A data-driven approach for high-accuracy tool wear prediction in machining Hastelloy C276

= 0.982, MAE = 0.004 mm, RMSE = 0.006 mm), effectively capturing nonlinear wear behavior associated with thermal-mechanical interactions. Experimental results confirmed the strong influence of lubrication environment on tool life, with nano-MQL reducing average flank wear by 28-35% compared to Dry machining due to enhanced cooling and tribo-film formation by hBN nanoparticles. Feature-importance analysis further identified lubrication condition, machining length, and feed rate as the dominant predictors governing wear evolution. The study demonstrates that reliable tool-wear prediction can be achieved using machining parameters alone-without additional sensors-highlighting the potential of ML-driven frameworks for future intelligent tool-condition monitoring and sustainable machining of difficult-to-cut superalloys.