Experience with the development and use of a simple DFN approach over a period of 30 years

Abstract

A contractual dispute on a tunnelling project more than 30 years ago was the impetus for the development of a program to generate discontinuity patterns. The purpose was to allow quantification of potential rockfalls, and determine appropriate rockbolt lengths and spacings. The simple discrete fracture network (DFN) program used measured or estimated statistical distributions of joint parameters to generate rock mass discontinuity patterns in any required two-dimensional section. The approach lends itself to a probabilistic analysis, and to use data determined quantitatively from the DFNs in the evaluation of risk. The development philosophy and application of the method will be described in the paper, supported by descriptions of several cases of practical engineering problems. Since a DFN does not represent the actual rock mass, but only one example of a ‘theoretical’ rock mass, multiple DFN representations need to be used to quantify engineering outputs.