The development and control of traffic jams caused by incidents in rectangular grid networks.

Urban traffic congestion is becoming a central issue in transport planning. If the present growth in car ownership and use continues, traffic jams are likely to increase in frequency and extent, particularly within the central areas of major cities. Whilst it is important to study the impact of congestion in the field, there is an urgent need for a fundamental understanding of the causes of congestion and the way in which it propagates. But, although a number of control schemes for controlling traffic congestion exist, no comprehensive rationale for an effective dispersal strategy has been developed.
This research is mainly concerned with the properties of incident-induced traffic jams on rectangular grid networks, and possible measures for preventing and controlling them. The research investigates the underlying structure of such jams using a combination of theoretical and simulation models developed for this purpose. Using these models, gridlock is identified as a crucial stage in the evolution of traffic jams.
However, most conventional traffic management measures aim to increase capacity and hence postpone the onset of gridlock and are unsuitable when gridlock has already set in. This thesis develops several alternative strategies for protecting networks from gridlock and dissipating traffic jams once they have formed. The treatment focuses on the installation of bans at specific network locations. The bans come in two forms: turn or ahead. Turn bans are imposed on selected links to break gridlock cycles at the nucleus of the traffic jam. By contrast, ahead bans are implemented around the traffic jam envelope to reduce input into critical sections of the road.
The control strategies are tested extensively using the simulation model and as a result, some general control principles have emerged. These are not intended to be immediately applicable to real networks since they incorporate some simplifying assumptions. However, they point to certain characteristics of traffic jam growth and dispersal which would not be accessible in any other way.

A thesis submitted to Middlesex University in partial fulfilment of the requirements for the degree of Doctor of Philosophy.