Physical modeling has played an important role in studies related to excavation of tunnels in soft ground. A variety of modeling techniques have been developed by researchers all over the world to study ground response to tunneling. These techniques range from the two-dimensional trap door tests to the miniature tunnel boring machines that simulate the process of tunnel excavation and lining installation in a centrifuge. This paper presents a review of selected physical models that have been developed and used in soft ground tunneling research. Furthermore, this paper discusses some of the various approaches used to record soil deformation and failure mechanisms induced by tunneling. Experimental setups and sample results are presented for each technique as described by original authors. A summary of the advantages and disadvantages of each method is also presented.
۱٫ Introduction

Due to the increase in urbanization found all over the world, tunneling has become a preferred construction method for transportation and underground utility systems. With so many tunnels being built, it is important to have a comprehensive understanding of the tunneling induced displacements and stresses and their impact on nearby structures. Tunneling technology has significantly advanced in the past few decades. Nevertheless, tunnel engineers are often relying on empirical methods (e.g. Schmidt, 1974; Attwell, 1978; O’ReiIIy and New, 1982; Mair et al., 1993; etc.) based on limited field data in calculating surface settlement or lining stresses. These methods assume plane strain conditions and often do not account for the three-dimensional (3D) nature of the tunnel construction process. Numerical modeling (e.g. Mair et al., 1981; Rowe and Lee, 1989; Swoboda et al., 1989; Lee and Rowe, 1990; Leca and Clough, 1992; Chen and Baldauf, 1994; etc.) allows one to conduct more realistic analyses that take into account the tunnel-lining interaction, construction sequence and 3D face effects. Analysis of instrumented projects and field trials (e.g. Peck, 1969; Attwell and Farmer, 1974; Rowe and Kack, 1983; Lo et al., 1984; Harris et al., 1994; etc.) has yielded useful information. However, results are difficult to interpret. In addition, field investigation is limited by (a) expense of instruments and (b) safety concerns that prevent access to tunnels near collapse. Full-scale experiments are very expensive, difficult to run, and are hard to repeat. For all these reasons, ground response to tunneling should also be studied using reduced physical models.

Laboratory model tests conducted under gravity or in a centrifuge allow one to investigate the most relevant factors influencing the tunnel behavior. Testing results also provide valuable data for refining the chosen numerical model. Several 2D and 3D models have been proposed to investigate different aspects of tunneling in soft ground. Tunnels are usually modeled by either placing soil around and over a pre-installed tube and controlling the supporting pressure or precutting the tunnel opening and installing a lining system. Models have also been developed to study the face stability of tunnels in soft ground including the trap door method, a pre-installed tube with vinyl facing, a dissolvable polystyrene foam core, or a miniature tunnel boring machine.

This paper summarizes selected physical model experiments that have been developed and used in soft ground tunneling research. Furthermore, this paper will discuss the various approaches used to record soil deformation induced by tunneling. Testing setups are presented for each technique as described by the original authors. A summary of the advantages and disadvantages of each technique is also presented.