Abstract Accurately assessing machine health in Industry 4.0 smart manufacturing is challenging due to complex, heterogeneous data streams. Existing diagnostic models often focus on isolated faults or individual sensor signals, lacking a comprehensive, multidimensional evaluation of overall equipment condition. To address this gap, this study proposes a hybrid Principal Component Analysis (PCA) and Entropy weighting framework to generate a robust Composite Machine Health Score (CMHS). Empirically... Accurately assessing machine health in Industry 4.0 smart manufacturing is challenging due to complex, heterogeneous data streams. Existing diagnostic models often focus on isolated faults or individual sensor signals, lacking a comprehensive, multidimensional evaluation of overall equipment condition. To address this gap, this study proposes a hybrid Principal Component Analysis (PCA) and Entropy weighting framework to generate a robust Composite Machine Health Score (CMHS). Empirically validated using a smart manufacturing dataset comprising 10,000 records and 15 operational, environmental, and production variables, the framework integrates PCA for variance based dimensionality reduction with entropy weighting for objective, information theoretic indicator evaluation. Results demonstrate that the CMHS effectively differentiates machine conditions, successfully separating highly efficient operations from severely degraded states. Entropy analysis revealed that downtime duration and environmental humidity possess the highest discriminatory power in distinguishing machine health states. Furthermore, the hybrid fusion approach proved significantly more robust than single method evaluations, notably identifying latent anomalies where machines exhibited extreme performance degradation without triggering explicit fault alarms. Ultimately, this framework offers a highly interpretable, data driven decisionsupport tool for predictive maintenance. By synthesizing multidimensional manufacturing data into a single actionable metric, it enables organizations to accurately prioritize maintenance interventions, mitigate hidden inefficiencies, and enhance overall operational resilience. This advancement directly supports the critical transition from reactive maintenance to proactive, condition-based strategies in modern production environments. Read More Read Less

Abstract Accurately predicting cutting tool life is essential for minimizing unplanned downtime and optimizing predictive maintenance in modern milling operations. While existing tool condition monitoring studies focus primarily on wear detection, they often lack integrated degradation trajectory modeling and reliability assessment. This study proposes a comprehensive Prognostics and Health Management (PHM) framework combining vibration based condition monitoring with stochastic state space modeling to evaluate tool... Accurately predicting cutting tool life is essential for minimizing unplanned downtime and optimizing predictive maintenance in modern milling operations. While existing tool condition monitoring studies focus primarily on wear detection, they often lack integrated degradation trajectory modeling and reliability assessment. This study proposes a comprehensive Prognostics and Health Management (PHM) framework combining vibration based condition monitoring with stochastic state space modeling to evaluate tool wear progression. Utilizing the "Roughness in Milling Process" dataset, which pairs tri-axial vibration signals with surface roughness measurements across eight workpieces, the research constructs a normalised degradation index to track tool deterioration. The methodology incorporates wear state classification, Hidden Markov Models (HMM), and Hidden Semi-Markov Models (HSMM) with Weibull duration distributions to capture realistic state transitions. Furthermore, an absorbing Markov chain framework is applied to reliability analysis, alongside Monte Carlo simulations for probabilistic estimation of Remaining Useful Life (RUL). Results demonstrate a strong monotonic correlation between increasing vibration magnitudes and worsening surface roughness. The HSMM effectively modeled prolonged intermediate degradation stages, while probabilistic RUL forecasts provided actionable insights for maintenance scheduling. Although constrained by the dataset's limited size, this integrated approach successfully bridges raw sensor data with actionable reliability metrics. Ultimately, the proposed framework establishes a robust, data-driven foundation for next-generation predictive maintenance strategies aligned with Industry 4.0 intelligent manufacturing paradigms. Read More Read Less

Abstract Traditional predictive maintenance often lacks interpretability and fails to uncover the underlying physical mechanisms of equipment degradation. To address this critical gap, this study proposes a dynamic causal network framework that transitions maintenance strategies from purely predictive to fully prescriptive. Utilizing a comprehensive industrial machine sensor dataset, the research employs a multi stage analytical methodology integrating correlation analysis, mutual information, Granger causality testing, and... Traditional predictive maintenance often lacks interpretability and fails to uncover the underlying physical mechanisms of equipment degradation. To address this critical gap, this study proposes a dynamic causal network framework that transitions maintenance strategies from purely predictive to fully prescriptive. Utilizing a comprehensive industrial machine sensor dataset, the research employs a multi stage analytical methodology integrating correlation analysis, mutual information, Granger causality testing, and rolling window dynamic network analysis. The empirical results reveal exceptionally strong linear and nonlinear dependencies among core operational variables, particularly temperature, vibration, and pressure. Furthermore, Granger causality analysis establishes directional information flows, successfully identifying vibration as a critical precursor to defect generation. Crucially, the rolling window analysis demonstrates that causal connectivity is non stationary and fluctuates significantly over time. Periods of elevated causal density correspond to degradation accumulation or abnormal system behaviors, while reduced connectivity indicates post maintenance recovery or stable operation. These findings indicate that machine health is reflected not only in individual sensor magnitudes but also in the evolving structural interactions among variables. Consequently, dynamic causal connectivity serves as a novel, highly interpretable health indicator. By uncovering the directional propagation of degradation, the proposed framework provides actionable insights for root cause diagnosis and early fault detection, ultimately supporting more proactive, targeted, and efficient maintenance decision making in smart manufacturing environments. Read More Read Less