Models For The Traffic Network Equilibrium With Link Capacity And Tradable Credits Constraint

The tradable credits system is a new form of the transportation demand management, which provides the way to alleviate the urban traffic congestion problem.Previous studies of tradable credits system, link capacity constraints and the bi-level programming are summarized in this paper. On the basis of the review, the target and content of this paper were proposed.Firstly, the paper outlined the traffic network assignment problems under the constraint of the tradable credits which is one of the transportation demand management system, it can be realized through the freely tradable rights for using the link. Economic principles of the tradable credits were analyzed in this paper by considering the externality of urban traffic congestion. Then the uniqueness of the tradable credits model was proved. Finally, a comparison of the tradable credits with the congestion pricing was implemented according to the equity and the applicability.In order to solving the tradable credit problem, there were two algorithm proposed in this paper:one was designed to find the best price of the credit while the other was to solve the best credits charge of each link. According to the margin cost between the user equilibrium and the system optimal, the credit charge amount of each link can be got as well as the total credit amount of the entire network. The extra penalty method was used to transfer the constrained problem to unconstrained problem which can be solved using the gradient projection algorithm. Programming with C#, the results proved the correctness of the algorithm.Another problem is about the congestion flows of the network. The link capacity constraint UE problem can be expressed as an equivalent mathematical model. Then the Enhanced Lagrange multipliers can be used to get the equilibrium link flows. Adding the link capacity constraint to the tradable credit system, a bi-level programming model under these two constraints was got. By replacing the lower level problem with its Kuhn-Tucker conditions, the bi-level programming model can be equivalently transformed into single-level nonlinear programming problem which is easier to solve. The conclusion showed that a rational credit charge policy can effectively alleviate the traffic congestion.