<p>The present study explores the machinability of graphene-reinforced aluminium metal matrix composites (Gr–Al MMC), focusing on performance metrics such as surface roughness and chip width under minimum-quantity lubrication (MQL) with liquid CO₂ and dry machining. While the mechanical properties of Gr–Al MMC have garnered attention, its machinability characteristics remain underexplored, particularly in varying cooling environments. We propose a novel machine learning framework based on experimental data to predict and optimize these machinability responses. Different machine learning (ML) models, namely random forest, linear regression, decision tree, XGboost and support vector regression are used to forecast optimal machining conditions, confirming better machinability. The intricate relationships between machining factors and how they affect machinability results are also captured. The experimental procedure was carried out with cutting speed, feed rate, depth of cut and cutting condition as input parameters in accordance with the design of the experiment (DOE) outlines. The ML models were authenticated utilizing cross-validation techniques, comparing experimental and estimated values. The machine learning model demonstrates high predictive accuracy, providing actionable insights for optimizing machining conditions. This investigation highlights the potential of machine learning in developing adaptive machining strategies tailored for advanced composites, paving the way for enhanced manufacturing processes. The outcome of this research work emphasizes the implementation of Gr–Al MMC material in advanced applications while focusing on the role of innovative cooling/lubrication strategies in modern manufacturing practices.</p>