A Rock Mass Classification System For Hard Rock Platinum Mines Within The Great Dyke

Abstract

Zimplats mine, a hard rock platinum mine operating within the Great Dyke of Zimbabwe, uses the Tunnel Quality Index (Q system) for its rock mass classification. The Q system was directly adapted from Barton et al. (1974) due to its overall acceptability and positive global record. The system, however, fails to align the observed rock mass conditions to the calculated ratings. The calculated ratings present nearly 70 % of the mining environment as poor ground, whereas from the underground observations, 50 % of the exposed rock mass is either fair or good rock mass. The research attempts to answer the question of the Q system's adequacy as a mining rock mass classification system. The parameters influencing mining rock mass stability were understood by analysing data from appropriate rock mass classification systems and other relevant parameters. This was done through geotechnical logging and using four rock mass classification systems: the RQD, the RMR, the Q, and the Q" systems. The parameters were analysed by reviewing previous fall-of-ground (FOG) accidents and incidents, including brows, monitoring data from instruments, and numerical modelling. From the assessment, the field stress and the orientation of the planes relative to the excavation were identified as worthy parameters to include in the research. Data analysis using descriptive statistics for each parameter in the assessment was done. Regression analysis was also used to determine the relationship between the parameters and each classification outcome. The outcomes of the rock mass classification systems used in the analysis were converted to both Q and RMR for comparison purposes. A review of the support determination chart was also done after noting that the Q system is more inclined towards defining extreme conditions than fair conditions. The support determination chart by Grimstad and Barton (1993) has a wider range of Q values describing poor conditions than the scale describing fair and good conditions. This overrates the risk, resulting in support overdesign through reduced extraction and extra costs on support. The support determination chart by Grimstad and Barton (1993) was altered to suit the hard rock platinum mining environment based on FOG data. The chart's upper limit for "Q =D" was moved to 3, the upper limit for "Q =C" was moved to 5, and the upper limit for fair conditions was moved to 10. The modification presented a better analysis when using the Q system. The research justifies the need for a mine specific classification system for hard rock platinum mines in the Great Dyke.