International Journal of Geomechanics

An analytical solution is presented for plane-strain poroelasticity induced by surface-normal loading within a finite rectangular fluid-saturated domain. In this work, the porous medium is assumed to be isotropic, homogeneous, and composed of compressible constituents. Appropriate Fourier and Laplace transforms and inversions are used to derive the analytical solution. In particular, lateral boundary conditions for the displacement field are specifically prescribed with care to match the Fourier and Laplace transforms. The analytical solution is presented in terms of summations of multiple series. Then, a case study for the consolidation of a water-saturated soft clay layer is carried out. The analytical solution is compared to the finite-element (FE) solution through a nonlinear transient FE analysis software program. The excellent agreement between the FE solution and the analytical solution demonstrates the accuracy and reliability of the analytical solution presented in this paper. Meanwhile, an analysis of coupled hydromechanical behavior is conducted herein. It depicts that the Mandel-Cryer effect occurs, which is a distinctive feature of poroelastic theory. The presented analytical solution can act as a benchmark for the validation of numerical solutions in plane-strain consolidation. It is helpful also for a further understanding of time-dependent hydromechanical behavior of finite two-dimensional fluid-saturated porous media.

Determining a suitable distance between two adjacent panels’ working faces (DBTF) is an important issue that should be considered in underground mining methods. A suitable DBTF causes a decrease in stresses on the pillar, reduces the width of the pillar, and controls the ground subsidence. The main purpose of this study was to determine the suitable DBTF in shortwall mining of the Tabas central coal mine. For this purpose, at the beginning, an algorithm is presented for numerical modeling. In this algorithm, front abutment stresses first and then side abutment stresses are evaluated on the chain pillar. Numerical modeling is performed by finite-difference method software in different distances of P5 and P6 panels’ working faces (5–45 m). The result of this research shows that the front abutment stress on the chain pillars is not different for a distance of 25 m or more. Moreover, by increasing the distance up to 25 m, the side abutment stress in the vicinity of the P5 panel face is decreased on the chain pillar and is increased on the chain pillar in the vicinity of the P6 panel’s working face. However, when the DBTF increases to 30 m, the lowest side abutment stresses are applied on the chain pillars and then increased for greater distance. It can be concluded that a suitable DBTF in the Tabas central coal mine is 30 m.

When the limit-equilibrium method is employed in the stability assessment of complicated landslides, some eccentric results are often encountered with the use of rigorous methods. These eccentric results may be attributable to irrationally assumed interslice forces. Furthermore, engineers should verify the rationality of the thrust-line position in a landslide after the safety factor of the landslide is solved with the use of such rigorous methods. In this paper, a new global analysis method is proposed based on the Morgenstern–Price assumption for interslice forces. The global equilibrium equations, normal stresses on the sliding surface, and position of the thrust line were employed as constraint conditions to construct an optimization model. The relationships between the constraint conditions and the safety factors and thrust line were investigated through the analysis of some typical landslides with some interesting results.

A high arch dam project causes an intense disturbance to its geological environment, with high in situ stress in the critical instability state, and the creep and damage of abutment rock mass resulting from the process of nonequilibrium evolution have a serious impact on the long-term safety of an arch dam. In this study, the long-term stability of an arch dam was evaluated with time-dependent deformation reinforcement theory (TDRT), in which plastic complementary energy (PCE) was used to judge the steady state or viscoplastic flow of the structure, the overstresses beyond the yield criterion were the driving force for nonequilibrium evolution of rock structures, and the unbalanced forces were the required reinforcement forces. Based on a viscoplastic model and asymptotic stability analysis, the principle of minimum plastic complementary energy was proved for perfect and hardening yielding, in which viscoplastic structures deform to the limit steady state at which the PCE is minimized under time-invariant loading and boundary conditions. Thus, the PCE is a reasonable and quantitative criterion for stability evolution, and unbalanced forces can be used to determine the reinforcement because they have a completely mathematical basis. The expression of PCE and unbalanced forces for the finite-element method (FEM) were programmed, the theory was implemented in a parallel FEM code, and the long-term safety of Jinping Arch Dam in China was evaluated.

