Evaluating the predicting performance of indirect methods for estimation of rock mass deformation modulus using inductive modelling techniques.

The rock mass deformation modulus is an important parameter in numerical modeling for the stability analysis of tunnels and underground excavations. This parameter can be determined by direct and indirect methods. The direct method includes in-situ tests which are costly, timing consuming and the reliability of the result is also questionable. In indirect method different empirical models are used for estimation of rock mass deformation modulus. In this paper Rock Mass Rating (RMR), Geological Strength Index (GSI), Young Modulus of Elasticity and Uniaxial Compressive Strength (UCS) were used as input parameters in empirical models for determination of deformation modulus for rock mass. The Multi Liner Regression (MLR) and Artificial Neural Network (ANN) were used for assessment of the prediction performance of different established empirical models for estimation deformation modulus for rock mass. After analysis and comparison of results obtained from MLR and ANN, it was concluded that, the ANN based model predicting performance is better as compared to MLR model for all five data sets and the performance of both models is much better for those data sets which are collected from empirical equation containing three input variables. [ABSTRACT FROM AUTHOR]