstructions behind the MSE walls cannot always be avoided, but with coordination between the civil, structural, and MSE wall engineer the obstructions may be modified to compliment the MSE wall and could eliminate concerns of internal stability and construction issues. This paper will discuss and present feasible solutions that consultants could consider when developing the contract plans. These feasible solutions could help avoid challenging circumstances in both the design and construction of MSE walls.

Mechanically Stabilized Earth (MSE) walls are a very common type of retaining structure in the recent world. Many countries use them to stabilize highway slopes and hill slopes or vulnerable soil masses. In this study, numerical analysis on the construction of an MSE wall located in Abilene, Texas, has been carried out. The backfill is reinforced with geogrids. Finite-Element Method (FEM) based software PLAXIS 2D is used for numerical modeling. The objective of the study is to analyze the effects of various geogrid parameters on the overall stability and performance of the wall. Two design parameters namely, the factor of safety and wall deflection were considered for the evaluation. The effects of geogrid spacing, length and strength, and friction angle of the backfill soil have been investigated on the performance of the MSE wall. It was observed that the factor of safety increased with the increase in strength and length of geogrids. A significant decrease in displacement was observed for an increase in geogrid length to wall height ratio from 0.5 to 0.7. Moreover, with the increment of vertical spacing of geogrid, the horizontal wall deflection decreased. Finally, three optimum design combinations are proposed which can be helpful for the engineers by providing them a suitable starting point for the design of an MSE wall.

This paper presents a case history of excavation support cutoff wall system using interconnected jet grout columns reinforced by a combination of multiple rows of tie-back anchors and H-steel beam piles. To define the optimal grouting parameters and to determine the mechanical properties of jet grout columns, several trial jet grout columns were executed in coarse grain soils. Taking the presented case history into consideration can be pointed out that how effectively the jet grouting technology can be utilized on various geotechnical applications specifically in gravel and cobble strata. Moreover, the extensive quality control and assurance methods conducted on the site, before and during the construction phase are elaborated here.

Cracks are often found at the crest of a slope in backfill soil behind retaining structures prior to their failure. The presence of cracks increases lateral earth pressure incurred in retaining walls by reducing the soil shear resistance. This paper employs an analytical solution for determining the tension crack depth (TCD) in unsaturated backfills and quantifying the effect on active earth pressure (AEP). Tension cracks occur at a depth where normal stress transitions from compression to tension. The maximum TCD is determined by evaluating the normal stress distribution along the critical log spiral failure surface using a limit equilibrium effective stress analysis. A parametric study is performed to investigate the effects of cohesion, angle of friction, air entry suction, and flow rates on TCD and AEP. Results suggest that tension cracks significantly increase AEP. TCD is found to be dependent on soil properties. Fine-grained soils with lower air entry suctions could generate deeper cracks. Increase in soil cohesion increased TCD while soil friction angle had minimal effect on tension crack formation. Finally, deeper cracks were developed during evaporation when higher suction developed within the soil layer. The proposed solution can be used in forensic studies of failed structures to determine the depth and strength of required reinforcement layers to be placed in marginal backfills to prevent the formation of cracks or rehabilitate the existing cracks.

The study investigates the complicated states of stresses around a large opening in an underground shaft and surrounding soil for a wide range of K_o (coefficient of earth pressure at rest) ground conditions by performing nonlinear 3D finite-element analysis. For this study, the magnitude of K_o ranged from 0.50 to 1.50 with an increment of 0.25, that is, a total of five cases were considered (K_o = 0.50, 0.75, 1.00, 1.25, and 1.50). The chosen geometry of the shaft is based on a construction site of North East Boundary Tunnel (NEBT) Project by DC Water. The original layered soil is simplified, and a uniform homogeneous silty-sand and clayey-sand (SM-SC) underlain by a bedrock layer was considered for the analysis. In the present study, the performance of the shaft and adjacent soil was analyzed when using 5.2-m (17-ft) diameter breakout into a 7.6-m (25-ft) diameter underground shaft at an approximate depth of 19.8-m (65-ft) from ground level. Ground improvement by soil-freezing was implemented to stabilize the adjacent soil mass at the breakout location prior to excavating the opening on the shaft wall. Sequential construction stage analysis (seepage and stress analysis) was performed in 3D nonlinear finite-element platform (using Midas GTS NX software) to simulate the construction sequences followed in the field. The most critical construction stage in terms of stability of shaft and soil is observed when the soil is excavated inside shaft and breakout opening on shaft wall is executed. Maximum hoop stress in the shaft is very sensitive at crown and invert locations of the opening to the variation of K_o magnitudes. It was observed that increasing the K_o from 0.5 to 1.5 (300% increase) caused the stress at the invert to increase by about 170%. In longitudinal direction, the influence of variation of K_o on stresses is observed to be not very significant. Formation of plastic zone in improved soil located around the breakout opening zone occurred and the extent of the plastic zone expanded to larger distances with the incremental increase of K_o magnitudes. The soil located in the bottom plug of shaft experiences cap failure due to unloading of overburden pressure for excavation.

