Application and comparison of squeezing estimation methods for Himalayan tunnels

The reliable prediction and estimation of squeezing and ground response to tunneling in squeezing ground remains a challenge. Various studies have attempted to address this problem by developing a variety of methods. However, to date, none of the suggested methodologies single-handedly provides an acceptable reliability in terms of squeezing prediction and, even more so, of the ground response to tunneling. In addition, the results of many of these studies differ from each other due to the different assumptions made and parameters considered. This variability in squeezing estimation methods led us to perform a comparative study with the aim to achieve more reliable estimates of rock mass behavior to tunneling. In this study, we applied a wide variety of available squeezing prediction methods to geotechnical data obtained on 35 tunnel sections in the Himalayan region that have experienced squeezing in the past. The geotechnical data were collected in a literature survey and include Barton's Q (Rock Mass Quality), Goel's N (Rock Mass Number), tunnel size, overburden depth, observed convergence, and support pressure. We propose a method for predicting the strength and deformation parameters of rock mass using available empirical correlations with the aim of applying more elaborate semi-analytical and analytical methods of squeezing estimation. The results of our analysis show that though the application of empirical methods is simple, the results vary substantially. The results from the semi-analytical and analytical methods are generally consistent for a given set of input parameters. The relation between support pressure and convergence allows assesment of tunnel behaviour for different support types. However, the major disadvantage of these methods is the estimation of required strength and/or deformation parameters of the rock mass; these parameters are difficult to estimate and are generally not evaluated in practice.