In this study, a novel analytical approach is proposed to calculate the passive earth pressure against a rigid retaining wall subjected to the translation mode. Effects of arching in cohesive backfill soils as well as friction mobilized along the wall–soil interface are considered. Analytical expressions of the earth pressure, passive resistance, and its height of application on the backface of the retaining wall were derived based on static equilibrium of forces under passive conditions acting on the failure wedge. A parametric study was undertaken to assess effects of internal friction angle of backfill, wall–soil interface friction angle, surcharge pressure, cohesion, unit weight, and inclination angle of the slip surface on the active earth pressure as well as on the slip-surface angle. The results of the proposed method were then verified against the existing test data as well as the predictions by Coulomb theory and Rankine theory. The results show that the proposed method yields satisfactory results.

A tunnel excavation-damaged zone (EDZ) will gradually form with increasing release ratios of displacement (RRD) in the rock mass surrounding a tunnel. This paper presents a study of the longest subway tunnel in China, constructed with tunnel-boring machines (TBM) and the new Austrian tunneling method (NATM) in soft and weak rock ground. On the basis of the damage constitutive model, some studies are conducted using the FEM. The spatial distribution rules of EDZ and RRD are studied with a three-dimensional (3D) numerical simulation. The results show that the spatial support effect of the tunnel face is concentered in the range of approximately two times that of the tunnel diameter. A two-dimensional numerical simulation is carried out to study the relationship between the EDZ and RRD. The results show that the EDZ initiates when the RRD is more than 0.5, and the EDZ suddenly increases when RRD is more than 0.9.

This paper presents the results of large-scale numerical modeling of cellular mattress–reinforced fly ash beds overlying soft clay using a finite-element program. The cellular mattress was a honeycomb structure consisting of interconnected multiple circular cells. The influence of the height, diameter, and tensile stiffness of the cell and the width of the entire mattress on the pressure-settlement response of footing, surface deformation during footing settlement, and mobilization of hoop tension in the cell walls are illustrated. Results from the numerical analyses indicate an improvement in footing capacity of approximately1.4 times greater over fly ash bed by inclusion of a single geotextile separator representing jute geotextile in between the fly ash bed and underlying clay. The cellular mattress–fly ash composite bed produced an approximately sevenfold increment in the footing capacity compared with the unreinforced fly ash bed both in presence of the jute separator. The mattress-reinforced beds produced better footing capacity with an increase in the height and width of the mattress and the tensile stiffness of the cell wall. It is satisfactory to acquire the optimization for the height and width of mattress and the tensile stiffness of the cell wall. For a particular mattress width and height, the footing capacity increased with a reduction in the cell diameter. The cell at the mattress center mobilized maximum hoop tension that was lesser in the cells successively toward the mattress periphery. Also, more hoop tension was mobilized with an increase in the tensile stiffness of cell wall. Small-scale finite-element models were created using the same material models and properties as those used for the large-scale modeling to validate the program with laboratory small-scale model tests comprising the same model conditions. The finite-element results were found to be in good agreement with the experimental results.

In this study, using the background of Qiyueshan high-risk karst tunnels, flow characteristics and escape-route optimization after water inrush in a backward-excavated karst tunnel are researched. First, five case studies are simulated and investigated under different velocities of water inrush. One special probing line is selected, respectively, in the left tunnel, cross passage, right tunnel, and along the height direction of the tunnel centerline. For each probing line, the corresponding velocity and pressure curves are obtained based on the numerical simulation results. The change rules of velocity and pressure under the five water-inrush velocities are analyzed and concluded. Then, focusing on one velocity of water inrush, the variation laws of velocity and pressure in a plane are analyzed and discussed by selecting a series of probing lines. Flow characteristics of water after inrushing in a backward-excavated tunnel are summarized. Finally, for such cases, the optimized escape routes are made in the tunnels. The results show that (1) as the water-inrush velocity increases, the flow velocity in the tunnels also increases proportionately; (2) as the water-inrush velocity increases, the pressure in the tunnels also increases, but there exists no direct linear relationship between them; (3) there is little flow velocity close to the tunnel side wall as well as at the working face without water inrush and at the high location of the cross passage; and (4) the pressure at the intersection area of the cross passage and the tunnels changes greatly. Research results provide a theoretical basis for making scientific and rational escape routes.

