In CANDU-6 type reactors, a multiple steam generator tube rupture (MSGTR) accident is characterized as a reactor building (RB) bypass scenario. Although the probability of an MSGTR accident in the CANDU-6 type Wolsong plants is very low, a direct release of radioactive nuclides to the environment can cause severe radiation exposure to residents around the plants. For this reason, a best-estimate analysis has been carried out based on the currently available best-practice knowledge in the Wolsong plants. The present study is divided into three steps. First, the existing parameter file of generic CANDU-6 plants for the MAAP-ISAAC 4.03 code was revised to incorporate its design-specific features for the Wolsong plants. A comparative analysis between the existing and modified parameter files showed that the progress of a severe accident was delayed or mitigated in the case of the modified parameter file. Secondly, variables related to the steam generator decontamination factors (DFs) were improved. It was confirmed that applying the steam generator DFs based on the ARTIST test significantly reduced the release fraction of each nuclide element to the environment. Finally, severe accident management guidance (SAMG) developed for the Wolsong plants was used to manage the progression of severe accidents and to limit the release of fission products during a MSGTR accident. A mitigation action of closing the main steam safety valves (MSSVs) after meeting a SAMG entry condition can significantly reduce the amount of released fission products compared with simply supplying feedwater to a broken steam generator (SG). To meet the domestic regulatory requirement for the release of Cs-137 to the environment (100 TBq), it was identified that an operator should supply feedwater to the broken SG within 1.5 h after meeting the SAMG entry conditions. The best-estimate analysis methodology and relevant accident management strategy for the mitigation of severe accidents, which have been developed through the study, can be applied to all similar CANDU plants and provide valuable insights into Level 2 probabilistic safety assessment and development of relevant SAMG.