The Geological Strength Index (GSI) is a classification system based on assessing the lithology, structure, and condition of discontinuity surfaces in the rock mass (Marinos et al., 2005). The main application of the GSI classification is estimating rock mass properties by reducing the material constants σci and mi.
Hoek and Brown (1997), Marinos and Hoek (2000), Marinos et al. (2005), and Hoek et al. (2013) provided guidelines for the selection of the GSI for various geological conditions. Estimating the GSI values from direct visual observations of the rock conditions in the field should always take the highest precedence.
However, in most practical mining or civil conditions, direct exposure mapping is either unavailable or limited to isolated locations throughout the rock mass. Information from geotechnical drill holes is often the primary source of data available for rock mass classification and establishing a GSI value.
The method proposed by Cai et al. (2004) should preferably be used to derive the GSI value from the geotechnical drill hole data. In cases where RQD is the only data available for the representation of joint spacing, the method presented by Hoek et al. (2013) can be used to quantify the GSI. This method uses the ratings applied in the most widely accepted rock mass classification systems, Rock Mass Rating (RMR) and Tunnelling Quality Index (Q), developed by Bieniawski (1976) and Barton et al. (1974). This technique has been used during the majority of recent validation studies carried out for the IUCM.
In cases where the large-scale rock mass stability is under study, e.g. mine scale assessments, large open pit stability assessments, caving assessments, etc., if the variability in the GSI values throughout the rock mass is not sufficiently represented in the models (for example,, through a geotechnical block model), the GSI value will usually need to be downgraded towards the lower percentile values. This is to account for the effect of rock mass variability.
As a rule of thumb, the median GSI value is recommended for underground models, while for open pit models, a 35th percentile GSI value is recommended.
References:
