Cox Regression模型在交通事件持续时间研究中的应用

有效的交通事件管理要求交通管理者全面了解交通事件的各种特征才能准确估计事件持续时间,从而及时地疏导交通拥堵。利用某高速公路应急指挥中心管理系统中记录的近3 a的交通事件持续时间数据,建立Cox Regression模型探索影响持续时间的危险因素并评价其作用强度和方向。研究表明,日夜、报警方式、事件类型、占用车道数、涉及车辆数、涉及死亡、救护车、牵引车、吊车、驳车、涉及货车等11项是交通事件持续时间的显著影响因素,因此,交通管理者对这些因素进行改善可有效提高交通事件管理效率和安全性。

Application of Cox Regression Model in Traffic Incident Duration

To make decisions more efficiently to reduce the impact of non-recurring congestions and delays due to traffic incidents,traffic managers must have a full understanding of various characteristics of incidents to accurately estimate incident duration.This paper utilizes a dataset containing nearly 3-year incident duration data,and adopts the Cox Regression model to analyze the effects of several factors on incident duration.The results show that 11 factors including day/night time,reporter type,incident type,number of lanes occupied,number of vehicles involved,fatality,ambulance,tow vehicle,crane,barge vehicle and truck,have significant impacts on incident duration.Thus,improvements to these factors are beneficial to the efficiency and safety of incident management.