Master of Science in Civil Engineering (MSCE)
Civil and Environmental Engineering
Microscopic simulation models are more and more widely used to support real-time control and management functions in the field of transportation engineering. However, even with today¡¯s advancement in computing power, microscopic simulation modeling remains a computationally intensive process that imposes limitations on its potential use for modeling large-scale transportation networks. While microscopic features of a simulated system collectively define the overall system characteristics, it is argued here that the simulation process itself is not necessarily free of redundancy which, if reduced, could substantially improve the computational efficiency of simulation processes without substantially compromising the overall integrity of the simulation process. The idea of this research is to explore the concept of scalability for microscopic traffic simulation systems in order to improve their computational efficiency and cost-effectiveness. In an attempt to strike the balance between simulation performance and computational resources, we present an optimized downsampling procedure to transform the full-scale simulation system into an equivalent reduced-scale system. The primary goal of this research is to maximize the fidelity to microscopic simulation properties while maintaining the same macroscopic properties, such as flow rate, speed, and density. Experimental analysis was conducted on a homogeneous freeway corridor to examine the behavioral scalability of sophisticated nonlinear car-following models. A methodology to address lane-changing scalability is also included in this research study.
Document Availability at the Time of Submission
Release the entire work immediately for access worldwide.
Zhang, Yan, "Scalability of car-following and lane-changing models in microscopic traffic simulation systems" (2004). LSU Master's Theses. 3677.