HIERARCHICAL FRAMEWORK FOR REAL-TIME TRAFFIC CONTROL

With the availability of faster computers and communications systems in the traffic control environment and more reliable monitoring and control hardware, better real-time control of traffic should be possible. The intelligent vehicle-highway system program now being proposed and implemented by transportation agencies, practitioners, and researchers will (a) need better real-time control methods for effectiveness in dealing with vehicle traffic and (b) allow the implementation and effectuation of better real-time traffic control. A framework for a hierarchical design of a real-time traffic control system is presented. The goal of the design concept is to respond to and monitor the various stochastic components of the traffic process with appropriate controls, frequencies, and sampling rates. The design is based on the decomposition of the traffic control problem into decision subproblems defined over different time and distance horizons. At the highest level of the hierarchy, the component process is considered that describes how over extended periods of time, travelers become aware of travel times and delays associated with the routes of a network and equilibrate into making routine route choices. At the middle level of the hierarchy, the faster dynamics are considered, for example, those dealing with traffic flows and queues during rush hours or traffic accidents. At the lowest level, the second-by-second dynamics in the traffic process are considered: the stochastic behavior of individual drivers and their responses to traffic controls at individual intersections. The conceptual design of RHODES, a prototype hierarchical traffic control system being developed for the city of Tucson, Arizona, and a comparison of its envisioned attributes with existing systems are described.