The U.S. Nuclear Regulatory Commission (NRC) is consolidating all its severe-accident codes into one code, MELCOR, and making an effort to bring it into a state of parity with SCDAP/RELAP5/MOD3.3 (S/R5/M3.3) to model a Three Mile Island Unit 2 (TMI-2)-like accident. In this regard, this cooperative research project seeks to help the NRC to assess S/R5/M3.3 associated with case studies of the TMI-2 lower-head creep rupture. The results of the simulations clearly demonstrate that the TMI-2 lower-head failure occurs. Thus, solely using the S/R5/M3.3 models of the molten pool and debris-to-vessel contact resistance, without implementing the gap cooling model, cannot explain the conservation of the TMI-2 lower head during the accident. These studies also conclude that the results calculated with the UNIX and Microsoft PC versions of S/R5/M3.3 are comparable, and hydrogen productions as well as lower-head creep ruptures vary with different time steps for the alternative accident. Further, those results for the base case and alternative accident are alike; thus, the models cannot differentiate between the base-case and alternative accident scenarios.