As a mechanical property of rock materials, brittleness is a critical parameter for risk analysis of rock explosion and stability of borehole-wall in the shale gas industry. It is also vital in the design of hydraulic fracturing and many other rock engineering applications. Currently, many methods of measuring brittleness require laboratory tests to measure the mechanical parameters (strength and deformation characteristics). It has been well characterized that the mechanical properties of the rock material are strongly dependent on petrographical properties. Because, the petrographical properties are reflective of the setting of the rock material during deposition, diagenesis, and weathering. Despite the high importance of brittleness in engineering applications, there are few studies on the effect of petrographic properties of rock materials on their brittleness. In this study, to quantifying the effect of petrographic properties on brittleness, statistical techniques were used to develop inexpensive predictive models to estimate the brittleness of five different types of sandstones. Based on strength (B1, B2, B3, and B4) and deformation (B6), brittleness indices measured for these sandstones are divided into two groups. Petrographic parameters that have a significant effect on the brittleness include cement and matrix content, the grain area ratio (GAR), packing proximity (Pp) and packing density (Pd), feldspar percentage, concavo-convex, long and tangential contacts. Using Backward Multiple Linear Regression analysis (BMLR), polynomial relationships were established to investigate the combined effect of petrographical properties on the brittleness index. To estimate the resistance to stress-induced damage (fracture toughness) in the assessment of excavation stability, a semi-empirical procedure was used to develop the rock tenacity rating system. For this purpose, the Hoek–Brown brittle parameter (s) was calculated based on the specific rock type intrinsic prope
