The characterization of carbonate formations is challenging as compared to sandstones, yet carbonate reservoirs hold over 60 of the world's hydrocarbon reserves. Carbonate reservoirs exhibit a high level of heterogeneity at every scale; from core to field. To be able to manage heterogeneity for reservoir modelling, the formation has to be discretized into a few rock types, each of which having somewhat similar flow properties. Recently, the interest in extending the rock-typing approaches is increasing with the aim to identify the potential layers in complex lithology like carbonates. The approach becomes more rigorous if the geological description is coordinated with petrophysical data, an approach that has been followed in this study. The hydraulic flow units in Arab-D formation were identified and interpreted using both geological facies and petrophysical data. All three methods; histogram analysis, normal probability plot and least-squared regression were utilized to determine the optimum number of hydraulic flow units across Arab-D carbonate formation. Published routine core analysis data from ten wells of Arab-D formation was analyzed and six optimum hydraulic flow units were identified. The average porosity and average permeability of each hydraulic flow unit was then computed. The results were found to be in good agreement with the geological facies data of the Arab-D formation, thus validating the identified flow units. © Published under licence by IOP Publishing Ltd.