Abstract
The strength of rock has significant influence on its performance, and is, therefore, a key input during modelling and analysis of mining and geotechnical engineering structures. The uniaxial compressive strength (UCS), which is a popular parameter to quantifying rock strength can be determined in the laboratory using suggested method by International Society of Rock Mechanics (ISRM). However, the laboratory determination of UCS consumes time, it is costly, and sometimes may not be feasible to perform because of different conditions of rock. Hence, this study attempts to employ Monte Carlo simulation (MCS) approach to estimate UCS, and to overcome various uncertainties associated with UCS estimation. To use MCS approach for UCS estimation, block punch index (BPI), Brazilian tensile strength (BTS), point load index (IS (50) ), and P-wave velocity (Vp) were selected as the model inputs. A multiple linear regression (MLR) equation was developed and used to predict UCS by the MCS approach. The methodology was applied to estimate UCS using real BPI, BTS, Is (50) , and Vp data as inputs. The proposed approach simulated UCS values that are consistent with UCS values measured in the laboratory. The mean of the UCS values simulated through the MCS approach is 119.10 MPa, while the mean of the UCS values measured in the laboratory is 118.42 MPa. In addition, hypothesis testing revealed that the Brazilian tensile strength (BTS) is the parameter with the most influence on UCS of rock for the site investigated.