Estimation of urban scale network-wide emissions based on fundamental properties of the network

One of the main criteria for evaluating traffic control systems, especially in large urban networks, is vehicular emissions, since it is the major contributor of urban air pollution. Decision makers in the cities want to minimize fuel consumption and emission of pollutants from traffic in addition to managing average speed and delay in the network. In order to optimize these parameters, a consistent estimation of the network-wide parameters is essential. Since existing macroscopic emission models based on average speed cannot provide accurate estimates for large-scale networks, there is a need for reliable emissions estimation based on the traffic features of a network. In the first part of this study, in order to have a more reliable, network-wide emissions estimation, the Integrated Traffic Emissions Model (ITEM) has been introduced to integrate macroscopic traffic parameters that have a major influence on emissions with microscopically calculated emission factors. This mesoscopic emission model takes three aggregated traffic parameters, which include: number of vehicle stops, duration of time spent cruising, and the time spend idling, all expressed per Vehicle Mile Traveled (VMT), and multiply them with corresponding emission factors to estimate overall emissions of the network. In the next step of this study, the macroscopic connection between three aggregated traffic parameters needed for ITEM and the average density on an idealized ring shape model has been investigated. The Macroscopic Fundamental Diagram (MFD) is used to understand the existence of a robust relationship between the average flow with the average number of vehicles circulating on the network. An analytical approach has been proposed to estimate the macroscopic traffic parameters based only on fundamental properties of the network. In the final step, the proposed model and analytical approach have been tested on a more realistic grid network with heterogeneous traffic states across links of the network caused by turning movements. The results show that there is not only a robust and reproducible relationship between the aggregated traffic parameters and resulting total emissions with the average density on the network, but also the comparison of the analytical estimates with detailed calculations shows that the errors are in an acceptable range, especially for not completely jammed traffic conditions.Ph.D.Includes bibliographical referencesby Rooholamin Shabihkhan