In light water reactors, particularly the pressurized water reactor (PWR), the severity of a loss-of-coolant accident (LOCA) would limit how high the reactor power can operate. Although the best-estimate LOCA licensing methodology can provide the greatest margin on the peak cladding temperature (PCT) evaluation during a LOCA, it generally takes much more resources to develop. Instead, implementation of evaluation models required by Appendix K of 10CFR50 on an advanced thermal-hydraulic platform such as RELAP5, TRAC, etc., also can gain significant margin for the PCT calculation. Through compliance evaluation against Appendix K of 10CFR50, all of the required evaluation models have been implemented in RELAP5-3D. To verify and assess the development of the Appendix K version of RELAP5-3D, nine kinds of separate-effects experiments and eight sets of LOCA integral experiments were adopted. Through the assessments against separate-effects experiments, the success of the code modification in accordance with Appendix K of 10CFR50 was demonstrated. Besides, one set of a typical integral large-break LOCA from Loss-of-Fluid Test Facility experiments (L2-5) has also been applied to preliminarily evaluate the integral performance of the Appendix K version of RELAP5-3D. The PCT predicted by the evaluation models is greater than the one from best-estimate calculation in the whole LOCA history with the conservatism of 150 K, and the measured PCTs of L2-5 are also well bounded by the evaluation model calculation. Another seven sets of integral-effect experiments will be further applied in the next step to ensure the reasonable integral conservatism of the newly developed LOCA licensing analysis code (RELAP5-3DK/INER), which can cover all the phases of both large- and small LOCA in one code.