Citation

Haight FA, Bisbee EF, Wojcik C. Highw. Res. Board bull. 1962; 356: 1-14.

Copyright

(Copyright © 1962, National Research Council (U.S.A.), Highway Research Board)

DOI

unavailable

PMID

unavailable

Abstract

As roads and highways become capable of carrying higher traffic volumes, the perturbations introduced by vehicles traveling at speeds or in paths that differ from the norm become increasingly harmful to safe and efficient operation of the road network. Some of this individual variation can be removed, or diminished, by sensible efforts to educate and control drivers. The necessity remains for accelerating, decelerating, weaving, and merging. Perhaps the most important example of such a situation is the freeway on-ramp and acceleration lane. At these points, which must be provided fairly frequently in urban areas, the smooth flow of traffic is perpetually harassed by new arrivals. Location and design of on-ramps and acceleration lanes are closely connected with the influence they exert on traffic stability. A complete mathematical model for merging cannot be claimed, but it is hoped instead to point out the problems in formulation of such a model, and solve a few of them. The merging problem has some interest beyond the simple question of waiting for a suitable gap in traffic. For example, a car traveling along an acceleration lane while waiting for the opportunity to merge is mathematically equivalent to a car waiting at a stop sign, or the difference resides only in the moving coordinate system. However, the driver on the acceleration lane is able to control the traffic stream with which he wishes to merge by changing his own speed, thereby increasing or decreasing his headway and spacing relative to the main stream. The stop sign problem does not contain this important ingredient, and therefore questions of driving policy do not arise. There is only one possible policy at a stop sign: Wait for a suitable gap. Therefore, a mathematical model for a stop sign is purely descriptive and its principal result consists of a probability distribution for delay. There is a much more varied and interesting collection of problems available when the driver is allowed to alter (within limits) his attitude with respect to the main stream.