Sensor-Based Frameworks for Predictive Condition Monitoring and Operational Assessment

The growing complexity of industrial systems, particularly in discrete manufacturing and assembly environments, has increased the need for monitoring approaches capable of identifying degradation before critical failures occur. Traditional non-destructive testing (NDT) frameworks primarily focus on machines and structural components, often overlooking human operators whose performance variability, workload, and environmental exposure influence system reliability. Within the NDT 4.0 paradigm, this study proposes a sensor-based conceptual framework that integrates human factors with mechanical components in a unified operational assessment structure. The framework incorporates multi-modal sensing technologies, including wearable, environmental, vibration, and acoustic sensors, to capture real-time signals of equipment condition and human operational states. These heterogeneous data streams are fused within a data integration layer to generate a unified operational health index. A layered architecture structures the framework into four stages: data acquisition, preprocessing, feature extraction, and predictive assessment. Sensor data are analyzed to identify anomalies and deviations from baseline operational signatures associated with early-stage degradation, including mechanical faults and operator-related conditions such as fatigue, repetitive strain, and environmental stressors. To demonstrate practical applicability, the framework is contextualized within an automotive assembly workstation, where operator fatigue indicators are integrated with vibration and acoustic signals from assembly tools to support early detection of operational risks. By incorporating human–system interactions into predictive monitoring, the framework enhances operational reliability, supports preventive maintenance planning, and strengthens safety management in industrial environments undergoing digital transformation.