The acquaintance relations between rock mass classification methods, RMR and Q, and their parameters are analyzed. The correlation coefficients between parameters and the matrix of acquaintance relations are derived. The acquaintance relation matrix can prescribe the proper correlations between rock parameters and give good reliability in the estimation of rock mass characteristics. This matrix of the acquaintance relations can be further extended to the form of network, and could be used to analyze the correlativity and to enhance the utility of existing rock mass classification methods.
Various methods such as RMR, Q, GSI, etc. have been used for the estimation of rock mass quality and the prediction of rock mass properties. These rock mass classification methods have different characteristics in rock parameters considered, and so they show some differences in results and the limitation in correlativity each other (Goel et al. 1995, Kwon et al. 2001, Marathe 1999, Synn et al. 2002).
For the analysis on correlations between parameters, the concept of acquaintance relations is developed. This network basically means the correlation between the rock parameters such as strength, deformability, discontinuity, groundwater, stress, etc.
The concept of acquaintance relation network is based on the ‘small-world theory’ or ‘small-world net- work’, which is a kind of model of analytical statistics recently applied to several fields that prescribes the correlativity between the elements within an organization and determines the acquaintance of any two elements (Case 2001, Watts 1998).
The Concept of small-world theory is shown in Figure 1 and its application to the correlation analysis (Figure in full paper) of rock mass classification methods and rock property parameters is shown in Figure 2. The symbols with numeric number represent rock mass classification methods such as RMR, Q, RCR, N, M-RMR, RMi, GSI, etc and the symbols with alphabetic letter represent rock properties such as strength, deformability, discontinuity, groundwater, stress, geophysical data, etc.
The present method of rock mass classification has the system similar to regular & random network, while the acquaintance relation network can prescribe the proper correlations between rock parameters and
(Figure in full paper)
For the construction of various roads and tunnels, many site investigations are performed in early stage. Rock mass classification is very important in the precise design of these rock structures, and so more than two of rock mass classification methods are often used in the same site.
The four sets of the site investigations, where RMR and Q methods are used together, are collected in this study and the relations between two methods are plotted in Figure 3. For the comparison, the relations by other researchers over the world are plotted together (Goel et al. 1996, Synn et al. 2002). According to the site and/or investigators, the results of rock mass classifications are different each other. The coefficient of correlation between two methods shows the range of 0.7~0.9.