Probabilistic Analysis of Borehole Data for Evaluation of Engineering Properties of Rock Mass

Abstract

Rock mass in nature is generally heterogeneous and anisotropic. Due to this, the variation in engineering properties of the rock mass is quite high. The engineering behaviour of rock mass depends on several factors, and sometimes, the evaluation of these factors is quite complex. Hence, probability-based approaches are used for prediction of engineering properties of the rock mass. Probabilistic analysis is a powerful tool for analysing the inconsistent data and predicts the behaviour of materials. With the help of probabilistic analysis, a solution of varying data can be obtained which may be used in the design and analysis of rock structures. The possibility of failure of structures built in rock mass can also be predicted using probabilistic analysis. Prediction of engineering properties using probabilistic analysis is summarized in the present paper. Borehole data were collected from a project site in the Himalayan region of northern India. The rock mass exposed at the site consisted of genesis with some bands of quartzite and phyllite. The borehole was drilled in vertical direction up to 127 m from ground surface. The average diameter of collected cores was 54.8 mm (NX size). From these cores, RQD and joint characteristics (JCS and JRC) were estimated. Uniaxial compression tests were also conducted on some specimens to increase the accuracy of data. Probabilistic analysis was performed with collected data sets. The results were analysed, and probability-based prediction of RQD, UCS and joint characteristics have been made. The main advantage of this analysis is that for any chosen value of probability, evaluation of these parameters can be easily made which helps in predicting the possibility of failure of rock structures.