基于生存分析的信号交叉口非机动车穿越行为研究

发布时间：2018-06-18 02:04

本文选题：信号交叉口 + 闯红灯　；参考：《北京交通大学》2014年博士论文

【摘要】：机动车与非机动车的混合交通是当前我国城市交通的主要特征,也是造成我国城市交通拥堵和事故频发的一个重要原因。非机动车出行灵活、准时性高,是解决中短距离出行和接驳换乘的理想交通方式,作为符合我国国情并拥有广泛群众基础的代步工具,在现阶段依然具有不可替代性。以前的非机动车主要指人力自行车,近年来随着技术的发展,电动自行车的使用已越来越普遍。与人力自行车相比,电动自行车可到达的距离更远,速度更快,但伴随的交通隐患也更高,这给我国城市交通带来了一些新的问题。此外,非机动车是一种健康、环保、低能耗的出行方式。发展城市非机动车交通是预防和缓解交通拥堵、减少大气污染和能源消耗的重要途径之一,关系人民群众的生产生活和城市可持续发展。然而,城市交通中,非机动车是一个相对弱势的群体,涉及到非机动车的交通事故比重一直居高不下。非机动车交通事故中,闯红灯违规是导致事故发生的主要原因,由于较差的法律约束和人们较低的安全意识,当前非机动车的闯红灯行为在我国极为普遍。目前,研究信号交叉口非机动车闯红灯行为的文献很少,更鲜有文献运用生存分析方法来分析非机动车闯红灯行为。生存分析的优点是可以考虑删失数据,并能将事件的结果和出现此结果所经历的时间结合起来分析,非常适合用来研究信号交叉口闯红灯行为。据此,本论文结合当前我国交通的特点,以非机动车为主要研究对象,基于生存分析方法,重点分析城市干道信号交叉口非机动车的穿越行为,揭示信号交叉口非机动车骑行者的违规风险、等待忍耐时间及其关键影响因素,并对特殊的管控措施进行评价。具体来讲,本论文的主要工作如下： (1)非机动车穿越行为的实证调研。通过选取典型城市干道信号交叉口对非机动车穿越行为进行实地拍摄,获取基础数据。分析非机动车和行人穿越信号交叉口过程中的等待时间分布、等待区域空间分布、运行轨迹、运行速度、穿越间隙等,揭示信号交叉口自行车、电动车与行人的微观行为差异。结果表明电动车的违规率明显高于行人和自行车；来自两侧非机动车的违规率明显高于直行到达的；行人不等待直接违规的比例显著低于自行车和电动车；违规者中行人的等待时间明显比自行车和电动车长,电动车运行速度比自行车和行人的快,而穿越的安全界限则比自行车和行人小。 (2)非机动车骑行者的等待忍耐时间分布规律。基于实证调研数据,建立信号交叉口非机动车穿越前的等待忍耐时间持续模型,对红灯期间到达的非机动车在穿越前的等待忍耐时间进行估计,探索不同等待时间下非机动车闯红灯违规率的分布。结果表明随着等待时间的增加,非机动车的违规概率逐渐增大,18.2%的骑行者几乎不等待就直接闯红灯违规；20.6%的骑行者愿意等待时间120s,甚至更长。并指出对删失数据的不当处理会导致明显高估非机动车骑行者的闯红灯违规率。 (3)非机动车闯红灯行为的Cox风险模型及其影响因素分析。构建非机动车骑行者等待忍耐时间的Cox风险模型,运用调查数据对模型参数进行估计,系统地分析各个潜在因素对骑行者的违规风险和等待忍耐时间的影响。结果表明交通方式、等待位置、高峰期、从众行为和机动车流量等对非机动车闯红灯行为有显著影响。电动车骑行者比自行车的违规风险更高,愿意等待的时间更短；等待位置越靠前(靠近路)违规风险越大；平峰期的违规风险大于高峰期；正在违规的人越多,机动车流量越小,则非机动车骑行者越容易违规,等待时间越短。并指出建立的Cox风险模型可用来预测或评估这些交通运营、管理和政策的变化对非机动车闯红灯行为的影响。 (4)非机动车通勤者的安全穿越可靠性建模与分析。运用可靠性思想和加速风险模型理论,结合非机动车穿越问题,建立非机动车通勤者的安全穿越可靠性模型,基于实证数据,找出最优可靠性模型的数学形式,并揭示影响通勤者安全穿越可靠性的关键因子。进一步把非机动车通勤者分为等待者和不等待者两类,分别探讨了他们的安全穿越可靠性问题。结果表明在各个潜在影响因素中,骑行者的等待位置、来自方向和从众行为等行为特征因素是影响非机动车通勤者安全穿越可靠性的最主要因素；Gompertz模型最适合用来拟合非机动车等待人群的安全穿越可靠性问题。 (5)交通协管的管控效果评价。通过运用Logistic模型、方差分析、协方差分析和生存分析等方法,比较信号交叉口有无交通协管时非机动车的穿越行为特征、闯红灯违规率和等待忍耐时间的差异,据此来评价交通协管对非机动车闯红灯行为的管控效果。结果表明交通协管对闯红灯行为有显著影响。有协管时非机动车和行人的闯红灯违规率都显著低于无协管的情形,有协管时的等待忍耐时间则比无协管时更长；不同来自方向中,交通协管对直行人群的闯红灯行为具有很好的管控效果,而对来自左右两侧人群的管控效果相对较弱。
[Abstract]:The mixed traffic of motor vehicle and non motor vehicle is the main feature of urban traffic in China. It is also an important cause of traffic congestion and frequent accidents in China. Non motor vehicle is flexible and punctual. It is an ideal way to solve short distance travel and transfer. With the development of the technology, the use of electric bicycles has become more and more common in recent years. Compared with the human bicycle, the electric bicycle can reach far distance, faster speed, but the accompanying traffic danger is higher, This has brought some new problems to urban traffic in China. In addition, non motor vehicles are a healthy, environmentally friendly and low energy consumption mode. The development of urban non motor traffic is one of the important ways to prevent and alleviate traffic congestion, reduce air pollution and energy consumption, which is related to the production and life of the people and the sustainable development of the city.
However, in urban traffic, non motor vehicles are a relatively weak group, and the proportion of traffic accidents involving non motor vehicles has always been high. In non motor traffic accidents, violation of red light is the main cause of the accident. Due to poor legal constraints and people's lower safety awareness, the current non motor vehicles break the red light line. It is very common in China. At present, there are few literature to study the behavior of non motor vehicles running red lights at signal intersection. There are few literature to use survival analysis to analyze the behavior of non motor vehicles running red light. The advantage of survival analysis is to consider censored data and to combine the result of the event with the time experienced by this result. According to the characteristics of traffic in our country, this paper takes non motor vehicle as the main research object, and based on the survival analysis method, this paper focuses on the analysis of the non motor vehicle crossing behavior of the signal intersection of the urban trunk road, and reveals the violation risk of the non motor vehicle riders at the signal intersection. Endurance time and its key influencing factors, and evaluation of special control measures.