Mechanically Stabilized Earth (MSE) walls have seen significant growth over the past decades due to their economics and reliable performance. Even though the Departments of Transportation are widely implementing them all over the United States, there is a gap in the current state of knowledge regarding sliding resistance of MSE walls. The objective of the current study is to evaluate a relatively novel approach of minimizing or eliminating sliding failure of MSE walls using Recycled Plastic Pins (RPP). Three field test sections were constructed where the base of the sections was reinforced with RPPs. About 3 m long RPPs were driven into the wall foundation as shear keys to improve the lateral resistance of the wall base. A control section without RPP reinforcement was also constructed to compare the effectiveness of the approach. Based on preliminary vertical inclinometer results, the lateral displacement of the reinforced MSE wall base was reduced by 76%–89%.

Construction of retaining walls on weak foundation soil is usually susceptible to a huge amount of ground settlement and bearing capacity failure. A variety of techniques can be applied to address this issue, such as removal and replacement of existing soil with competent soil, staged construction, dynamic compaction, stone columns, and geosynthetic reinforcement. A pile-supported system in combination with geosynthetic reinforcement can provide a sustainable solution for ground improvement. The use of Recycled Plastic Pins (RPPs) can be a viable alternative to other conventionally used piles. The objective of the current study is to compare the field results and numerical analysis results of an RPP reinforced foundation system. A field-scale study was conducted to evaluate the effectiveness of RPPs for reducing the settlement of wall foundation. A numerical analysis was conducted using the 2D finite element program PLAXIS 2D. About 30%–88% reduction in the vertical movement of the wall foundation was observed due to the presence of RPP reinforcement.

This paper presents the effects of excavations in Los Angeles on the surrounding ground surface. Three large excavations with varied support systems for stations constructed as part of the K (Crenshaw/LAX) Line Transit Project provide an opportunity to acquire data through an extensive geotechnical investigation and field monitoring program to further our understanding of soil-structure interaction and excavation-induced ground deformations. The ground conditions encountered on site include alluvial deposits of silts, clays, and sands. The data acquired illustrates that ground displacement and wall deflection largely depend on the stiffness of the excavation support system. Support systems with relatively high stiffness are able to successfully limit surface settlement behind excavations. The data shows that heaving caused by unloading governs the soil response given the relatively small lateral support wall deformations. Current empirical models are unable to capture this ground behavior. An update of these empirical relations is needed to represent this behavior in support of the design of future excavations.

Current approaches for seismic design of retaining structures can be broadly characterized as analytical, numerical, or experimental. All three approaches involve simplifying assumptions to provide a tractable solution. Therefore, use of the methods involves interpretation of the output in the context of the underlying assumptions, which necessitates understanding the inputs and their meaning. Misunderstanding the inputs and assumptions often produces poor comparisons amongst the methods when used to “validate” the applicability of a method over another. In addition, comparison of analysis results to case histories and identifying discrepancies in the assumptions is generally not performed in a consistent manner to ascertain the validity of the assumptions used in the analyses. In this light, the focus of this paper is to highlight the consequences of the basic assumptions inherent in the different methods of analysis.

