The Key Laboratory of Road and Traffic Engineering, Ministry of Education, Tongji University, Shanghai, 201804, China; College of Transportation Engineering, Tongji University, 4800 Cao'an Highway, Shanghai, 201804, China.
Accid Anal Prev. 2020 May;139:105499. doi: 10.1016/j.aap.2020.105499.
Previous studies have focused on the impact of visibility level on drivers' behavior and their safety in foggy weather. However, other important environmental factors such as road alignment have not been considered. This paper aims to propose a methodology in investigating rear-end collision avoidance behavior under varied foggy conditions, with focusing on changes in visibility and road alignment in this study. A driving simulator experiment with a mixed 2 × 4 × 6 factor design was conducted using an advanced high-fidelity driving simulator. The design matrix includes two safety-critical conditions, four visibility conditions, and six road alignment situations (in terms of the road curve and slope). Behavior variables from different dimensions were identified and compared under varied conditions. To estimate the safety of drivers, a time-based measurement, speed reduction time, is selected among the variables as a measure of safety. The survival analysis approach was introduced to model the relationship between environmental factors and driver safety, using speed reduction time as the survival time. Both the Kaplan-Meier method and the COX model were applied and compared. Results generally suggest that reduced visibility leads to more dangerous rear-end collision avoidance behavior from different aspects. Though findings are mixed regarding the road alignment, the impact of the road alignment was found to be significant. Interestingly, conditions of downward slope were found to be safer. Overall, the COX model outperformed the Kaplan-Meier method in understanding the impact of environmental factors, and it can be applied to investigate other contributing factors for freeway safety under foggy weather conditions.
先前的研究主要集中在雾天能见度水平对驾驶员行为及其安全性的影响上。然而，其他重要的环境因素，如道路线形，尚未得到考虑。本文旨在提出一种在不同雾天条件下研究追尾碰撞规避行为的方法，重点研究能见度和道路线形的变化。本研究采用先进的高保真驾驶模拟器，进行了一项混合 2×4×6 因素设计的驾驶模拟器实验。设计矩阵包括两个安全关键条件、四个能见度条件和六个道路线形情况（道路曲线和坡度）。在不同条件下，从不同维度识别和比较行为变量。为了估计驾驶员的安全性，从变量中选择基于时间的测量，即减速时间，作为安全性的度量。引入生存分析方法，将环境因素与驾驶员安全性之间的关系建模为减速时间的生存时间。应用并比较了 Kaplan-Meier 方法和 COX 模型。结果普遍表明，能见度降低会导致更危险的追尾碰撞规避行为。尽管道路线形的结果存在差异，但道路线形的影响被发现是显著的。有趣的是，下坡条件被发现更安全。总的来说，COX 模型在理解环境因素的影响方面优于 Kaplan-Meier 方法，可用于研究雾天条件下高速公路安全的其他影响因素。
