Impact Of Pedestrian Violation Behavior On The Traffic Flow At Signalized Intersection

Signalized intersections are the bottlenecks of traffic flow on urban road network. While serving mixed traffic with conflicting objectives, the occurrence of jaywalking is quite frequent, particularly in heavily congested urban areas. Illegal jaywalking at signalized intersections is a widely spread problem. Based on observations at the signalized intersections in Beijing, more than 50% of pedestrians walk across a street under red light, which undoubtedly affects intersection capacity and consequently aggravates traffic congestion problems. The current traffic performance evaluation techniques only consider the normal situation(assuming all users obey traffic light).Assessing the effect of pedestrian illegal crossing under various traffic conditions can quantitatively demonstrate the gravely of problem and lead to better improvement strategies to solve the problem. Therefore, the researches for the effect on the traffic by pedestrians’ violation behavior on different conditions need to be deeply analyzed and quantified. The pedestrian violation, which has affected the traffic seriously, may be caused by road factors, traffic and pedestrians themselves. It is great value to analyze these factors and to see whether they have significant and direct effects on the pedestrian violation behavior. Therefore, to quantify the impact on the traffic efficiency, to explore the factors affecting the pedestrian violation behavior and to model them, are important for the traffic flow theory in the mixed traffic and useful to manage congested and signalized intersections in metropolitan areas.One of the obvious and significant effects by pedestrians on the traffic is the decline of vehicles’ speed. Thus the effects of the pedestrians are first evaluated from this point of view. Firstly, how the speed of traffic flow is affected by illegal pedestrian crossing is studied. The average operating speed at the saturation flow rate with and without pedestrian traffic is used as a direct measurement to evaluate the effect of pedestrians. Based on the statistical analysis, following conclusions are gotten: saturation flow rate is mainly affected by the position of pedestrian in the crosswalk and the distance between the vehicle and pedestrians, while the number of pedestrian has little influence on operating speed. Using the iterative significant analysis which increases analysis unit by a fixed increment, the degree of pedestrian influence is classified into four levels. The impact at each level is quantitatively studied. It is discovered that the speed of saturation flow could decrease by 20% at the most serious level. Finally, the speed-time profile at different level is fitted according to the investigated data.The traffic capacity is affected by pedestrians also. The start-up time, acceleration time and saturation flow rate with and without pedestrian traffic is used as a direct measurement to evaluate the effect of pedestrians. Then, the model of capacity influenced by pedestrians is established. Again by using the iterative significant analysis which increases analysis unit by a fixed increment, the pedestrian influence to the capacity is classified into four levels and the effect is quantitatively estimated at each level. The result shows that the saturation flow rate could be decreased by 14% at the worst case(the first level). Lastly, the relationship between vehicular speed and time-headway is analyzed based on the investigated data. Then, a model of speed-flow relationship(S-shaped curve) is developed, which is a power exponential function. Based on the characters from the model, the results of influence to the saturation flow by pedestrians are verified.Traffic delay is associated with many factors, such as signal phase, vehicle arrival rate, leaving rate. It is very difficult to make a comparison on the traffic delay with and without pedestrians’ effects. Based on the relevant researches and the saturation flow rate affected by pedestrians, the traffic delay affected by pedestrians is calculated, which takes the actual signal intersection as an example. The results show that: for the steady state delay, when the ratio of the vehicle arrival rate and departure rate(kd1) is greater than 0.9, the traffic delay affected by pedestrians increases significantly. The requirements to the vehicular arrival rate under the corresponding pedestrian effect level are calculated by the numerical analysis method proposed in this paper. Under oversaturated flow condition, the higher pedestrian influence level, the severely increased delay. The proportion of delay increase corresponds to the proportion of reduced saturated flow rate.In order to provide suggestions and measures to prevent and correct pedestrian violation behavior, which affects the traffic seriously, the factors related to this behavior are investigated, including the length time of red light, length of crossing, time-headway of vehicles, gender, age and waiting time of the pedestrians and others. Using the correlation analysis which gradually increases analysis time unit by univariate factor, the factors that directly and significantly affect the individual pedestrian violation behavior and the group violation are determined. Then the model of pedestrian violation behavior is constructed. For the individual pedestrian violation behavior, the direct and significant influence factors are time-headway of vehicles and the waiting time of pedestrian, and the result of the model reveals that: When a pedestrian waiting time reaches 45 seconds(t=45s), the possibility of the individual pedestrian’s violation behavior increases greatly. For the group pedestrian violation behavior, the direct and significant influence factors are the number of the pedestrians and the longest waiting time for the group pedestrians. When the number of pedestrians is 4 or the longest waiting time reaches 46 seconds, the possibility of group pedestrians’ violation behavior increases greatly.The conclusions obtained in this paper are valuable for a better management to the signal intersection. The speed-time profile of an “S” shape for vehicles from starting to reaching the stable saturation flow and the power exponential function between speed and time-headway can enrich the traffic flow theory. The correlation analysis which gradually increases analysis time unit by univariate factor and the iterative significant analysis which increases analysis unit by a fixed increment are good reference methods for the similar problems.