Abstract: Uniaxial compressive strength of rocks is widely used in geotechnical engineering, as it can be directly estimated using relatively straightforward and cost-effective methods. In the present study, an empirical equation to predict the uniaxial compressive strength for anisotropic rocks is developed. The proposed equation has been used to fit the data for four types of anisotropic rocks, i.e. sandstone, phyllite, slate and shale. Three different statistical parameters (regression R-square value R2, relative error Dp and average absolute relative error AAREP) have been used to assess the predictive ability of the empirical equation. Results show that the predicted uniaxial compressive strength values agree well with the experimental values. Statistical evaluation of performance of the proposed expression has also been carried out using a uniaxial compressive strength database, which includes 274 uniaxial compressive tests conducted worldwide on anisotropic rocks. Further, predictive capabilities of the proposed method have been compared with those of three commonly employed methods. Statistical analysis result shows that it is possible to predict uniaxial compressive strength of anisotropic rocks even if three data of uniaxial compressive strength tests at orientation angle ? 0°, 30°and 90°are available.

Key words: anisotropic rocks, uniaxial compressive strength, empirical equation, statistical evaluation