Some normally consolidated soft soils manifest strength sensitivity, ie these soil manifest strain softening when shear in an undrained mode. These soils, referred to as sensitive soft soils, have the typical features of strain hardening in drained shearing and strain softening in undrained shearing. The consolidation lines of these soils are also curved (concave upwards) in the semi-log space. However, under high consolidation stress or upon large shearing, these soils re-gain the features of re-constituted soil. Ground improvement methods like stone columns were reported as not effective when installed in the sensitive soft clays. But mechanism of the un-effectiveness of the stone columns remains unknown because of lack of a suitable and simple model for simulating the stress-strain behaviours of sensitive soft soils. Although these soils have a meta-stable micro-structure, models that developed for simulating structured firm soils are not suitable for simulating sensitive soft soil features. Thus, a new model was formulated. The new model can degenerate back to a Modified Cam Clay model. The ability of new model in simulating a range of behaviour was verified by using the finite difference (FD) method in solving the partial differential equations of the soil model for a range of tri-axial test conditions. The model was further implemented in coupled analysis formulation and coded into FEM program AFENA. Various cases with different soil parameters were then simulated and compared with the FD solutions for various triaxial tests so as to check the stability of the FEM code. The coupled FEA was then used to simulate the performance of geosynthetic-encased stone columns. A new stone column element and a geo-encasement element were developed and coded into AFENA. The stone column simulations were then done for both non-sensitive soils (represented by Modified Cam Clay model) and sensitive soft soil (represented by the new model). Parametric study was conducted to examine the performance of the geo-encased stone columns in both types of soils. Furthermore, two different installation methods: wished-in installation and full displacement installation were studied numerically. Cross comparison was done to investigate how the sensitive soft soil features interact with the installation method in affecting the performance of the geo-encased stone columns. A range of factors that influence the geosynthetic-encased stone columns performance installed in soft soils were also made clear.

Braunkohle ist aktuell immer noch ein wichtiger Rohstoff für die Energieversorgung in Deutschland. Ein großes Abbaugebiet befindet sich in Nordrhein-Westfalen. Im sogenannten rheinischen Braunkohlenrevier liegen mächtige Braunkohleflöze, welche seit einigen Jahrzehnten im Tagebaubetrieb abgebaut werden. Allerdings fallen auch große Mengen an Abraum (die Deckschichten über der Braunkohle) an, welche meist auf der, dem Abbauort, gegenüberliegenden Seite, wieder verkippt werden. Dort bilden die verkippten Massen eine Böschung, die im Fall des Tagebaus Hambach ein Böschungssystem mit über 400m Höhe erreicht hat. Diese sich ständig verändernden Böschungen müssen, wie alle hohen Böschungen, durch Fachpersonal auf ihre Standsicherheit überprüft werden. Für diese Standsicherheitsberechnungen bedarf es Kennwerte, u. a. die undrainierte Scherfestigkeit für die Nachweise der kurzfristigen Standsicherheit, für die vorliegenden verkippten Böden. Diese liegen nach der Verkippung meist heterogen vor und verändern durch eine fortschreitende Konsolidation ihren Zustand. Diese Arbeit beschäftigt sich mit der Bestimmung des Kennwertes der undrainierten Scherfestigkeit in fein- und gemischtkörnigen Böden. Dazu werden zunächst die Klassifikations- und Zustandseigenschaften während der Verkippung der Böden für die in den Tagebauen Inden und Hambach verwendeten Materialklassen vorgestellt. Dabei wird auch auf die Besonderheiten der Böden im Tagebau, wie z.B. die Zustandsänderung während des Transportes oder die vorliegende Heterogenität, eingegangen. Es werden analytische und numerische Prognosemodelle zur Bestimmung der undrainierten Scherfestigkeit vorgestellt. Die Grundlage dieser Modelle bilden die Konsolidationstheorien, mit welchen sich die Zustandsvariablen, wie die Porenzahl oder die effektive Spannungen, berechnen lassen. Über verfügbare Stoffmodelle, wie z.B. das hypoplastische Modell nach Mašín können anhand der Zustandsvariablen die undrainierten Scherfestigkeiten abgeleitet werden. Mit den in dieser Arbeit verwendeten Modellen lassen sich undrainierte Scherfestigkeitskennwerte zu verschiedenen Zeitpunkten prognostizieren, mit denen wiederum z.B. Standsicherheitsberechnungen