Contemporary manufacturing systems face the persistent challenge of simultaneously addressing conflicting performance metrics, specifically the reduction of surface roughness (Ra) while enhancing material removal rate (MRR). Traditional optimization approaches often struggle with the inherent trade-offs between surface quality and production efficiency. This investigation presents an innovative methodology that combines high-order polynomial regression with Deep Reinforcement Learning (DRL) to achieve multi-objective parameter optimization in machining operations. The experimental framework employed a Box-Behnken statistical design systematically evaluate the relationships between cutting velocity (Vc), feed per tooth (fz), and axial depth of cut (ap) with respect to surface finish and material removal characteristics. High-order polynomial models were constructed to establish predictive relationships between process parameters and response variables. Regression analysis revealed substantial parameter sensitivity in determining machining performance outcomes. The optimization strategy utilized DRL methodology to navigate the complex parameter space while addressing competing objectives through a Markov Decision Process formulation. A carefully constructed reward function facilitated the exploration of parameter combinations that balance surface integrity with productivity requirements. The algorithm systematically generated Pareto-optimal solutions representing the frontier of achievable performance trade-offs. Experimental validation demonstrated notable performance boundaries: optimal surface quality achieved Ra = 0.393 µm using Vc = 89.398 m/min, fz = 0.101 mm/z, and ap = 0.118 mm, while maximum productivity yielded MRR = 183.851 mm3/min at Vc = 168.770 m/min, fz = 0.195 mm/z, and ap = 0.482 mm. These findings establish the viability of DRL-based approaches for resolving multi-criteria optimization challenges in precision manufacturing. The developed framework offers significant potential for integration within adaptive CNC control systems, contributing to the advancement of intelligent manufacturing technologies. Future applications may extend this approach to broader sustainability considerations and energy-efficient production strategies.