摘要
忍耐时间是分析电动自行车骑行者不安全过街行为的重要参数,是信号交叉口交通管控的重要约束条件.以红灯期间到达信号交叉口处违章过街的电动自行车为研究对象,应用多元线性回归分析和生存分析法中的Cox回归方法,统计分析骑行者在信号交叉口处的过街忍耐时间.该方法运用中国电信的"全球眼"网络视频监控技术,获取不同影响因素下宁波市电动自行车的实时视频数据,共采集了57 213个电动自行车过街忍耐时间样本.统计发现:电动自行车忍耐时间与5种影响因素存在强相关关系;7种影响因素对电动自行车忍耐时间的影响程度差异性较大,其中天气、人行横道长度和有无交警执法等3种因素的影响程度最大,出行时间和时段的影响最小;电动自行车过街忍耐时间的均值为48.600 s,标准偏差为300.341.
The waiting endurance time(WET) is an important parameter to analyze the electric bicycle(ebike) rider's unsafe crossing behavior, which is also the important constraint of traffic control at intersection simultaneously. This paper focuses on the e-bikes crossing illegally who arrived at signalized intersection during red-light period. Moreover, the multiple linear regression analysis method and Cox regression method are statistical analyzed for the WET of riders at signalized intersection. The real-time video data of e-bike riders under different influence factors are collected by the Global Eye network video surveillance technology of China Telecom. A total of 57 213 e-bike riders' WET samples were observed in Ningbo,China. The results show that: there is a multiple linear regression relationship between endurance time and 5 influencing factors; seven influences factors are found to have significant impacts on e-bike riders' WET,and the influence degree of weather, crosswalk length and whether have traffic police factors are the maximum, while the travel time and time interval are the minimum; the mean value of e-bike riders' WET is48.600 s, the standard deviation is 300.341.
出处
《交通运输系统工程与信息》
EI
CSCD
北大核心
2017年第5期242-249,共8页
Journal of Transportation Systems Engineering and Information Technology
基金
浙江省公益技术应用研究计划项目(2016C33256)
浙江省自然科学基金(LY17E080013)
浙江省社会科学规划课题(17NDJC130YB
18NDJC107YB)~~
关键词
城市交通
忍耐时间
统计分析法
实时视频数据
电动自行车
urban traffic
waiting endurance time
statistical analysis method
real-time video data
electric bicycle