von Böschungen durchgeführt werden können. Die statistischen Auswertungen der Klassifikationseigenschaften und Zustände erlauben eine Beurteilung von im Tagebaubetrieb häufig verwendeten Kennwerten (z.B. Fließgrenze oder Wassergehalt) und deren Streubreiten innerhalb der Materialklassen. / Lignite was and still is an essential source for producing energy in Germany. One huge extraction site is located in North-Rhine-Westphalia (in between Mönchengladbach, Cologne and Aachen). The so called Rhenish lignite district contains very thick lignite deposits, which are mined in open pit mines since a few decades. There are huge overburdens (on top on the lignite) which are transported and dumped on the opposite to the excavation side. The dumped soil masses create slope systems, that can amount to 400m of height (in open pit mine Hambach). These large slopes undergo permanent changes and have to be assessed considering their safety and stability by specialists. Characteristic properties regarding shear strength of the heterogeneous fine- and mixed grained soils are required for such calculations. The main challenge is the description of the state of the soil, as it is permanently changing due to consolidation. One essential characteristic is the undrained shear strength. The main focus of this thesis is the determination of the undrained shear strength. Therefore, the soil properties are determined with classification tests and the state of the soils at the time of dumping is assessed. The results are categorised using the material classes of the open pit mines Inden and Hambach. The particularities of dumped soils in open pit mines are shown, for instance the change of state during transport or the heterogeneity. Moreover, analytical and numerical models are employed to predict the undrained shear strength. The basis of all these models are consolidation theories, which are applied to determine the state variables such as void ratio and effective stresses. The undrained shear strength can then be predicted by using constitutive models such as the hypoplastic model for clays according Mašín. Using the approach presented in this thesis the undrained shear strength can be predicted and employed for example for slope stability analyses. The statistic evaluation of the soil properties allows a review of the soil characteristics that are typically employed in open pit mining. Furthermore, the scattering of properties of the material classes in the open pit mines can be assessed.

Offshore developments for hydrocarbon resources have now progressed to water depths approaching 2500 m. Due to the difficulties and high cost in recovering high quality samples from deepwater site, there is increasing reliance on in situ tests such as piezocone and full-flow (i.e. T-bar and ball) penetration tests for determining the geotechnical design parameters. This research was undertaken in collaboration with the Norwegian Geotechnical Institute (NGI), as part of a joint industry project, to improve the reliability of in situ tests in determining design parameters and to improve offshore site investigation practice in deepwater soft sediments. In this research, a worldwide high quality database was assembled and used to correlate intact and remoulded shear strengths (measured from laboratory and vane shear tests) with penetration resistances measured by piezocone, T-bar and ball penetrometers. The overall statistics showed similar and low levels of variability of resistance factors for intact shear strength (N-factors) for all three types of penetrometer. In the correlation between the remoulded penetration resistance and remoulded shear strength, the resistance factors for remoulded shear strength (Nrem-factors) were found higher than the N-factors. As a result, the resistance sensitivity is less than the strength sensitivity. The correlations between the derived N-factors and specific soil characteristics indicated that the piezocone N-factors are more influenced by rigidity index than those for the T-bar and ball penetrometers. The effect of strength anisotropy is only apparent in respect of N-factors for the T-bar and ball penetrometers correlated to shear strengths measured in triaxial compression. On the other hand, the Nrem-factors showed slight tendency to increase with increasing strength sensitivity but were insensitive to soil index properties. These findings suggest that the full-flow penetrometers may be used to estimate remoulded shear strength and are potentially prove more reliable than the piezocone in estimating average or vane shear strength for intact soil but the reverse is probably true for the estimation of triaxial compression strength.

