Abstract
Shear strength of rock masses is critical for geotechnical tasks including tunnel, slope stability, and foundations. Various approaches comprising of back calculation, large-scale testing, experimental parameter, and a rock mass categorization system could be used to define the shear strength of a jointed rock mass. Furthermore, the excavatability features of rocks influence the choice of excavation technique and equipment in geotechnical and mining operations. In this work, the excavatability of rock is assessed using rock mass categorization techniques, such as residual shear strength, rock mass rating system, and updated excavatability graph. Therefore, the residual and peak shear strength envelopes of jointed magmatic rock masses are evaluated through a hybrid ELM-SVM model that combines two soft computing models: support vector machines (SVM) and extreme learning machine (ELM) (objective of study). The shear strength characteristics are studied and predicted using the findings of regression analysis using factors, such as root-mean-square error (RMSE) and coefficient of determination (R-2). The findings reveal that the suggested hybrid could be applied to find the peak and residual shear strength values of frequently fractured rocks as well as providing a meaningful relation between the values of shear strength variables of the geotechnical units.