A series of pullout tests were conducted on a model pile in a soil with different initial water contents and different pile surface roughnesses to study their influences on pile skin friction. To measure axial strains and skin friction of the pile, fiber Bragg grating (FBG) strain sensors were used. Data were collected during the pullout tests and are analyzed in this paper. Test results indicate that the skin friction and axial strain of the model piles decrease with the initial water content of the soil but increase with the degree of surface roughness of the piles. The axial strain induced in the upper segment of the tested pile is considerably greater than that in the bottom segment of the pile. In addition, the skin friction is distributed nonlinearly along the length of the model piles.

The use of p-multipliers in the analysis of the lateral loading behavior of pile groups is based on the concept of modifying the single-pile p-y curve to obtain the p-y curve for a pile in a group. The p-multipliers account for the reduced soil resistance mobilized at a given deflection as a result of overlapping of shear zones. Different factors can influence the p-multipliers; however, because of the lack of experimental data, most researchers and design guidelines consider only the normalized pile spacing in the direction of loading in their recommendations. In this study, the effects of pile spacing and clay stiffness on the p-multipliers were investigated using the results of two series of centrifuge tests. The soil profile consisted of four lightly overconsolidated clay layers overlying a dense sand layer. The pile groups had a symmetrical layout consisting of 2 × 2 piles spaced at center-to-center distances of 3.0 and 7.0 pile diameter (D) and were driven into improved (stiff clay) and unimproved soft clay. Ground improvement was done in situ using simulated cement deep soil mixing (CDSM). Computer analyses were performed to back-calculate the p-multipliers. There was very good agreement between the measured and computed responses for both the leading and trailing rows of piles in the unimproved and improved soft clay. The results reveal that increasing the clay stiffness and pile spacing in the direction of loading increase p-multipliers. No pile–soil–pile interaction was observed for the 7D spacing. The proposed set of p-multipliers for the soft clay was found to be in close agreement with some current guidelines, whereas other design guidelines appear to recommend relatively conservative values of p-multipliers for both soft and stiff clay (improved ground).

A two-surface bounding surface plasticity model for the undrained axial pile response in fine-grained soils is described and used in the back-analysis of full-scale pile tests using the empirical one-dimensional t-z spring method to represent the axial soil resistance. A damage mechanics component incorporates degradation of resistance as a result of accumulated plastic displacement. The computed pile shaft-resistance damage along the pile allows estimation of its evolving capacity. The model produces evolving local cyclic and average resistances mobilized along the pile that are consistent with the observed test responses. The model is used to investigate the cyclic capacity of the same pile conditions under combinations of mean and variable loads. The implications of the test data and modeling results on the design of tension piles are discussed.

This paper reports a novel concept of ground improvement technique using cemented stone columns for improvement of the soft clay ground. The experimental investigations and analytical studies led to evolution of the technology of soft ground improvement using such stone columns almost to the fullest extent. The study included the following aspects: relative degree of effectiveness of cemented stone columns over other reported approaches, theoretical analysis for predicting the ultimate load-carrying capacity of soft clay ground treated with cemented stone columns, analysis to estimate the probable short- and long-term settlements of a given structure founded on the treated ground under the applied foundation pressure intensity, and an illustrative design problem using this technology. The failure mode and the load-transfer mechanism for this system were conceptualized on the basis of the unit-cell concept. The mathematical formulation resulted in equations enabling quantitative evaluation of the required design aspects related to the actual design of cemented stone column–soft clay systems.

