This dissertation research proposes the development and evaluation of a new concept for high-coverage point-to-point transit systems (HCPPT). The system is a radically new operational scheme for mass transit that relies on a large number of small transit vehicles routed using advanced communication technology and real-time stochastic control algorithms. The defense presentation covers the three major contributions of this research. First, the details of the concept are described, outlining its features and the flexibility of potential operational schemes. Second, the real-time routing algorithms are presented. A strict optimization formulation and solution for such a problem is computationally prohibitive in real-time. The design proposed in this dissertation is effectively geared towards a decomposed solution using detailed rules for achieving vehicle selection and route planning. If real-time update of probabilities based upon modeling the future dispatch decisions is included, then this scheme can be considered as a form of quasi-optimal predictive-adaptive control problem. Finally, some numerical results obtained from a simulated case study in Orange County are shown. This study was carried out using a multi-purpose simulation platform developed in the context of this research. The final simulations required point-to-point vehicle simulation, which is not possible with off-the-shelf simulators. The simulation framework uses a well-known microscopic traffic simulator that was significantly modified for demand-responsive vehicle movements and passenger tracking. A brief demonstration (animation) of the simulation tool applied to the HCPPT scheme is also included in the presentation.
