Abstract: The phenomenon identification and ranking table (PIRT) is an important basis for nuclear reactor thermal-hydraulic analysis, and it is mainly used in the construction of the code assessment matrix as well as the identification of important input parameters in the best estimate plus uncertainty (BEPU) analysis. It would be difficult to perform the parameter identification process when lacking of expert experience since the development of the PIRT relies heavily on relevant expert experience. One way to decrease to reliance on expert experience in the phenomenon identification and ranking process is to use the quantitative sensitivity analysis technique, such as the Sobol or the moment-independent method. In this paper, the research on determining the parameter importance rankings in short of expert experience was carried out by using an efficient moment-independent global sensitivity analysis method. Since the small break loss of coolant accident (SBLOCA) scenario is an important case in the safety assessment of the nuclear power plant, it was adopted as the scenario of interest in the research. Therefore, the target scenario was determined to be the SBLOCA in a typical three-loop pressurized water reactor (PWR), and five different break sizes were analyzed. Firstly, the nominal cases were simulated to analyze the accident phases as well as potential thermal-hydraulic phenomena. Then, the figure of merit (FOM) was determined to be the minimum collapsed level (MCL) in the active core section. A total of 54 uncertain input parameters with potential influence on the FOM were then screened and added by referring to some PIRTs and by analyzing the base case calculation. And then, the efficient moment-independent global sensitivity analysis method was utilized to calculate the sensitivity measure of each input on the FOM. The importance rankings of the inputs were then transferred into the Savage scores and the inputs can be grouped qualitatively based on the Savage scores. The results show that there are two phases where the collapsed level deceases significantly, namely the blowdown phase and the loop seal and clearing phase. The parameters that have significant influences on the MCL during the blowdown phase are mainly the initial power, the initial primary pressure, the critical flow model, and the flow resistance in the primary loop. In the loop seal and clearing phase, the core level drop is also related to the core boiling, so the interfacial friction model, various heat transfer models and the auxiliary feedwater system delay also have large influences on the MCL. Finally, a parameter importance ranking table of the SBLOCA scenario was obtained, which can provide guidance for similar cases.