The paper begins by noting that many of the contributions to the EUROCK2013 Symposium deal directly with risk. An explanation of epistemic and aleatory uncertainty then follows with further notes on risk analysis techniques and ‘black swan’ events. Three illustrative rock engineering application examples are discussed in some detail: shale gas extraction, CO2 storage, and radioactive waste disposal—with the emphasis being on successfully achieving the project objectives. The paper concludes with ten recommendations, concentrating on the further development of the subjects of uncertainty and risk as they apply to rock mechanics and rock engineering. Especially important is the need to validate coupled numerical modelling and to make the programs more accessible to practitioners. This, together with the wider application of the techniques used for the design of radioactive waste repositories, will significantly reduce the impact of uncertainties and risks inherent in all rock engineering modelling, design and construction.
In order to be able to coherently design an underground rock engineering project, one has to be able to predict the future. For example, what will happen when a tunnel of this diameter, at this depth, excavated in this direction, in this rock mass, is constructed? Will rock blocks fall into the tunnel, will the concentrated rock stress damage the tunnel, will the tunnel encounter any major faults? If one is unable to answer questions such as these, indicating that the future cannot be predicted, the design of the project is subject to some level of uncertainty and risk. Since, we cannot yet obtain full information about any rock mass, all rock engineering projects are subject to uncertainty and hence risk. Reduction of the risks is directly related to reduction of the uncertainties and so, for improved rock engineering design, we must concentrate on the uncertainties involved. In the worst case, lack of attention to this subject could endanger lives and prejudice the functionality of the completed project.