Congestion charge setting needs to take into account the reactions of transit operators, as these will influence traveler costs, which will in turn change the optimal congestion charge. This thesis proposes a model to explain at an aggregate and generic level the relationships between the three principal types of decision-making unit involved with congestion charging, namely the Government, the transit operator(s) and the travelers.
The proposed inter-modal equilibrium model links an urban road network subject to a congestion charge to a parallel urban transit market with two modes, bus and train, with a view to finding the optimum congestion charge consistent with the commercial decisions of the transit operator(s). A congestion charge is sought which maximizes social surplus. Travel behavior is assumed to conform to elastic-demand user equilibrium traffic assignment. The transit market is assumed to be either a profit maximizing monopoly or a profit maximizing duopoly consisting of a bus operator and a train operator competing non-cooperatively. The operator(s) set the fares to maximize profits and the supply of transit services are determined by the resulting demand. The choice between private and public modes is assumed to conform to a logit model, whereas the choice among transit modes is assumed to conform to either a logit model, whereas the choice among transit modes is assumed to conform to either a logit model or the user equilibrium principle. The problem has been formulated as a bilevel programme with the determination of the congestion charge on the upper level and the setting of transit fares on the lower level. In the case of non-cooperation operators, the Bertrand-Nash equilibrium fares are sought. This reveals the importance of competition in
06-Abstract.doc
the transit market for the trade off between the Government, the transit provider(s) and the travelers.
Various aspects of the model are examined. We compare cases where travelers have the same and different values of time. The importance of sensitivity to travel cost, which influences the decision to make a trip and the choice of mode, is also investigated. An increase in this sensitivity is equivalent to a reduction in the random component of utility, and may therefore be interpreted as synonymous with the effects of improving traveler information. Also investigated are fares which maximize welfare with or without a constraint that profits should be non-negative. Finally, the model is extended to incorporate the effect of stop sharing to assess the benefits of interchanges in a transit network. Conclusions drawn from the model provide useful guidance on congestion charge setting policy.