CONTACT US: Contact info
|
Citation |
Ferreira S, Kokkinogenis Z, Couto A. Transp. Res. F Traffic Psychol. Behav. 2019; 60: 25-36. |
|
Copyright |
(Copyright © 2019, Elsevier Publishing) |
|
DOI |
|
PMID |
unavailable |
|
Abstract |
Professional drivers are particularly exposed to drowsiness and distraction inasmuch as they drive for long periods of time and as a daily routine. Therefore, several studies have been conducted to investigate driverś behavior, supported by controlled experiments (e.g. naturalistic and driving simulator studies). However, due to emerging technologies, new study methods can be developed to complement existing studies. In this study, retrospective data gathered from a driver monitoring system (DMS), which monitored 70 professional drivers from different companies, was used to investigate the effect of journey characteristics on the number of alerts due to either distraction or drowsiness. Two separate negative binomial models were developed, including explanatory variables describing the continuous driving time (sub-journey time), the journey time (a set of sub-journeys), the number of breaks and the breaking duration time. Dummy variables were also included. Interesting results were observed such as increasing continuous driving time, the number of distraction and drowsiness alerts increase too. In contrast, the journey time has the opposite effect decreasing the number of alerts. In the case of distraction alerts, stopping the vehicle during the journey (break) was not statistically significant and the increase in the breaking duration time showed an unexpected effect as the number of alerts increased. This was not the case of drowsiness alerts in which the frequency of breaks and the breaking duration time decreases the alerts. The companies (for which the drivers work) affect the alert frequency differently. The study shows that there is potential in terms of using the data obtained by the new technologies to complement other type of studies based on controlled experiments but also to enhance the development of technologies taking into account the driver profile and the type of journey. Language: en |
|
Keywords |
Alert-based data; Distraction; Driver monitoring system; Drowsiness; Real-world driving; Safety |