The association of schedule characteristics of heavy vehicle drivers with continuous eye-blink parameters of drowsiness

•Ocular parameters (Johns Drowsiness Scores, JDS) could detect real-time drowsiness in HVDs on the road.•The risk of drowsiness in HVDs was measured by the occurrence of drowsiness events (JDS ≥ 2.6 events/hour) and assessed by logistic regression models.•Night-time drive, shift start time in the early afternoon, prolonged work hours, short break durations (7–9 h) and short sleep time (<6 h) increased the risk of drowsiness after controlling for the effects of other work schedule metrics.•The magnitude of driver drowsiness was measured using hourly rates of drowsiness events (JDS ≥ 2.6 events/hour) and assessed by mixed linear regression models.•Night-time drive, shift start time in the early morning, prolonged work hours, break times (<9 h), and shift of 8–15 h increased the hourly rate of drowsiness events after adjusting for the effects of other work schedules metrics.•Night-time drives (9 pm- 2 am), 18 to 21 h into the shift, or breaks shorter than 7 h elevated drowsiness event rates substantially relative to “alert state (4.9 events/hour) after controlling for the effects of other work schedule metrics.