Subnetwork analysis : methodology and application
MetadataShow full item record
The focus of this dissertation is to create robust tools that enable efficient and comprehensive subnetwork analysis for Dynamic Traffic Assignment (DTA) and a microscopic simulation setting. A DTA subnetwork can potentially replace a large urban transportation network that experiences a change in only a small fraction of the whole network. However, DTA mainly uses Cell Transmission Model (CTM), which lacks many details provided through microscopic traffic simulation. Also, there is very little research done on the balance between the computational time and the subnetwork size. Computational time increases when using a larger subnetwork, but the simulated result is more similar to that of the entire network. Conversely, the computational time decreases when using a smaller subnetwork, but the simulated result might not replicated the entire network. Currently, extracting a subnetwork is a manual and time-consuming process, requiring an entire coded urban network in ArcGIS. Therefore, to overcome these shortcomings this study automated the process of extracting a subnetwork. Moreover, to further the transition between long-term and short-term traffic analysis, the study integrated a DTA simulator and a microscopic traffic simulator so that together they can assign traffic and provide detailed traffic result. This study also defined an appropriate sub-arterial size for the microscopic simulator, which is not the same as the size of the DTA subnetwork. Furthermore, this study analyzed several factors which significantly influence computational time, and developed optimization models to find the balance between the computational time and error resulting from sub-area size. Ultimately, this study developed two programs that can automatically extract a subnetwork from a regional DTA network, and automatically develop an identical subnetwork in a microscopic simulator from this DTA network of an appropriate size. The methodologies this study built promote the efficient analysis of traffic conditions and facilitate the implementation of advanced models that were previously limiting in terms of the amount of time required to compute results; also, the automatic tools this study developed will contribute to the depth and the breadth of dynamic transportation systems analysis.