Reliable assessment of the Geological Strength Index (GSI) is essential for estimating the strength and deformation of rock masses. This study proposes a more objective and data-driven model for GSI estimation. To achieve this, a method that combines the Hoek-Brown (HB) failure criterion and the 3DEC-based synthetic rock mass (SRM), namely the HB-SRM method, is proposed to quantify the effects of rock mass structures and joint surface condition parameters on GSI values. Based on the analysis of the numerical results, a quantitative GSI model in terms of the volumetric joint count ( J v ) and the joint friction angle ( φ ) is proposed and validated. It is found that the values of GSI increase non-linearly with the decrease of J v and increase of φ ; therefore, GSI contours are nonlinear and exhibit variable spacing. In addition, the reliability of the proposed GSI model is further tested through back-analysis of in-situ rock mass deformation tests conducted in exploration tunnels at the dam site. Results show that the proposed model provides a conservative estimate of GSI, which is advantageous for ensuring safe and reliable engineering design. Finally, this study primarily focuses on blocky rock masses; however, the proposed HB-SRM method together with the use of J v and φ can also be used for developing quantitative GSI charts for various rock types.
Shen et al. (Fri,) studied this question.