Uddin Majbah, Huynh Nathan
University of South Carolina, Department of Civil and Environmental Engineering, 300 Main St, Columbia, SC 29208, USA.
University of South Carolina, Department of Civil and Environmental Engineering, 300 Main St, Columbia, SC 29208, USA.
Accid Anal Prev. 2017 Nov;108:44-55. doi: 10.1016/j.aap.2017.08.009. Epub 2017 Sep 6.
This paper investigates factors affecting injury severity of crashes involving trucks for different lighting conditions on rural and urban roadways. It uses 2009-2013 Ohio crash data from the Highway Safety Information System. The explanatory factors include the occupant, vehicle, collision, roadway, temporal and environmental characteristics. Six separate mixed logit models were developed considering three lighting conditions (daylight, dark, and dark-lighted) on two area types (rural and urban). A series of log-likelihood ratio tests were conducted to validate that these six separate models by lighting conditions and area types are warranted. The model results suggest major differences in both the combination and the magnitude of impact of variables included in each model. Some variables were significant only in one lighting condition but not in other conditions. Similarly, some variables were found to be significant in one area type but not in other area type. These differences show that the different lighting conditions and area types do in fact have different contributing effects on injury severity in truck-involved crashes, further highlighting the importance of examining crashes based on lighting conditions on rural and urban roadways. Age and gender of occupant (who is the most severely injured in a crash), truck types, AADT, speed, and weather condition were found to be factors that have significantly different levels of impact on injury severity in truck-involved crashes.
本文研究了农村和城市道路上不同光照条件下涉及卡车碰撞事故的伤害严重程度的影响因素。它使用了来自公路安全信息系统的2009 - 2013年俄亥俄州碰撞事故数据。解释性因素包括乘员、车辆、碰撞、道路、时间和环境特征。考虑到两种区域类型（农村和城市）的三种光照条件（白天、黑暗和有路灯），开发了六个独立的混合逻辑模型。进行了一系列对数似然比检验，以验证按光照条件和区域类型划分的这六个独立模型是合理的。模型结果表明，每个模型中包含的变量的组合和影响程度都存在重大差异。一些变量仅在一种光照条件下显著，而在其他条件下不显著。同样，一些变量在一种区域类型中显著，而在其他区域类型中不显著。这些差异表明，不同的光照条件和区域类型实际上对涉及卡车的碰撞事故中的伤害严重程度有不同的影响，进一步凸显了根据农村和城市道路的光照条件来研究碰撞事故的重要性。乘员的年龄和性别（碰撞中受伤最严重的人）、卡车类型、年平均日交通量、速度和天气状况被发现是对涉及卡车碰撞事故中的伤害严重程度有显著不同影响水平的因素。
