Tool condition monitoring (TCM) is crucial in modern manufacturing for ensuring product quality, minimizing downtime, and enhancing productivity. The researchers largely overlooked the integration of multiple algorithms in developing these TCM systems. Accordingly, this paper presents an ensemble-learning-based integrated diagnostics and prognostics framework for TCM. First, we inventively formulate the diagnostic and prognostic modules that work in collaboration to determine the current health state of cutting tools and predict their Remaining Useful Life (RUL). Herein, the diagnostic module employs an ensemble of classifiers like Logistic Regression, Support Vector Machine, and Multi-Layer Perceptron with majority voting strategies to achieve accurate and robust multi-level wear classification. On the other hand, the prognostic module leverages the Random Forest algorithm to estimate RUL, focusing specifically on when the tool is in a critical wear zone. Subsequently, a case study involving a CNC machining system is conducted to evaluate the effectiveness of the proposed approach in enhancing manufacturing productivity. We have comprehensively evaluated the approach based on its reliability, robustness, and applicability. Extensive validation using the experimental dataset reveals high diagnostic accuracy (91%) and reliability. Additionally, the prognostic module demonstrates low prediction error (mean absolute error of 4.886 and R-squared value of 0.923 in critical wear zones), emphasizing its suitability for real-world applications. The findings confirm the proposed system’s effectiveness in improving manufacturing productivity and maintenance planning. In essence, by using advanced ensemble learning, this study offers a scalable and reliable solution, contributing significantly to the goals of intelligent manufacturing and Industry 4.0 initiatives.