Slope stabilization using soil nails is implemented in homogenous soil slope subjected to seismic loading and considering a planar failure surface passing through toe of the slope. Closed form expressions, for determining the equivalent pullout capacity of the nails and factor of safety, are proposed considering the influence of surcharge load, seismic forces, angle of the failure plane, slope angle, inclination angle of the soil nails, and spacing between the nails. It is seen that the increase in soil nail inclination angle increases the equivalent pullout capacity of the nail and the corresponding failure angle of the wedge. An increase in horizontal spacing between the soil nails reduce the equivalent pullout capacity of the nails and angle of the failure wedge. The simplified trial wedge method serves as an accessible method for initial analysis of nailed soil slopes and also a check on advanced methods employing complex techniques to compute the factor of safety at critical conditions. The proposed design charts will be fruitful for geotechnical engineers in assessing the stability of nailed soil slopes in earthquake prone areas.

Many geotechnical engineers rely on corrosion control guidelines provided by transportation agencies when providing corrosion control recommendations for residential and commercial structures without realizing that highway guidelines focus on carbon/structural steel and concrete with little thought regarding aluminum, copper, brass, or stainless steel. As highways deal mainly with rebar and carbon steel piles, general (large area) corrosion is more of a concern than localized pitting corrosion which can produce leaks. Most agencies suggest that soil resistivity is the prime factor to categorize soil corrosivity only recommending additional testing of soil pH, minimum resistivity, water soluble sulfates and chlorides, if the resistivity is below 1,500 ohm-cm. These tests have come to be known as the “Corrosion Series” by geotechnical engineers with only sulfate testing being required by the International Building Code (IBC). By not also testing for ammonia, nitrates, sulfides, and REDOX the potential corrosion risks for other important general construction materials for residential and business projects are missed. These other factors affect pitting and intergranular corrosion of stainless steel and copper based alloys greatly as well as probability of corrosive bacteria presence also known as MIC. In order for the soil side corrosion to occur, there must be moisture present to allow ion exchange in the oxidation reduction reactions. Few remember that underground condensation can occur on metal surfaces and that significant dew can fall in the evenings. In the opinion of this author, it is hoped that the 1,500 ohm-cm limit is removed or raised to 5,000 ohm-cm and that four extra tests are added to the “Corrosion Series” typically ordered by geotechnical engineers to avoid pipeline leaks, failure of copper water lines, copper electrical grounding systems, aluminum hardware, stainless steel hardware, ductile iron pipe, and brass hardware and fittings.

Occasionally, in the design of deep foundations for a structure, a single deep foundation element may be used to support a building column. The most common type of element used in such cases is a drilled shaft (also referred to as a drilled pier, caisson, or bored pile). However, as installation capabilities have improved, large diameter augercast piles are increasingly more common. While drilled shafts and augercast piles differ significantly in construction methods, in the eyes of the International Building Code (IBC), both are treated as cast-in-place deep foundations and must comply with many of the same code provisions. Deep foundations are required to be braced such that lateral stability is provided in all directions, however isolated, cast-in-place elements that have a diameter (D) of at least 2-ft and a length (L) not more than 12 times the diameter can be an exception. In response to the concerns raised by some member professionals, the DFI Augercast Pile Committee took a closer look at the code and its applications, the historical development of the code language through the years, and performed analytical modeling to evaluate the impact of this ratio on a pile’s stability. The question is whether the limiting ratio included in the IBC 2018 code provisions is justified, or whether different criteria should be used to determine when isolated deep foundation members are allowed.

Tunnel and underground constructions usually involve various uncertainties. Besides the uncertainty resulting from ground material, the design and construction variabilities also create another form of uncertainty. Current tunnel design practices, however, have not routinely and directly included uncertainty analysis into a systematic procedure, which in some cases may lead to an unsafe or over-conservative design. This study proposed an approach for reliability assessment of sequential excavation method (SEM) tunnel construction in urban soft ground settings. Key geotechnical properties were selected as random variables and support parameters were considered as design variants. Three-dimensional (3D) numerical modeling was used to better model ground-tunnel interaction. A surrogate approach was developed to substitute costly 3D simulations. The limit state function integrates both deformation tolerance and liner capacity, and the failure probability is calculated via the crude Monte Carlo simulation (MCS) and the first-order reliability method (FORM). The reliability results indicate that the increase in soil secant modulus E^ref₅₀ and cohesion highly improves the settlement reliability. An increase of shotcrete thickness is favorable to both settlement and liner reliability. The shotcrete parameters are more sensitive to the early-stage liner reliability, implying the critical role to maintain sufficient shotcrete strength and thickness at an early stage.

