A master's thesis at the College of Engineering, University of Basra, discusses NUMERICAL MODELING FOR GROUND IMPROVEMENT OF KHOR AL ZUBAIR IMMERSED TUNNEL

Researcher Ahmed Dawood Aziz's master's thesis was discussed at the College of Engineering, Department of Civil Engineering, University of Basra, entitled NUMERICAL MODELING FOR GROUND IMPROVEMENT OF KHOR AL ZUBAIR IMMERSED TUNNEL

‏This research numerically investigates the performance of ground improvement techniques for the Khor Al-Zubair immersed tunnel project, a critical infrastructure link between Al-Faw Grand Port and Umm Qasr in southern Iraq. The tunnel alignment crosses soft to very soft clay layers, presenting major geotechnical challenges related to settlement control and load transfer.
‏Deep Soil Mixing (DSM) was adopted as the primary stabilization method to improve the foundation soils beneath the tunnel approaches. A comprehensive parametric study was conducted using PLAXIS 3D finite element software to evaluate the effects of DSM column diameter, length, and configuration (single, twin, quadruple), as well as the area replacement ratio (as%) on immediate vertical settlement and stress redistribution beneath the tunnel invert. The study also included a comparative analysis with an alternative ground improvement technique using stone columns under identical geotechnical and structural loading conditions. Furthermore, the research incorporated a critical comparison between two-dimensional (2D) and three-dimensional (3D) modeling ,with the aim of highlighting the fundamental differences between them in the modeling of ground improvement.
The results demonstrated that increasing the length of DSM columns, especially when penetrating stronger bearing layers, leads to a significant reduction in both total and differential settlement, due to the activation of end-bearing resistance and side friction. While increasing column diameter under a constant value of as% improved the overall system stiffness and load transfer capabilities, it simultaneously reduced the number of columns per unit area, potentially leading to untreated zones and stress concentration. Notably, twin and quad column configurations provided better lateral confinement and stress overlap, making them more effective under asymmetric or eccentric loading conditions, despite their higher implementation complexity.
Additionally, the study showed that variations in as% value had a nonlinear effect on settlement reduction; technical benefits became marginal beyond 20%, indicating a performance-cost tradeoff. The optimal as% range was found to be between 15% and 20%, achieving a practical balance between geotechnical performance and construction economy. From the perspective of comparing techniques, DSM technology demonstrated a clear superiority over(stone columns) in terms of stress distribution homogeneity and resistance to vertical deformations, especially in highly compressible soil.
Regarding numerical modeling, the comparison revealed that simulations using two-dimensional (2D) models consistently underestimate settlement and cannot accurately represent the complex three-dimensional (3D) interactive effects between adjacent columns and the soil, especially in the critical tunnel edge regions. In contrast, three-dimensional models provided a more accurate representation of the resulting deformations and stress distributions, thereby confirming the crucial importance of using 3D analysis in numerical modeling.
From an engineering design perspective, this study offers a detailed framework for optimizing DSM column parameters based on actual soil behavior and stratigraphy. It emphasizes the importance of explicitly modeling the soil-column interfaces as heterogeneous rather than employing homogenized equivalent soil properties, which often result in unrealistic underestimation of peak stresses and deformations. The findings contribute to improving the predictive accuracy of numerical models and provide a valuable reference for engineers and practitioners involved in similar projects through soft ground conditions.