Multi-level Fundamental Diagram Of Traffic Flow For Beijing Based On Multi-Source Data

Fundamental Diagram (FD) model of traffic flow is a method that uses diagrams to describe flow-speed-density relationships, and is an important foundation for exploring traffic regulation and developing traffic management strategy. Due to the huge ownership, high concentration and intensive usage of vehicles in urban area, infrastructure construction cannot solve the issues of traffic congestion and pollutions in Beijing. Studies on the basic traffic regulations and analysis on network performance are critical issues to improve road and network performance. It is also the way to improve the level of service of the road network. However, due to the limitations of data collection techniques and data analysis method on urban traffic flow data, most existing studies of traffic flow fundamental diagram are based on the data collected on the segments of highway or expressway. In this context, the study in this thesis strives to develop fundamental diagram models on multi-levels of road segment, class-specific road network, and regional road network for Beijing based on floating car data, Remote Traffic Microwave Sensors (RTMS) data and screen line survey data.First, the research conducts the literature review on the classical microscopic and macroscopic traffic flow models and the relationships of traffic flow parameters on urban road. Subsequently, the research scope, objective and contents are determined.Secondly, this study discusses the collections method of these data and analyzes the data characteristics for the floating car data, RTMS data and screen line survey data. Furthermore, the method of data integration and multi-source data matching is designed in this study.Thirdly, the fundamental diagram of traffic flow for single segments is developed based on speed data from RTMS and FCD respectively. According to the comparative analysis of the two proposed model, it is found that the model based on the space mean speed data of the FCD can capture traffic flow characteristics on urban roads more accurately. Moreover, the fundamental diagrams of the traffic flow for various road types are constructed. The flow data are obtained by calculating the average value of multiple sections of the same road type and the corresponding travel speed data is provided by FCD. The traffic flow characteristics of different road types are analyzed quantitatively by comparative analysis of the traffic flow parameters.Then, the relationship diagrams between weighted flow, weighted density and average travel speed are constructed for different districts (Xicheng, Dongcheng, Haidian, Chaoyang and Fengtai District) in Beijing. Van Aerde single-regime flow model is applied to calibrate the characteristic parameters. The network performances of different districts are compared and evaluated based on the proposed macroscopic fundamental diagram models.Finally, the traffic strategy of low emission zone (LEZ) in Beijing is utilized as a practical application case of the proposed model. By using the fundamental diagrams of traffic flows on expressways, major arterials and minor arterials, the speed variations before and after the implementation of LEZ are estimated. It is therefore shown that the developed model can be used for evaluating macroscopic traffic management strategies.