Variability of the GSI Index Estimated From Different Quantitative Methods

Abstract

The main purpose of this paper is to present the results of a comparative analysis of the GSI values predicted by different empirical equations currently available in literature which apply the input parameters used in the best known rock mass classification systems, namely the RMR1989, the Q and the RMi. For this aim, probabilistic analyses based on Monte Carlo simulations have been developed using, as inputs, the geomechanical field data collected for a real rock mass. Afterwards, the GSI outputs obtained from the different empirical approaches have been statistically analyzed and compared. The results of simulations indicate that the diverse relationships may predict dissimilar values of the GSI for the same rock mass. In general, the highest values have been obtained from the equations which apply the RMR1989 input ratings, while the methods based on RMi produced the lowest results. Sensitivity analyses performed on the simulation outputs show that, for the examined case study, the input parameters reflecting the degree of jointing of the rock mass, namely the RQD and the rock block volume, have the largest effects on the calculated GSI, while those describing the characteristics of discontinuities show lesser influence and may depend on the equation adopted. In particular, the GSI estimated from the methods based on the Q and RMi inputs is especially sensitive to the infilling and roughness of discontinuities, while in the methods based on the RMR1989 ratings the discontinuity characteristics influencing the GSI value the most demonstrate to be the aperture, the roughness and the length. The analyses have mainly highlighted the possible uncertainties still related to the quantitative estimation of the GSI and allowed for the recognition of the input rock mass parameters that may have the highest impact on the GSI in the different estimation methods.