Abstract:

In order to find a scientific approach to the decision-making of toll rate, this paper proposes an optimal model of multi-class toll rate based on the bi-level programming theory for the road network consisting of toll roads and adjacent roads. The upper-level programming of the proposed model describes the objective of network management for the maximum social benefits, and uses the revenue of proprietors as the restraint. The lower-level programming adopts a multi-class multi-criterion traffic equilibrium model to simulate the route choice behavior of road users. By solving the proposed model, the optimal charge rate of each toll road for different types of vehicles and the corresponding traffic assignment results can be obtained. Moreover, the solution steps of the model via the hybrid genetic/simulated annealing algorithm are presented. The results of a numerical example indicate that the proposed model can effectively improve the efficiency of the entire road network and better implement the pay equity among road users with different vehicle types.

Key words: road pricing, optimal model, bi-level programming, traffic equilibrium, user heterogeneity