Ensuring worker attentiveness on construction sites is crucial for safety and efficiency. This study explored how AI and machine learning can help monitor attentiveness and identify potential distractions in real time. Through behavioral observations and physiological data analysis, key patterns in worker focus and inattentiveness were uncovered. The most common distraction was engaging in unnecessary conversations (30.0%), while failure to wear PPE was the least observed (12.5%). Physiological data showed that inattentive workers had lower heart rate variability (HRV = 46 ms) and higher galvanic skin response (GSR = 5.9 µS), indicating increased stress and fatigue. A strong negative correlation (-0.73) between HRV and inattentiveness suggests that workers with lower HRV are less focused, while the positive correlation (0.69) with GSR highlights the impact of stress on inattentiveness.

The AI-driven monitoring system performed with high accuracy (91.3%), demonstrating its potential as a reliable tool for identifying inattentive behaviors and improving site safety. By providing real-time feedback and alerts, this technology can help reduce workplace hazards and encourage better adherence to safety protocols. Moving forward, refining these machine learning models and expanding their application to different work environments could further enhance workplace safety and overall worker well-being.