Analysis of the effect of thermal comfort on driver drowsiness progress with Predicted Mean Vote: An experiment using real highway driving conditions

Abstract

Drowsy driving causes serious accidents. Driver drowsiness is affected by the thermal environment, but drowsiness predictions in previous research are based on a limited thermal environment and do not consider realistic thermal conditions, including warm environments and individual differences. This study uses the Predicted Mean Vote (PMV) index, a personal thermal environment indicator, in drowsiness prediction to treat individual differences. Furthermore, it provides a model of thermal-drowsiness progression characteristics that comprehensively covers both cold and warm environments. The drowsiness data were collected from 29 subjects driving on a highway under six levels of thermal comfort conditions. Based on the collected data, a model of the amount of change in drowsiness after 15 min was built. The results show that the drivers were most drowsy in the slightly warmer condition (PMV + 0.2) and hardly drowsy in both the cooler and warmer conditions after 15 min of driving. This inverted U-shaped characteristic was numerically modeled as the effect of the thermal environment on driver drowsiness. In addition, in models predicting drowsiness after 15 min, PMV had a greater effect on improving accuracy than conventional driving time. These results suggest that for short periods of driving, the personal thermal environment may have a greater influence on drowsiness progression than driving time. Whilst limited by the small sample size, this study provides insight into the relationship between thermal comfort and drivers’ drowsiness under real-world highway driving conditions.