Distinct mental fatigue mechanisms under prolonged work and night shifts Implications for safety management in a real-world oil drilling site.

OBJECTIVE

This study aimed to investigate how shift work and prolonged working hours affect mental fatigue under real-world working conditions, focusing on oil and gas drilling operations.

METHODS

Physiological data from drillers, collected via wearable sensors, were analyzed. EEG features and heart rate data were integrated, and a semi-supervised machine learning model was developed to classify mental fatigue.

RESULTS

The study revealed distinct heart rate changes under prolonged work and night-shift conditions, reflecting parasympathetic regulation and circadian disruption. EEG analysis showed significant alterations in the α band and increased functional connectivity in the left frontal lobe. The machine learning model achieved a 93.46% classification accuracy.

CONCLUSIONS

Night-shift work induces different fatigue characteristics from day shifts, necessitating tailored safety management strategies. The fatigue recognition method offers a proactive approach for early intervention in complex operational environments.