A rational design method for embankments on soft soils that takes care of both bearing capacity and stability mechanisms is proposed in this paper. Two types of embankment-construction techniques are used on soft soils: a single-stage construction technique for quickly constructing relatively low-height embankments and a staged-construction technique for constructing high embankments. In this paper, a limit-equilibrium approach is used to estimate the critical slip-surface depth that is assumed to be the bottom line of a possible collapse mechanism under undrained loading conditions corresponding to single-stage construction. In contrast, to design a stage-constructed embankment on soft soils, both undrained and drained conditions must be taken into account. The proposed method can address this behavior and therefore is applicable to embankments constructed in either a single stage or multiple stages. The staged-construction technique, however, necessitates adding the tool of stress-path application to the method to consider both undrained and drained behaviors. This is a property that is not found in limit-state approaches, and it gives the proposed method the capacity to follow and interpret changing stress–strain states and their influence on mobilized strength states.

Dynamic loading is a major cause of landslides with significantly different behavior from static loading. A series of centrifuge model tests was conducted to investigate the progressive failure behavior of slopes under dynamic loading conditions, with a focus on the effect of cyclic loading. The cyclic effect induced significant deformation and final failure of the slope, which was influenced by the cyclic loading patterns. The slip surface developed from the top of the slope downward to the lower part with the application of vertical loading. The failure process of the slope exhibited a significant coupling with the deformation localization under dynamic loading conditions. The deformation localization developed in the slope during loading and resulted in the formation of the slip surface. Alternately, the local failure caused stronger deformation localization near the slip surface. The deformation was mainly concentrated in a limited zone. In the zone, a reversed triangular zone under the loading plate exhibited negligible deformation localization. The cyclic loading pattern had a significant effect on the slip surface and limit-bearing capacity of the slope, which was essentially dependent on the diversity of the deformation behavior of the slopes under different cyclic loading conditions.

Very few studies have addressed the systematic study on the stress-dependent behavior of cement/lime-treated structured and reconstituted soils from the viewpoint of the constrained modulus. The influence of cement/lime on the nonlinear stress-strain behavior is investigated with relation to the constrained modulus. Based on the idealized modulus-stress mode, the modulus-based parameters are defined and discussed taking into consideration cement/lime content and curing time. This contributes to establishing quantitatively the modulus-related correlations, such as M_L-M₀, m-ω, ${\sigma }_{\mathit{vy}}^{\prime }$-${\sigma }_r^{\prime }$ and C_c-M_L on structured soils. A proposed novel concept of ΔE_oed is feasible to identify the effect of soil structure, and the variation in shape and position of stress-strain curves attributes to the soil structure developed with chemical reactions. The consolidation stress at the point of ΔE_oed = 0 can be considered as the yield stress, and this is proved to be appropriate in contrast to the result determined by a particular method. Comparison between structured and reconstituted soils indicates that the difference in the constrained modulus in the preyield regime is mainly caused by the existence of soil structure, and the difference of modulus in the postyield regime is closely related to an important difference in water content. Soil structure is proved to be completely destroyed from the viewpoint of material stiffness when the consolidation stress is higher than the yield stress.

To investigate the influence of the axial force and its distribution along the pile shaft on the response of laterally loaded piles, a generalized solution is proposed based on the transfer matrix approach, in which the transfer matrix coefficients for piles in the free, elastic, and plastic zones were analytically obtained through Laplace transformation. This proposed method can handle piles in multilayered soil deposits with any form of p-y curve. The proposed methodology is validated by comparing its predictions with the laboratory model pile test results. A good match between model prediction and the laboratory model pile test results implies that the proposed method can be used to evaluate the pile response under combined loads effectively. Moreover, the axial force distribution along the pile shaft is simplified as a constant, which equals to the vertical load applied at the pile head. Finally, the authors investigated the influence of vertical loads and the pile-embedded ratio on the ultimate lateral bearing capacity of piles under combined loads. The ultimate lateral bearing capacity will decrease significantly as the embedded ratio is reduced and also will decrease with increased vertical load at the pile head.

