As known to all

the fact that the stability of underground rock is mainly controlled by joint; and the stochastic distribution of geometric and mechanical parameters of joint causes high degree of uncertainty; a reliability analysis method about stability of rock blocks considering uncertainty in geometric parameters and mechanical parameters of joint combined with key block theory is presented; and a proper

combined evaluation model for failure probability is developed. The main objective of this research is to develop a probabilistic approach that can be implemented to assess the probabilistic occurrence and probability of failure of wedges within excavations. This approach is based on the use of stochastic fracture networks FracSim3D to simulate the fracture network and key block theory to identify the wedges and calculate the factor of safety. A programming code based on Fortran 95 has been implemented to perform the main probabilistic wedge occurrence and wedge failure analysis. To evaluate the feasibility of this new probabilistic approach

the procedure is applied to a practical example

a cooper mine in Adelaide

Australia

with considering the uncertainty of dip angle

dip direction

cohesion and friction angle. Finally

the analysis method of conditional probability is proposed to find the failure probability of wedge in plane failure. The calculation results show that the occurrence probability of wedge in plane failure is 11.0%

and the probability of failure is 3.85%

which is greater than the general engineering acceptable risk level. Results of the probabilistic method indicate that the method can be used as a basis model to evaluate the reliability of wedge.