이분적 터널 암반 분류를 위한 정성적 자료의 지구통계학적 연구 II. 응용 = A Geostatistical Study Using Qualitative Information for Tunnel Rock Binary Classificationll- II. Applcation

다국어 초록 (Multilingual Abstract)

In this paper, the application of the rock classification method based on indicator kriging and the cost of errors, which can incorporate qualitative data, was presented. In particular, the binary classification of rock masses was considered. To this end, a simplified RMR system was used. Since most of subjectivity in this analysis occur during the estimation of loss functions, a sensitivity analysis of loss functions was performed. Through this research, it was found out that an expected cost of errors could successfully be used as an indication for how well a sampling plan was designed. In certain conditions, qualitative data can be more economical than quantitative data in terms of expected costs of errors and sampling costs. Therefore, an additional sampling should be carefully determined depending upon the surrounding geologic conditions and its sampling cost. The application method shown in this paper can be useful for more systematic rock classifications.