In this paper, preliminary support design of the main underground opening (i.e., mine adit) located at the Artana lead-zinc mine, Kosovo, was examined by employing both conventional and numerical methods for safe underground excavation and design. In order to conduct field studies including discontinuity surveying and sampling for laboratory testing two empirical methods, namely rock mass rating (RMR) and geological strength index (GSI) were employed. For the purpose of determining necessary support units RMR system was utilized. However, these kind of systems can take into account for neither the depth of underground opening nor in situ field stresses. For this reason, empirical design methods (i.e., RMR system) failed to investigate the performance of rock support units; therefore, a 2D finite element analysis program was used to assess the performance of the proposed support systems. This indicated that RMR system might not be applicable for poor and very poor rock masses located in deep environment (i.e., 300 and 400 m). Moreover, this is linked to the fact that the RMR system does not consider in situ stress conditions. This study showed that when empirical methods are supported by numerical analysis, the preliminary support system design will be much more reliable.