Abstarct: Rheological parameters of drilling mud, including plastic viscosity and yield point, are among the most important properties of drilling fluid and play a key role in optimizing oil well drilling operations and reducing total drilling costs. In this study, an artificial neural network model was used for real-time estimation of rheological parameters baxsed on operational drilling data. The data used in this study were collected from real field measurements obtained from seven oil wells located in southern Iran. Data from Wells 1, 2, 4, and 5 were divided into 70% training and 30% testing subsets for model development and initial performance evaluation, while unseen data from Wells 3, 6, and 7 were used as an independent validation dataset to assess the real-time prediction capability of the developed model. After data preprocessing and outlier removal, True Vertical Depth, mud weight, Marsh funnel viscosity, pH, hardness, solids content, and chloride concentration were selected as model inputs to predict plastic viscosity (PV) and yield point, considering the critical role of drilling mud rheological properties in formation pressure control, cuttings transport, wellbore stability, and hydraulic performance, as well as the limitations of conventional measurement techniques and existing predictive models. For the artificial neural network model, the coefficient of determination for predicting plastic viscosity in the training, testing, and validation stages was 0.9616, 0.9399, and 0.9095, respectively. For yield point prediction, the corresponding values were 0.9794, 0.9519, and 0.9204. In addition, the low error values obtained in all stages indicate the acceptable accuracy, stable performance, and strong generalization capability of the developed model for real-time prediction of drilling mud rheological parameters.