FCD-based Regularity Studies For Urban Expressway Congestion Triggered By Flooding Under Bridge

In recent years, the traffic congestion caused by the inclement weather like rainstorms and heavy snows has become increasingly frequent. The inclement weather has not only led to traffic congestion in a large scale, but also caused severe economic losses and casualties. It has been a new task for cities such as Beijing to mitigate the flooding caused traffic congestion. Nowadays, scholars and experts studied more on the recurrent traffic congestion, but relatively fewer on the traffic congestion triggered by the inclement weather by using floating car data (FCD). In this context, taking Beijing as an example, this thesis focused on the FCD-based multi-angle analysis on the flooding segments under key bridges, and summarized the evolution regularities of the flooding caused traffic congestion. After that, the thesis proposed an identification method for the flooding congestion segments based on three indicators, and built the propagation and dissipation speed model for the flooding caused traffic congestion.First, from the definition and characteristics of the traffic congestion, the thesis described the speed level division of road traffic operation, and the differences and similarities between recurrent congestion and occasional congestion. Then based on the existing studies on incident detection and congestion identification for traffic incident, the thesis analyzed the applicability of the traditional identification methods for flooding caused traffic congestions. Combined with the existing characteristics of traffic congestion, literature review is conducted on the influence of inclement weather to traffic congestion. After that, the research scope, objective and approch were determined.Secondly, after analyzing comparatively the traffic flow characteristics based on the FCD between the flooding days and the normal days, the phenomenon of FCD serious samples loss was found, and then the index of ratio of FCD samples was proposed. Meanwhile, by comparing the ratios of FCD samples of the flooding congestion segments with those in traffic incidents, traffic control, the result showed that the proportion of sample loss of west3rd ring in Beijing was nearly80%, and that of west4th ring reached to97%. Then through the analysis and calculation of the propagation and dissipation speed of the flooding traffic congestion, and the vehicle kilometers traveled (VKT) of the flooding congestion segments, the thesis summarized the temporal relationship between flooding depth and congestion spread speed, and the influence scope of congestion propagation.Finally, according to the influenced scope of the flooding traffic congestion, the speed and the spatial-temporal distribution of congestion segments, the thesis proposed an identification method for the flooding caused congestion segments based on the ratio of the FCD samples, the speed variations of traffic flow, and speed difference. Then the identification method was programmed and applied on the ArcGIS platform. After that, based on the comprehensive analysis of the traffic flow characteristics of the flooding days, the influence of road capacity reduction, the ratio of volume over capacity, expressway ramp and other factors, the propagation and dissipation speed model of flooding caused traffic congestion was developed. Meanwhile the thesis calibrated the parameters of the model by using VISSIM simulation data and the FCD of the flooding traffic congestions.