Slope stability analysis has traditionally been performed using a deterministic approach. However, it has strongly been debated that the use of only the factor of safety in slope stability analysis does not explicitly account for all the uncertainties in soil parameters. Therefore, to investigate the effect of uncertainties in the stability of a fill slope, this paper uses the finite-element limit analysis methods to conduct a simple probabilistic-based analysis of two-layered undrained clay slopes. The results obtained show that although the recently developed stability charts for two-layered undrained clay may be convenient tools, the factors of safety calculated from the charts may not be sufficient. Particularly, it is shown that the probability of failure of this type of slope can be highly influenced by uncertainties in the soil properties. Various probabilistic charts based on different coefficients of variation in soil properties have been proposed in this paper. Based on this study, the recently developed charts for two-layered undrained clay slopes can be better applied. Additionally, this study also demonstrates the applicability of the finite-element upper and lower bound limit analysis methods to conduct probabilistic slope stability analyses.

A soil–rock mixture consists of soil and broken rocks mixed proportionally, and it has more complicated seepage characteristics than pure soil or broken rocks. Current research on the seepage characteristics of soil–rock mixtures is still limited to empirical equations. This paper presents an improved theoretical compound seepage model for the permeability coefficient of soil–rock mixtures and relative concentrations of individual constituents (i.e., pure soil and pure broken rocks). This model considers the gap filled by particle grains in a soil–rock mixture under the same compaction level, as well as the reduced porosity of the mixture and the decreased permeability caused by the fine particles serving as a filler. A revised version of the seepage model of soil–rock mixtures consisting of a combination of the proposed model and the Kozeny-Carman seepage model is also suggested, and the precision of the model is verified against experimental results. The model provides a straightforward and effective quantification for the permeability coefficient of soil–rock mixtures.

The present paper deals with the proposition of an original analytical permeability model of compacted soils. The model involves the microstructure of the material through the porosity and the pore-size distribution obtained by mercury intrusion porosimetry (MIP), as well as the degree of compaction. Their effects on the morphological parameters of the porous network (tortuosity and interconnection of pore network) are studied. The model was developed and tested on various types of soil: a loamy sand, a gravelous sand, a clay, and an alterite. Samples were compacted with various degrees of compaction: 85, 95, 100, and 105% of the optimum dry density as determined by a standard compaction method. The experimental results obtained for both loamy and gravelous sands and for clay were well reproduced by the model, except for the alterite. Such results might be explained by the high brittleness of the alterite, leading to a crumbling phenomenon rather than to its densification during the compaction process.

This paper presents a numerical investigation into the macromechanical and micromechanical behavior of unsaturated structural loess under constant stress increment ratio compression using the distinct-element method (DEM). Loess is typically a kind of clayey silt with a metastable structure composed of interparticle bonding and large voids. To represent the interparticle bonding, a published bond contact model proposed by the authors was used. Three samples with different water contents were tested under six incremental stress ratios to study the influence of water content and stress increment ratio on the mechanical response of loess. The DEM results qualitatively agree with available experimental observations in the literature. The observed macromechanical properties can be explained by the velocity field, force-chain distribution, bond breakage events, and contact orientation at the microscopic scale. When the stress increment ratio is low, the force chains exhibit a column-like pattern. As the stress ratio increases, bond breakage occurs uniformly and the force chains form a web-shaped pattern. As the water content increases, the force chains become denser and the number of broken bonds becomes larger. The compression and shear behavior of the specimens is related to the bond breakage at the microscopic scale. The main failure types are tension failures in the stress range studied irrespective of the stress increment ratio. The proportion of tensile failures increases with an increase in water content. In addition, the contact orientations and deviator fabrics of contacts are also studied.

A two-dimensional elastoplastic analytical model of stress–strain analysis of the pipeline under strike-slip faults was established based on the plastic flow theory and elastoplastic-beam theory, considering the effects of service loads and the nonlinearities of the pipeline and the pipe–soil interaction. According to the plastic flow theory, the axial, radial, and hoop plastic strains were derived, and the stress–strain relation was established under the plane stress state. The axial stress and strain of the pipeline induced by fault displacements were analyzed based on the elastoplastic beam theory. Additionally, the axial stresses, hoop stresses, and strains induced by the service loads (i.e., the internal pressure and the temperature variation) were also considered. The proposed model was verified by comparing the analytical results and the finite-element results. The results show that the proposed model can predict accurately the mechanical response of the pipeline subjected to strike-slip fault displacements, with minor deviations of within 10%.

