various methods have been proposed for reservoir rock typing; these approaches are often associated with limitations and simplifying assumptions that hinder accurate definition of flow units and dynamic classification of reservoir rocks. Studies indicate that the selection of an appropriate rock typing method depends on the intrinsic properties of the reservoir rock as well as the quantity and quality of the available data. Rock typing is a key strategy for mitigating these challenges by grouping reservoir rocks with similar petrophysical and flow properties. In this study, multiple rock typing methodologies—including the Flow Zone Indicator (FZI), Modified Flow Zone Indicator (FZIM), the Winland method, and permeability-based classifications—were applied to a carbonate reservoir in a southwestern Iranian oil field. To evaluate the effectiveness of these methods, statistical techniques such as the Elbow method and correlation analysis were employed. Among the evaluated models, the FZIM, the permeability-based classification, and the R-model exhibited superior performance based on relative permeability curve analysis. Consequently, these models were selected for use in dynamic reservoir simulation. In addition, two novel rock typing models were developed using relative permeability data as the basis for classification. Simulation results demonstrated that the proposed models achieved an acceptable history match for water production trends. However, rock type classification showed a limited impact on the history matching of gas-oil ratio and reservoir pressure decline.