Abstarct: Drilling through the Earth's laxyers is essential to accessing hydrocarbons buried deep underground. Consequently, minimizing drilling costs has always been a priority for drilling companies. Among the key factors influencing these costs is the drill bit, which plays a crucial role in reducing overall drilling expenses. Although the cost of drill bits accounts for approximately 10% of total drilling costs, extending the lifespan of the bit can significantly enhance the rate of penetration, reduce the frequency of bit replacements, and minimize non-productive time associated with such replacements. In this study, controllable drilling parameters were optimized to reduce the bit wear rate (thereby extending bit life) and improve the drilling rate. Petrophysical and mudlogging data were collected from three wells in the Azadegan field, located in southwestern Iran. Initially, geomechanical parameters for reservoir rocks were estimated by constructing a one-dimensional geomechanical model for the wells. This data was then combined into a databaxse of geophysical and upscaled geomechanical parameters. The wells were divided into two groups: training (Wells A and B) and testing (Well C). A second version of the Non-Dominated Sorting Genetic Algorithm (NSGA-II) was employed alongside a Random Forest (RF) algorithm to identify a subset of features with the greatest impact on the drilling rate and bit wear rate. Predictive models were developed for these metrics using the Gaussian Process Regression (GPR) and Random Forest algorithms, applied to normalized training data. These models were evaluated on the test dataset, and the best-performing models were selected for further use. Using the predictive models in NSGA-II, the controllable drilling parameters—weight on bit, rotational speed, and flow rate—were optimized across the studied depth intervals of the wells. Results demonstrated that GPR-baxsed models outperformed RF-baxsed models in terms of accuracy and generalizability for predicting both drilling rate and bit wear rate. The optimization process led to reduced bit wear rates across all studied wells and improved drilling rates in Wells A and C. Additionally, a comparison of the Mechanical Specific Energy (MSE) before and after optimization revealed a significant reduction in drilling risks.