PENGEMBANGAN METODE PENENTUAN RELIABILITAS KESTABILAN LERENG TAMBANG TERBUKA BATUBARA DI INDONESIA

• Main Author: Ahmad Azizi, Masagus
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/64808
• Institution: Institut Teknologi Bandung

Currently a number of open pit coal mine do an increase in the depth of excavation so that they provide the potential of increased slope failure and very serious impact threat. Slope stability method used today needs to be reinforced with risk-based indicator of stability. Therefore, the method of determining the reliability of open pit mine slope stability should be developed to minimize the risks that will happen. Characteristics of the rock mass show random nature resulting in the patterns of some particular distribution. On the other hand, the use of limit equilibrium method (Bishop) also results in variations due to the simplification of the method. This can be proved by the existence of slope with Safety Factor (SF) < 1 and SF > 1 experiencing failure. Variations in the physical and mechanical properties of the rock mass, as well as the used method has resulted in variations in the value of SF. Variations in input and output parameters produce a risk that can be measured using Failure Probability (FP) indicators. Determination of the appropriate distribution of random rock mass characteristics data can be approximated using Kolmogorov-Smirnov (KS) method by finding the smallest value of Dn max and compared with the critical value. If there are 2 or more Dn max values smaller than the critical values, then advanced test is conducted using Akaike Information Criteria (AIC) dan Bayesian Information Criteria (BIC) indicators. The highest AIC and BIC values indicate the best distribution that fits the empirical data distribution. The goodness of fit test results can be used for the analysis of data uncertainty level and estimated Safety Factor (SF) and Failure Probability (FP) estimations. The level of data uncertainty can be also known using the indicators of coefficient of variation (CV) after the best distribution is determined, which later can be used to perform the optimization of data. Estimated SF and FP use the goodness of fit test results by entering the type of distribution, mean value, standard deviation, relative minimum, and relative maximum. The SF and FP calculation use Monte Carlo (MC) sampling method and Latin Hypercube (LH). This research also results in acceptable SF & FP criteria for open pit coal mine in Indonesia by the use of back analysis results, determination of correction factor due to model error, determination of the true SF and FP, as well as evaluation of existing slopes. To conclude, the methods of determining the reliability of slope stability of open pit coal mine has been developed.