Loop data-based optimization methods for optimal traffic operations on freeways

Dramatically increasing travel demands and insufficient traffic facility have induced severe traffic congestion problems. Research and development of operational control and management strategies that optimize freeway traffic operations are essential to enhance traffic mobility and overall infrastructure efficiency. Rapid development of Intelligent Transportation Systems (ITS) has employed several advanced traffic detection technologies to provide a high-quality platform of data to support optimized control strategies. The predominant freeway detection comes from Traffic Monitoring Stations (TMSs) which consists of inductive loop detectors that cover each mainline and ramps. The goal of this dissertation is to develop ITS loop data-based optimization methods to manage the existing infrastructure more efficiently with advanced traffic control and management strategies.;Freeway loop data are paramount for traffic surveillance, control, and information systems for freeway management, yet little has been published on identifying or correcting the sources of data errors. This dissertation develops method to compare aggregated loop speeds with GPS probe vehicle speeds to investigate the inherent difficulty of the problem by identifying the error sources from loop and GPS probe vehicle speeds. Results show a weak relationship between loop and GPS probe vehicle speeds.;Further, the reliability and accuracy of loop data depends on the allocation and placement of loop stations. This research examines the effect of TMS spacing on travel time estimates and determines the optimal spacing of TMS. Results indicate that it is essential to deploy more TMSs to cover major bottleneck areas and fewer for free-flow regimes.;Another advanced traffic control strategy employed as an acceptable countermeasure to mitigate freeway congestion is the High Occupancy Toll (HOT) lane operation. However, literature on congestion pricing algorithm aimed at overall HOT lane operation optimization is still in early stage. This research develops an efficient feedback-based tolling algorithm that can systematically optimize traffic allocation between freeway lanes and enhances overall infrastructure efficiency. Results show an increase in HOV lane usage and average GP lane speeds by 90% and 25% respectively were achieved during peak periods. Overall, results show that the optimization methods developed in this dissertation have performed reasonably well in optimizing freeway traffic operations.