Column-supported embankments (CSEs) are constructed over soft soils to overcome common problems associated with such soils, namely: potential bearing capacity failures, intolerable total and differential settlements, large lateral pressures and movement, and instability of slopes. CSEs are often more economical than conventional foundation techniques and can be used to accelerate construction by considerably reducing the consolidation time. Highway embankments, highway widening, and bridge approach fills are some typical applications of CSEs. A wide range of column types may be used in CSEs, which is another advantage of this ground improvement technique. Driven piles, granular columns (also referred to as stone columns), deep-mixing method columns, vibro-concrete columns, geotextile encased columns, or other types of suitable columns can all be used in CSEs. Given the high density of piles needed in CSEs, as well as the large areas that often are covered by an embankment, flexible columns are typically more economically attractive than stiffer elements such as driven piles. Among flexible columns, the use of granular columns (GCs) in soft soil has become a major ground improvement technique during the last two decades. Granular columns subjected to compressive loads experience failure modes such as bulging (lateral displacement), general shear failure, and sliding. However, the most common failure mode for GCs in very soft soils is bulging due to the lack of required lateral confining pressure. In these situations, to provide the required lateral confining pressure and to increase the bearing capacity, GCs can be encased by a suitable geosynthetic to form a geosynthetic-encased column (GEC). Three-dimensional finite element analyses were carried out to simulate the behavior of a single GC with and without geosynthetic encasement in a soft clay soil using the commercial computer program ABAQUS. A comprehensive study was performed to better understand the mechanism of load transfer in both GCs and GECs. Numerical results confirmed that using a high-strength geosynthetic for confinement not only increases the strength of a GC, but also significantly reduces the lateral displacement of the column into the very soft surrounding soil. A series of extensive parametric analyses was also performed in order to investigate the influence of different parameters on the behavior of a GEC. As it is the volumetric response of the encased material that mobilizes tensile stresses in the encasement and thus allows for superior stress-displacement column performance, it was hypothesized that accurately capturing the shear-induced volume change that occurs in the encased granular material may be quite important in the numerical simulation of GECs. As a result, finite element analyses were performed using models possessing various levels of sophistication in order to examine the sensitivity of the results to the constitutive model that is used to simulate the behavior of the encased granular soil. Having completed a detailed numerical study of an isolated GEC, finite element analyses were performed to study the behavior of GECs when used as deep foundation elements in CSEs (with or without an overlying geosynthetic reinforcement layer). The validity of the commonly used unit cell concept in the numerical modeling of CSEs with geosynthetic reinforcement (i.e., geosynthetic-reinforced column-supported embankments or GRCSEs) was examined using full 3-d, 3-d unit cell, and axisymmetric unit cell analyses. 3-d unit cell analysis was selected to investigate the importance of geosynthetic encasement and the influence of both the granular column material and the soft soil constitutive models on the numerical simulation of CSEs with GECs. Finally, parametric studies were performed to determine the influence of various input parameters in the design of CSEs/GRCSEs when GECs were used as deep foundation elements.