This study uses dynamic centrifuge tests and three-dimensional (3D), nonlinear finite-difference analyses to: (1) evaluate the effect of soil-cement grid reinforcement on the seismic response of a deep soft soil profile, and (2) to examine the dynamic response of structures supported by shallow foundations on soft clay reinforced by soil-cement grids. The soil profile consisted of a 23-m-thick layer of lightly over-consolidated clay, underlain and overlain by thin layers of dense sand. Centrifuge models had two separate zones for a total of four different configurations: a zone without reinforcement, a zone with a "embedded" soil-cement grid which penetrated the lower dense sand layer and had a unit cell area replacement ratio Ar = 24%, a zone with an embedded grid with Ar = 33%, and a zone with a "floating" grid in the upper half of the clay layer with Ar = 33%. Models were subjected to a series of shaking events with peak base accelerations ranging from 0.005 to 0.54g. The results of centrifuge tests indicated that the soil-cement grid significantly stiffened the site compared to the site with no reinforcement, resulting in stronger accelerations at the ground surface for the input motions used in this study. The response of soil-cement grid reinforced soft soil depends on the area replacement ratio, depth of improvement and ground motion characteristics. The recorded responses of the structures and reinforced soil profiles were used to define the dynamic moment-rotation-settlement responses of the shallow foundations across the range of imposed shaking intensities. The results from centrifuge tests indicated that the soil-cement grids were effective at controlling foundation settlements for most cases; onset of more significant foundation settlements did develop for the weakest soil-cement grid configuration under the stronger shaking intensities which produced a rocking response of the structure and caused extensive crushing of the soil-cement near the edges of the shallow foundations. Results from dynamic centrifuge tests and numerical simulations were used to develop alternative analysis methods for predicting the demands imposed on the soil-cement grids by the inertial loads from the overlying structures and the kinematic loading from the soil profile's dynamic response. / Ph. D. / This study uses dynamic centrifuge tests and three-dimensional (3D), nonlinear finitedifference analyses to: (1) evaluate the effect of soil-cement grid reinforcement on the seismic response of a deep soft soil profile, and (2) to examine the dynamic response of structures supported by shallow foundations on soft clay reinforced by soil-cement grids. The results of centrifuge tests indicated that the soil-cement grid significantly stiffened the site compared to the site with no reinforcement, resulting in stronger accelerations at the ground surface for the input motions used in this study. It was also found from the results of the dynamic centrifuge tests that the soil-cement grids were effective at controlling foundation settlements for most cases; onset of more significant foundation settlements did develop for the weakest soil-cement grid configuration under the stronger shaking intensities which produced a rocking response of the structure and caused extensive crushing of the soil-cement near the edges of the shallow foundations. Results from dynamic centrifuge tests and numerical simulations were used to develop alternative analysis methods for predicting the demands imposed on the soil-cement grids by the inertial loads from the overlying structures and the kinematic loading from the soil profile's dynamic response.