Decision aids for tunnel exploration

Abstract

Tunnels are subsurface passages which are often constructed without removing the overlying rock or soil. It follows that the lack of a priori knowledge of subsurface conditions poses major challenges in their preliminary design and planning. Considerable construction savings may be achieved through the proper collection and interpretation of information obtained through site exploration. However, exploration results are often not completely reliable and site exploration in itself involves a cost. Exploration planning is therefore a process of decision making under uncertainty. Einstein et al. (1978) provide a model that applies decision analysis to the tunnel exploration problem. This thesis first describes the model devised by Einstein et al. and provides numerical techniques for implementing it in a programming package. A package in Visual Basic for Applications is presented which implements the model for a generic tunnel. The thesis concludes by applying the devised package to the North Kenmore Tunnel (Washington State).