Rapidly increasing traffic volume, congestion, and excessive delay are making the management, control, and guidance of traffic flow one of the most critical transportation problems in urban freeway corridors. Modeling demand diversion to less congested routes within acorridor is a necessary part of demand modeling efforts for improved simulation and control, as are guidance-navigation systems in realtime. Models for describing diversion at the trip origin and diversion at freeway entrance ramps are discussed. Data collected in a major metropolitan area have shown that diversion at the origin is a function of trip time, route length, and the number of intersections along the trip. However, trip time is the dominant determining factorand can be employed to estimate the decision in the absence of additional information. Diversion at freeway entrance ramps depends on the perceived trip time on the freeway and arterial and the perceivedwaiting time at the ramp queue. The data confirm that socioeconomicindicators do not play a role in the diversion decision. The purpose of developing these models is for dynamic simulation, on-line freeway corridor control, and demand forecasting suitable for guidance and navigation. This paper appears in transportation research record no. 1220, Forecasting.
Abstract