This report documents results of work completed at San Diego State University during the first year of a research project intended to validate a behavioural traffic-flow theory recently proposed by Daganzo. Work on this project completed at the University of California at Berkeley will be documented in a separate interim report. In the idealised form developed so far, Daganzo’s theory assumes two types of drivers, aggressive (rabbits) and timid (slugs), and two lane groups, shoulder lanes and passing lanes. In free flow, rabbits travel faster than slugs and the two groups are segregated in what Daganzo terms two-pipe flow, with the slugs all in the shoulder lane and the rabbits all in the passing lane. In high-volume uncongested flow, rabbits will follow one another with very small headways, so long as they are able to pass, and such drivers are referred to as being motivated because the very small headways are held to be motivated by the desire to pass slower vehicles. Whenever an event occurs that reduces speed in the passing lane to (or below) vf, the free-flow speed in the shoulder lane, the rabbits change lanes to equalise speeds, lose their “motivation,” and increase their headways. This results in one-pipe flow. Because of the difference in the headways, maximum uncongested flow rates in the passing lane exceed the queue discharge rate for this lane. Daganzo provides solutions for a number of cases involving transitions from two-pipe to one-pipe flow and vice versa. These solutions are presented for the most part in terms of wave speeds and traffic states in, upstream, and downstream of queues. (Author/publisher)
Abstract