Abstarct: Estimating the Rate of Penetration (ROP) is considered a fundamental indicator in drilling operations due to its direct role in enhancing drilling efficiency and reducing overall operational costs. In this research, one of the artificial intelligence techniques, namely the Artificial Neural Network (ANN), was utilized alongside the traditional physics-baxsed model, in addition to developing a hybrid model that integrates both approaches. This combination aims to achieve more reliable and highly accurate predictions of the penetration rate. The data used for developing the models are baxsed on real field data obtained from four wells located in southern Iraq. The data from Wells 1, 2, and 3 were used for training, while the data from Well 4 — which was not included in the training phase — were employed for model evaluation and performance verification under unseen conditions. Prior to modeling, comprehensive data preprocessing was carried out, including cleaning, removal of outliers, and normalization to ensure stable and accurate results. The findings demonstrated that the hybrid model, which combines the physics-baxsed model with the artificial neural network, delivered the best predictive performance, achieving a correlation coefficient of 0.94 for the test dataset. The artificial neural network alone also showed strong predictive capability, achieving a correlation coefficient of 0.93 in estimating the penetration rate of the studied formation. Therefore, it can be concluded that integrating physics-baxsed knowledge with artificial intelligence techniques significantly enhances prediction accuracy and improves result reliability compared to using each approach independently