抄録
It is well-know that unforeseeable geological condition in underground construction projects causes cost overrun and delay of construction period. The reason why unforeseeable geological condition occurs is the uncertainty of geological condition estimated by geological surveys. Authors have estimated construction cost in the case of optimistic, likelihood, and pessimistic scenarios and compared the estimation with an actual one in order to evaluate the risk of construction cost variation in mountain tunnel projects. However, it is also important to evaluate the case in which unforeseeable geological condition, which is seldom occurred, causes high cost overrun in risk management. From such a viewpoint, this study proposes the methodology to evaluate probability of each rock classification by modeling the results calculated by external drift kriging, which is one of Geo-Statistics theory, as random valuable. In addition, the methodology to evaluate geological condition with core evaluation point, which this study uses, depends on data obtained in limited areas because the methodology uses borehole data. Therefore, the result of the estimation changes drastically if there is non-typical data. From such a viewpoint, this study proposes the methodology to eliminate non-typical data through evaluating the relation between each evaluation item of core evaluation point and physical property and using data in other tunnel construction project.