This paper presents the results from a series of physical modeling and three-dimensional finite-element (FE) analyses in which the authors examined the uniaxial vertical capacity of suction caissons for offshore wind turbines. The experiments were carried out in quartz sand and involved monotonic application of vertical load. It was found that the drained capacity of suction caissons is dependent on embedment ratio. In contrast, predictions from conventional semiempirical depth factors were found to somewhat underestimate when applied to rough foundations. On the basis of the tests and FE analyses, new expressions for the depth factor of shallow foundations were validated for embedment ratios (aspect ratios) up to unity, calibrating the fitting parameters by using data from a range of soil profiles.

To design passive piles, it is critical to incorporate the impact of lateral soil movement (w_s) and its profiles. This may be conveniently realized by using appropriate input parameters and a three-layer analytical model developed by the first author. In this paper, 25 (1-g) model tests were conducted on single piles in sand, subjected to a uniform (U), inverse triangular (T), or arc (A) profile of sand movement, to a final sliding depth (l_m) of either 0.29l (l = pile embedment) or 0.57l, respectively. The measured response is subsequently simulated using the three-layer model to gain the input parameters and pile–soil interaction mechanism. The main conclusions (for l_m = 0.29l) are as follows. First, the limiting resistance per unit length (at pile-tip level; p_b) increases from uniform to inverse triangular and further to arc movement profiles at an increasing magnitude of w_s. These profiles may be superimposed together to mimic evolving soil movement profiles. Second, the p_b attains 30−60% of that on laterally loaded piles and increases by 22−60% owing to the vertical load. The p_b is proportional to the ratio of pile-head displacement (w_g) over the soil movement. The on-pile pressure is larger on smaller diameter piles. Third, the thrust taken by piles reduces by 50% from a uniform to a linearly increasing modulus. The bending moment (at approximately 4 times larger movement) is 2.3−7.5 times larger under the T movement (worst-case scenario) than those induced under U movement. Bending capacity based on w_g = 10 mm needs to be approximately tripled to warrant the safety of passive piles at a rotation angle of 5°. Finally, moving soil imposes rotational restraining and reduces bending moment and pile displacement. It may push the piles to a w_g of 1.5d (d = pile diameter) without failure. For the deep sliding of 0.57l, the p_b values may be doubled, and the arc profile may inflict the worst-case scenario as well. The results from model tests work well for three instrumented piles in glacial tills and clay, respectively, independent of evolving soil movement profiles.

Deep excavation clearly impacts existing underlying subway tunnels and threatens their operational safety. The prediction of existing tunnel deformation induced by nearby excavation is a major concern for urban construction. This research presents an analytical calculation method for predicting tunnel deformation induced by upside excavation and also discusses the role of dewatering in the deformation mechanism. First, the existing tunnel is assumed to be nonexistent in the soil, and the vertical unloading stress at the location of the existing tunnel caused by the upper excavation and associated dewatering is calculated. Second, the existing underlying tunnel is simplified as an elastic beam on a Pasternak foundation to calculate its vertical deformation. The proposed method was verified by the good agreement found between the predictions and the field measurements of the construction in the Shenzhen Chegongmiao hub project. Such an analytical method can provide fast and accurate evaluation results for similar engineering projects.

Soil clogging occurred in the early construction stage of Nanchang Metro Line 1, China, when an earth balance pressure (EPB) shield passed through the argillaceous siltstone with a total content of clay minerals of 40.5%. This paper presents a new scheme for determining soil conditioning parameters to avoid the occurrence of soil clogging, taking both the slump value and the liquidity index of the soil into account. As indicated by the results of slump tests, the optimum slump value for the soil conditioning ranged from 17 to 20 cm, and a fitting function of slump value against foam injection ratio (FIR) and water content (w) was obtained. Considering the requirement for a liquidity index equal to ∼0.4–0.75 to ensure the stability of the excavation face, a chart was proposed for the soil conditioning. Based on the chart, the optimum FIR and w for the soil conditioning for tunneling in argillaceous siltstone were determined to be ∼19.2–23.2% and ∼14.2–36.6%, respectively. Taking shield parameters, tunnel design, and soil parameters into account, the soil conditioning techniques were implemented in field conditions. The field implementation indicated that soil clogging was effectively avoided in the later section of Nanchang Metro Line 1, and the thrust and torque of the shield were reduced significantly.