Three-dimensional (3D) geological model, as an effective tool for modelling the complex nature of underground geological conditions, has been used for the design and construction of underground facilities. The accuracy and reliability of the geological information reflected by the 3D model are critical for engineering practitioners to make sound judgement. It is important to consistently update the 3D geological model with new borehole or geophysical data for better assessment of the underground uncertainties. In this study, an application of the NTU-BCA 3D geological model to an undergoing subway project in Singapore has been investigated. The established 3D geological model was updated using the new borehole data obtained in the project. A comparison of the virtual boreholes generated in the 3D model with the real borehole data in terms of rockhead positions was made as a verification of the uncertainties of the model. The uncertainties analyses show that the standard deviations of the predicted errors for the rockhead are improved after the new borehole data was incorporated into the model. Predictions made by spatial statistical analyses are also presented and compared with the borehole data and the results from the 3D geological modeling. Through the study, an example for using newly obtained borehole data to update the existed 3D geological model to improve the accuracy was given.

Sydney’s Central Station is the busiest railway station in Australia. The upgrade of the 115-year-old station to incorporate Sydney’s new metro lines involves the excavation and top-down construction of a 27-m-deep station box in the heart of the operating station. This paper describes the unique underpinning design for the existing pedestrianized Devonshire Street Tunnel (DST), which allowed it to remain open to customers throughout construction of the station box. This paper aims to present the structural and geotechnical design and staged construction methodology, describe the interface between geotechnical and structural works, and discuss risks and mitigation measures adopted during design development and construction.

This paper presents an implementation of Random Forest (RF), a supervised learning algorithm, to classify the encountered geologic conditions using continuous earth pressure balance machine (EPBM) operation data. This study was performed on a data set from State Route 99 (SR99) tunnel construction in Seattle, WA. Hyperparameter tuning was performed to investigate the effects of RF hyperparameters on the classification performance as well as to determine the best hyperparameter configuration. The role of the features in the classification model was investigated by evaluating the feature importance measures. This study demonstrates that, with straightforward hyperparameter tuning, RF could deliver good classification performance and could infer the geologic transition through the classification probabilities. This study indicates that although several EPBM features had relatively larger “weights” for the classification, it was the interactions among the features that contain the geologic information.

This paper presents the design and implementation of groundwater cutoff and dewatering scheme for cut-and-cover construction of an underground subway station in Manhattan, New York. The excavation was approximately 60 ft wide and 1,800 ft long with depths ranging from 60 to 70 ft below street level. The invert of the station rested partially on rock at one end but mostly on weak glacial deposits for the remaining portion of the station and a transition tunnel. Groundwater cutoff was achieved using secant pile walls and diaphragm walls. A dewatering system consisting of a combination of shallow sumps and pressure relief ejector wells was adopted. Hydraulic properties of the soils were evaluated from particle size distribution, in situ falling head tests, and a full-scale pumping test. Field pumping volumes were recorded using flowmeters at discharge points to ensure pumping capacity was adequate at all times. Piezometric head within the excavation was observed using selected ejector wells as open standpipe piezometers. Hydraulic pressure build-up beneath the invert slabs were closely monitored as the pressure system was progressively deactivated to ensure the partially completed underground structure was stable against uplift water pressures. The varved character of the glacial deposits resulted in varying residual pressures acting at the invert base after the pressure relief system was de-activated.

In underground mines, a stope can be used to supply workers and machinery with a safe and secure work area. It is important to properly assess the pressures applied to walls and barricades placed close to the center of the stope to preserve the filling materials in order to ensure the secure application of the backfill. An analytical formula describing the horizontal pressure of the barricades is therefore needed. However, the investigation of this analytical formula has not attracted many researchers, although it is important to avoid accidents due to the failure of the barricades. The proposed approaches to investigating the stress of barricades are discussed in this research. The authors then propose a new two-dimensional analytical solution based on the formula of Mastron. The solution presented depends on dividing the stope drive into three segments at three stages of the solution. This solution considers the transfer of load from the backfill to the walls of both the stope and the access drift. In general, the solution takes into account the effect of the stress state of the barricade position within the drift. The results are compared between the solution proposed and the test results for the physical model and the numerical simulations. A reasonable agreement between the different results suggests that this proposed formula provides a good forecast of the pressure on the barricade. For fully drained granular backfill, this proposed solution presented here applies to backfill without water-pore pressure.