(1) the empirical investigation of non motor vehicle crossing behavior. Through selecting the typical urban trunk road signal intersection to take the non motor vehicle crossing behavior on the spot, obtain the basic data, analyze the waiting time distribution in the process of the intersection of non motor vehicle and pedestrian crossing signal, wait for the distribution of the space, the running track, the running speed, the crossing gap and so on. The microscopic behavior differences between bicycles at signalized intersections, electric vehicles and pedestrians were revealed. The results showed that the violation rate of electric vehicles was significantly higher than that of pedestrians and bicycles; the violation rate from both sides of non motor vehicles was obviously higher than that of direct traffic; the proportion of pedestrians not waiting for direct violation was significantly lower than that of bicycles and electric vehicles; Compared with bicycles and electric cars, electric vehicles run faster than bicycles and pedestrians, and the safety limits crossing them are smaller than bicycles and pedestrians.
(2) the non motor vehicle riders wait for the time distribution of endurance. Based on the empirical research data, the waiting time model of waiting time for non motor vehicles crossing the signal intersection is established, and the waiting time of waiting time for non motor vehicles arriving at the red light during the crossing is estimated, and the rate of violation of the red light violation rate for non motor vehicles under different waiting time is explored. The results show that the probability of violation of non motor vehicles increases with the increase of waiting time, and 18.2% of the riders hardly wait for the red light violation. 20.6% of the riders are willing to wait for the time 120s or even longer. Regulation rate.
(3) the Cox risk model of the non motor vehicle running red light and its influencing factors. The Cox risk model of the non motor vehicle rider waiting for the time of patience is constructed, the parameters of the model are estimated by the survey data, and the effects of the potential factors on the cyclist's risk of violation and the waiting time are systematically analyzed. The results show the mode of traffic, The waiting position, the peak period, the crowd behavior and the traffic flow of the motor vehicle have a significant influence on the red light behavior of the non motor vehicle. The bicycle rider has a higher risk of infraction than the bicycle and is willing to wait for a shorter time; the longer the waiting position is near the Road, the greater the risk of violation; the risk of violation of the irregularities at the peak period is greater than the peak period; the people who are in violation are violating. The more the less traffic, the smaller the traffic, the easier the non motorist is to violate, the shorter the waiting time. It is pointed out that the Cox risk model can be used to predict or evaluate the effects of these traffic operations, management and policy changes on the behavior of non motor vehicles running red light.
(4) modeling and analysis of safety crossing reliability of non motor vehicle commuters. Using reliability thought and accelerated risk model theory and combining non motor vehicle crossing problems, the safety crossing reliability model of non motor vehicle commuters is established. Based on the empirical data, the mathematical form of the optimal reliability model is found out, and the safety wear of commuters is revealed. The key factor of the more reliability is to further divide the non motor vehicle commuters into two categories: the waiting and the non waiting ones. The problem of their safety crossing reliability is discussed respectively. The results show that the waiting position of the rider, the direction and the herd behavior are the factors affecting the non motor vehicle commuters. The most important factor in the reliability of full traversing is the Gompertz model, which is most suitable for fitting the safety traversing reliability of non motor vehicle waiting crowd.
(5) the evaluation of the control effect of traffic co management. By using Logistic model, variance analysis, covariance analysis and survival analysis, we compare the characteristics of non motor vehicle crossing behavior at the signal intersection without traffic coincident, break the violation rate of the red light and wait for the time of endurance, and evaluate the traffic coordination tube's behavior to the non motor vehicle running red light. The results show that the traffic coordination tube has a significant influence on the behavior of the red light. The violation rate of the red light of non motor vehicles and pedestrians at the time of cobing is significantly lower than that of the non cotube, while the waiting time of the cotube waiting for patience is longer than that without the cotube. Good control effect, but on the left and right sides of the crowd control effect is relatively weak.
【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：U491.225

【参考文献】