| Continuous increases in travel demand, coupled with the limited capacity of transportation systems, have given rise to massive traffic congestion and loss of productivity due to wasted time in traffic. Unfortunately, capacity expansion is not always a practical solution because of limited funding and the latent demand that would be generated. Thus, there is a need to manage congestion directly, and in recent years, there has been considerable interest in modifying traveler behavior through information provision.; This dissertation focuses on how Advanced Traveler Information Systems impact congestion in transportation networks. In the first part of the research, we evaluate route guidance systems that drive user behavior toward system optimality. We perform a parametric examination of system performance as a function of the fraction of the population that follows system optimal routes. We prove that there exist a critical mass fraction, below which no benefit can be derived from route guidance systems, and a threshold fraction, which represents the minimum route guidance subscription rate needed to realize the full benefit of route dictation.; Research and practical experience have shown that travelers use several travel choice criteria, such as travel time, distance, number of vehicles on the road, and habit. This being the case, we propose a model in which travelers minimize their travel time and some measure of the quality of travel, weighted by the travelers' "value of travel discomfort." We again perform a parametric examination of network-wide congestion level as a function of the distribution of the travelers' value of travel discomfort.; The last part of the dissertation examines the impacts of real-time information on the performance of transportation networks. The dynamics of traveler behavior will be examined using information updating and boundedly rational models of route choice. A simulation program has been written to evaluate the effects of information provision on different driver groups in terms of improvements in travel time. The results highlight the relationship among the level of market penetration of traveler information systems, the quality of the information they provide, the expected number of incidents in the network, and the pre-existing congestion levels in the system. |