TBM uses disc cutters to break the hard rock in mixed face conditions (MFC), which will induce dynamic loads both on the TBM and the working face. Besides, when the cutter moves from the soil into the hard rock, it will instantaneously bear an impact load at the soil-rock transition zone. These dynamic interactions involve the dissipation of energy into the ground and, therefore, cause significant levels of ground-borne vibration, potentially damaging the existing structures or causing human perception and disturbance. This work conducted field monitoring on the ground vibration induced by TBM excavating in MFC in Jinan, China. The monitoring scheme included five accelerometers buried in the ground along with depth. The waveform characteristic of ground-borne vibration was analyzed. Then the influence of the distance between the cutter and monitoring points on the vibration was investigated. Finally, the distribution and attenuation of ground vibration along with depth were fitted by the exponential function. The findings can help engineers better understand the ground vibration characteristics and conduct suitable countermeasures.

Concrete pipes are widely used as water retention/detention facilities in civil engineering projects, which are typically buried in a shallow depth (i.e., 0.6 m). Concrete pipes are typically installed at the early stage of projects to ensure the drainage capacity of the construction site, and the heavy construction vehicles may cause damage to the pipes. However, most of existing methods for calculating additional pressure on the pipe caused by live load could not consider the soil arching effect in the soil cover, which may result in underestimation of the pressure at pipe top. Numerical modeling was adopted in this study to investigate the performance of concrete pipes under a heavy crane load including pipe deflection, pressure at pipe top, and maximum bending moment. A parametric study was also conducted to evaluate the effects of pipe diameter, soil cover thickness, and magnitude of external load on the performance of concrete pipes.

Some corroded corrugated metal pipes (CMPs) buried under roadways have been reached the end of their service life. Maintenance, which includes excavation and replacement of the damaged pipes, will disrupt the traffic and require a considerable cost. Trenchless methods of rehabilitation, such as sliplining and cement mortar (spray-on) lining, have been increasingly utilized to repair buried damaged metal pipes. However, few studies have been carried out to assess the performance of the rehabilitated corroded pipes. In this study, a numerical analysis was carried out to assess the performance of buried corroded CMPs rehabilitated with sliplining and cement mortar lining methods. The available field test results of the corroded CMP before and after rehabilitation were used to verify the numerical models. The modeled CMP had an inside diameter of 600 mm and a wall thickness of 2 mm, with corroded invert, which was located at 0.4 m depth under the pavement layer. Typical properties of a low-viscosity grout were used to simulate sliplining and cement mortar liner grouts. For the sliplined CMP, a high-density polyethylene (HDPE) liner pipe with an outer diameter of 550 mm and a wall thickness of 20 mm was modelled. In the cement mortar lining method, the thickness of the cementitious liner was 50 or 75 mm. The load-carrying behavior of the rehabilitated CMPs, their horizontal and vertical diameter changes, and settlement of the pavement surface were investigated when the rehabilitated buries CMPs were subjected to surface loading. The results show that the cement mortar rehabilitation method provided approximately 10% higher load-carrying as compared with the sliplining method at a same vertical diameter change. The rehabilitation by sliplining and cement mortar lining methods notably reduced the displacement of the pavement as compared with no rehabilitation.

Ultimate bearing capacity (q_u) of footings on undrained soft clay stabilized with a granular trench was thoroughly investigated using the upper-bound limit state plasticity failure approach with discontinuity layout optimization. The investigations were performed in a parametric fashion to determine the ultimate bearing capacity for a wide range of practical configurations (i.e., geometric and material combinations). Solutions from the limit analysis using discontinuity layout optimization (LA-DLO) have provided a deeper insight into the influence of the various design parameters and their intricate relationships. Solutions from LA-DLO simulations were compared and found largely agreeable with solutions from published analytical derived-expressions, finite element analyses, and small-scale model tests. The critical depth of the trench, which defines the cut-off whether the critical shear failure is considered shallow or deep was successfully established. This critical depth was found to significantly depend on D_f/B, W/B, and ϕ. Finally, bearing capacity factors N_c, N_q, and N_γ were back-calculated and presented in a tabulated form.