The concept of reinforcement was developed in late 20th century and since then till the recent past there are many works carried out on the effect of fibers in imparting strength and stiffness to the soil. Experimental investigations on fiber reinforced soils showed an increase in shear strength and reduction in post peak loss of strength due to the reinforcement. Analytical/mechanistic models are developed to predict the stress-strain response of fiber reinforced soil (under discrete framework, energy dissipation methods, force equilibrium methods etc). Numerical investigations are also carried out, and it was observed that the presence of random reinforcing material in soils make the stress concentration diffuse more and restrict the shear band formation. Soil properties vary from point to point at micro level and influence stress mobilization. Hence, there is a need to carry out probabilistic analysis to capture the effects of uncertainties and variability in soil and their influence on stress-strain evolution. In the present thesis an attempt has been made to propose a mechanistic model that predicts the stress-strain response of fiber reinforced soil and also considers the effect of anisotropy of fibers. A stochastic/probabilistic model is developed that predicts the stochastic stress-strain response of fiber reinforced soil. In addition, probabilistic analysis is carried out to observe the effect of number of fibers across the shear plane in imparting shear resistance to soil. The mechanistic model and stochastic models are validated with reference to the experimental results of consolidated undrained (CU) triaxial tests on coir fiber reinforced red soil for different fiber contents. The entire thesis is divided into six chapters. Chapter-wise description is given below. Chapter one presents a general introduction to the works carried out on fiber reinforced soils and also the investigations carried out on probabilistic methodologies that takes into account the soil variability. Thus, the chapter gives an outline of the models developed under mechanistic and probabilistic frameworks in the thesis. The objectives and organization of the thesis are also presented. Chapter two presents a detailed review of literature on the role of fibers in fiber reinforced soil. The details of experimental investigations carried out and models developed are explained briefly. Also, the literature pertaining to the role of variability in soil on its engineering behavior is presented. Based on the literature presented in this chapter, concluding remarks are made. Chapter three presents the details of a new mechanistic model developed based on modified Cam-clay model. This model considers the effect of fiber content and also the effect of anisotropy due to fibers. The predictions from the mechanistic model are compared with the experimental results. Under anisotropic condition, as angles of inclination of fiber vary from 0° to 90° with the bedding plane, it is observed that the strength increment in the reinforced soil is not as significant as that observed in isotropic case. Horizontal fibers turn out to be most effective since they are subjected to maximum extension thereby inducing tensile resistance which in turn contributes for strength increase in fiber reinforced soil. Chapter four presents a new approach to predict the stochastic stress-strain response of soil. Non-homogeneous Markov chain (multi-level homogeneous Markov chain) modeling is used in the prediction of stochastic response of soil. The statistical variations in the basic variables are taken into account by considering the response quantities (viz. stress at a given strain or settlement at a given load level) as random. A bi-level Homogeneous Markov chain predicts the stochastic stress-strain response efficiently. The predicted results are in good agreement with the experimental results. An illustration of this model is done to predict the stochastic load-settlement response of cohesionless soil. A simple tri-level homogeneous Markov chain model is proposed to predict settlements of soil at a given load for an isolated square footing subjected to axial compression. A parametric study on the effect of correlation coefficient on the prediction of settlements is performed. Chapter five presents the results of probabilistic analysis carried out to determine the effect of number of fibers across the shear plane in improving the shear strength of soil. It is observed that as the percentage of fibers in the specimen increases, the probability of failure of specimen under the same stress condition reduces and thus the reliability of the fiber reinforced soil system increases. In Chapter six, a summary of the important conclusions from the various studies reported in the dissertation are presented.