Faculty of Engineering of the University of Porto, Research Centre for Territory, Transports and Environment, Rua Dr. Roberto Frias 4200-465 Porto, Portugal.
J Safety Res. 2020 Feb;72:279-285. doi: 10.1016/j.jsr.2019.12.024. Epub 2020 Jan 13.
Due to the negative impact on road safety from driver drowsiness and distraction, several studies have been conducted, usually under driving simulator and naturalistic conditions. Nevertheless, emerging technologies offer the opportunity to explore novel data. The present study explores retrospective data, which was gathered by an app designed to monitor the driver, which is available to any driver owning a smartphone.
Drowsiness and distraction alerts emitted during the journey were aggregated by continuous driving (called sub-journey). The data include 273 drivers who made 634 sub-journeys. Two binary logit models were used separately to analyze the probability of a drowsiness and distraction event occurring. Variables describing the continuous driving time (sub-journey time), the journey time (a set of sub-journeys), the number of breaks, the breaking duration time and the first sub-journey (categorical variable) were included. Additionally, categorical variables representing the gender and age of the drivers were also incorporated.
Despite the limitations of the retrospective data, interesting findings were obtained. The results indicate that the main risk factor of inattention is driving continuously (i.e., without stopping), but it is irrelevant whether the stop is long or short as well as the total time spent on the journey. The probability of distraction events occurring during the journey is higher than drowsiness events. Yet, the impact of increasing the driving time of the journey and stopping during the journey on the probability of drowsiness is higher than the probability of distraction. Additionally, this study reveals that the elderly are more prone to drowsiness. The data also include a group of drivers, who did not provide information on gender and age, who were found to be associated to drowsiness and distraction risk.
The study shows that data gathered by an app have the potential to contribute to investigating drowsiness and distraction. Practical applications: Drivers are highly recommended to frequently stop during the journey, even for a short period of time to prevent drowsiness and distraction.
由于驾驶员困倦和分心对道路安全的负面影响，已经进行了几项研究，这些研究通常是在驾驶模拟器和自然条件下进行的。然而，新兴技术提供了探索新数据的机会。本研究探讨了回顾性数据，这些数据是通过一款旨在监测驾驶员的应用程序收集的，任何拥有智能手机的驾驶员都可以使用该应用程序。
在行驶过程中发出的困倦和分心警报通过连续驾驶（称为子行程）进行汇总。数据包括 273 名驾驶员的 634 个子行程。分别使用两个二元逻辑模型来分析困倦和分心事件发生的概率。描述连续驾驶时间（子行程时间）、行程时间（一组子行程）、休息次数、休息持续时间和第一个子行程（分类变量）的变量被包含在内。此外，还纳入了代表驾驶员性别和年龄的分类变量。
尽管回顾性数据存在局限性，但仍获得了有趣的发现。结果表明，不注意的主要风险因素是连续驾驶（即不停顿），但无论停顿时间长短以及在旅程中花费的总时间如何，都无关紧要。在旅途中发生分心事件的概率高于困倦事件。然而，增加旅程中的驾驶时间和在旅途中停车对困倦发生概率的影响高于对分心发生概率的影响。此外，这项研究表明，老年人更容易困倦。数据还包括一组未提供性别和年龄信息的驾驶员，他们被发现与困倦和分心风险有关。
研究表明，应用程序收集的数据有可能有助于研究困倦和分心。实际应用：强烈建议驾驶员在旅途中经常停车，即使是短时间停车，以防止困倦和分心。