Wherever a hydraulic barrier is required, waterproofing membranes are often employed in construction. Because they sit at the boundary between the natural and built environments, interface modeling is an important part of their assessment. This highlights the value of maintaining an experimental test database on the behavior of waterproofing membranes in contact with various soils and environmental conditions. This study is part of a series of large-scale laboratory tests on soil-structure interactions focused on the friction mechanism at various interfaces with boundary conditions as similar to reality as possible. The aim of this paper is to investigate the interface shear resistance of the waterproofing membrane interface in contact with dry and saturated Ottawa sand. To this end, a spray-applied waterproofing membrane was sheared on Ottawa sand placed in a sandbox with interior plan dimensions of 305 × 305 cm and an interior height of 100 cm. The coefficient of sliding friction at the interface is characterized, and the surface failure mechanism is also discussed. The waterproofing membrane-sand interface friction angle was found to be independent of normal pressure and quite different from the internal friction properties of the Ottawa sand. It was observed that the shear resistance at the interface of saturated sand was greater than its dry counterpart. The data sets are being used to support the development of numerical models for sliding at waterproofing membrane-soil interfaces.

Using PTI DC10.5-19, geotechnical engineers can provide recommendations and design parameters which do not require extensive site work (such as removal and replacement) of expansive clays that result in cost-effective designs. Extensive site work is often a significant cost and time-consuming operation. The increasing use of the PTI DC10.5 method and post-tensioned reinforced designs demonstrate that post-tensioned slabs are an economical solution to building foundations. PTI DC10.5-19, Standard Requirements for Design and Analysis of Shallow Post-Tensioned Foundations on Expansive and Stable Soils is an IBC-referenced standard for the design of slab-on-ground foundations. Misconceptions exist regarding whether the potential vertical rise (PVR) index often determined in site investigations can be used in place of the value of differential soil movement, (y_m) defined in PTI DC10.5-19. This paper will outline the requirements of relevant IBC sections, explain the differences between PVR and y_m, discuss why they are not interchangeable or relatable, and review the geotechnical aspects of design of a post-tensioned slab-on-ground foundation using PTI DC10.5.

The design and construction of underground openings in shale type of rocks requires the understanding of coupled effect of anisotropy and polyaxial stress state. Therefore, in the present work, an attempt is made to develop a methodology for analyzing anisotropic rocks under a polyaxial stress state. The methodology developed in the present work depends on basic experiments (Uniaxial and Triaxial) and the Discrete Element Method (DEM) based numerical modeling. It is observed that the interaction properties between two layers of the anisotropic rock are significant for the numerical simulation of underground structures. The interaction behaviour of two layers is dependent on shear stiffness (Ks), normal stiffness (Kn), joint tensile strength, joint friction angle, and joint cohesion. Therefore, in first step the interaction properties are found by simulating experimental uniaxial and triaxial results. Finally, the obtained properties are used for the simulation of coupled effect of polyaxial stress state and anisotropy.

Engineered cavities in rock formations are often part of an underground transportation infrastructure. Three measures of roof stability in such cavities are discussed: the stability number, the factor of safety, and the support pressure needed to prevent cavity roof failure in weak rock. The stability number is a dimensionless combination of the rock properties and the size of the cavity when roof failure becomes imminent. While there exists substantial experience in application of the stability number and the factor of safety to soil structures, their use to define the safety of rock structures is intricate. This is because the strength envelope for rocks is a non-linear function of the mean stress. The specific function used in the analysis is the Hoek-Brown failure criterion. The kinematic approach of limit analysis is used, and the results are presented in charts. All measures of stability are strongly dependent on the Geological Strength Index, and, to a lesser degree, on other parameters in the Hoek-Brown failure criterion.

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 03 30 2022.