This paper presents model tests on the displacement characteristics of flat-ended piles, pipe piles, and piles with a cone tip of 45º in a transparent granular soil. The testing setup consisted of transparent granular soil, a penetration device, a linear laser, a charge-coupled device (CCD) camera, and an optical platform. The transparent granular soil used in this study were made of fused quartz and a pore fluid with the same reflective index. The pore fluid was a mixture of n-paraffin C₁₂ and food-grade White Oil 15. The positions of the granular-soil particles were registered by laser speckles, and the internal displacements of the particles during penetration were recorded by the CCD camera. The horizontal and vertical displacements of the granular soil and the zone of influence of penetration of different piles were determined by comparing the images of the granular soils before and after penetrations. On the basis of these data, the displacement field and sheltering effect of different piles are discussed, and the responses of the granular soil to the penetration of different types of piles are evaluated and compared.

An empirical nondimensional reduction factor is proposed in this paper, which can be used for determining the ultimate bearing capacity of eccentrically and/or obliquely loaded shallow strip foundations embedded in sand. This was developed after analyzing a series of laboratory model test results available in the literature on strip foundations. The depth of foundation (D), eccentricity of load (e), and inclination of load (α)varied from 0 to 1B, 0 to 0.5B, and 0 to 20°, respectively, where B refers to the width of foundation. The proposed reduction factor was found to provide reasonably well results for calculating the ultimate bearing capacity of shallow strip foundation under eccentric and/or inclined loads. The results obtained by using the proposed method have been found to be in good agreement with widely used traditional methods available in the literature.

This study investigated the effectiveness of estimating the maximum allowable annular pressure during horizontal directional drilling (HDD) in noncohesive soils by using two theoretical cavity-expansion solutions (small strain and large strain); several laboratory and field tests were used for verification. The results indicated that the measurements were overestimated when using the plastic radius equal to two-thirds of the overburden depth and underestimated when using the hoop strain equal to 2%. Although the calculated limit pressures based on the two cavity-expansion solutions were higher than the measured failure pressures, a high correlation (R² = 0.96) existed between the measurements and the calculations based on the large-strain cavity-expansion solution. This correlation may be used to estimate the maximum allowable annular pressure during HDD in noncohesive soils, provided that further validation is conducted with more field data for deep depths rather than shallow depths.

In this study, bonded-particle models were used to simulate the intact rock behavior of Westerly granite. Two three-dimensional models were tested: the enhanced bonded-particle model (EBPM) and the flat-joint contact model (FJCM). To calibrate both models, a new standard procedure was developed. The results of the simulations of the models were compared with the intact rock behavior determined by laboratory tests, including elastic parameters, peak and threshold envelopes, and postpeak behavior. From this study, it can be concluded that the FJCM better represents the mechanical behavior of Westerly granite rock, especially at the low to intermediate confining pressures. Based on this work, it is recommended to use the FJCM to model the behavior of brittle rock at the low to intermediate confining pressures.

A finite-element analysis program was developed to analyze the progressive failure of a soil slope based on the progressive failure mechanism. The shear element was developed to comprehensively characterize the shear zone, which describes the soil shear deformation localization of progressive failure. The formulation of the shear element was derived using a format similar to that of the soil-structure interface element and incorporated into the traditional FEM to macroscopically simulate the deformation and failure processes of the slope. A series of algorithms, including the combination and transformation of the shear element and the entity element, were proposed to obtain the full solutions for the progressive failure of soil slopes. The simulation results using the developed program for the slip surface, failure process, and displacement distribution match the observations of centrifuge model tests well.