Citation

O'Dea WP. Res. Transp. Econ. 1999; 5: 155-179.

Copyright

(Copyright © 1999, Elsevier Publishing)

DOI

10.1016/S0739-8859(99)80007-9

PMID

unavailable

Abstract

In this chapter, we evaluate the use of automated highways as a means of reducing peak period congestion costs. Their proponents argue that by transferring control of vehicles from drivers to a computerized control system, automated highways will enable cars to travel at high speeds in close formation. The perceived benefits are a reduction in travel time and an increase in highway utilization which would permit a given flow of traffic to be accommodated by a smaller investment in highway capacity. The Achilles heel of the system is safety. If one of the vehicles in the traffic flow is forced to stop, the potential for a catastrophic accident is clear. To evaluate the benefits of automated highways, we employ a version of the bottleneck congestion model which includes travel time downstream of the bottleneck. We assume that the highway authority controls the rate at which vehicles enter the roadway and the behavior of vehicles once on the roadway. If the rate at which vehicles arrive at the highway entrance exceeds the entry rate permitted by the authority, a queue of vehicles will build up. By varying the entry rate, the authority can trade off time spent in the queue for travel time on the road. In selecting the entry rate, the authority is constrained by the speed-flow curve governing the road. Since the authority controls the behavior of cars on the road, the speed-flow curve would approach an engineering ideal. Using a speed-flow curve based on engineering principles which takes safety into consideration, we show that, even if automated highways would not enable cars to travel at 100 mph in close formation during peak periods, they would enable rush hour traffic flows to be served with a smaller right-of-way at a sizable cost savings.