Physics-informed analytical intelligence framework for mechanism-driven decision support of tool wear in smart machining systems

Tool wear critically governs the reliability, quality, and economic efficiency of machining systems; however, existing modeling approaches remain predominantly prediction-oriented, empirically dependent, and insufficiently capable of extracting mechanism-driven knowledge for intelligent manufacturing decision support. This study develops a physics-informed analytical intelligence framework that transforms fundamental wear mechanisms into interpretable computational reasoning for smart machining systems. The framework integrates abrasion, adhesion, and diffusion within a unified force–temperature-coupled analytical formulation, establishing explicit relationships between wear evolution, material properties, thermo-mechanical fields, and process parameters. Unlike conventional approaches that primarily focus on numerical wear prediction, the proposed framework enables analytical interpretation of mechanism dominance, transition behaviour, and parameter sensitivity without experimental calibration. A gradient-based sensitivity analysis further converts the analytical formulation into decision-relevant knowledge, allowing systematic evaluation of tool–material compatibility and machining conditions during early-stage process planning. The framework is evaluated across four representative machining scenarios and experimentally validated using both uncoated and coated tools. The predicted mechanism transitions and coating-dependent behaviors exhibit consistent agreement with observed wear patterns, demonstrating the capability of the proposed framework to provide physics-based decision support beyond conventional wear prediction. By integrating cutting mechanics, thermo-physical material attributes, and multi-mechanism wear evolution into a unified analytical intelligence structure, this work establishes a foundation for interpretable and data-efficient digital manufacturing systems, enabling knowledge-driven tool selection and process planning in smart machining environments.