Highlights • We propose a combined model by joint link-based credit charging and capacity improvement. • We try to manage the demand by optimizing credits charging for using the links. • We try to expand the whole capacity of a road network by increasing the road capacity. • The experiments show that this model is efficient in alleviating traffic congestion.

Abstract In this paper, we will discuss the urban road network improvement problem from both supply and demand sides, and propose a bilevel programming model considering joint optimal link-based tradable credit charging scheme and road capacity improvement. The upper level decision-maker tries to minimize the total system travel time under a budget constraint by optimizing both link-based credit charging and road capacity improvement, whilst at the lower level considering the users’ route choice behavior through the generalized travel time including the travel time and the converted time from the value of credit charging for using the link. Therefore, this proposed model integrates the improvement of the urban road network according to improving the road capacity with the given budget constraint and decreasing the travel demand with the tradable credit scheme. After presenting a relaxation algorithm, the numerical experiments on the nine node network are illustrated. Analysis shows that the proposed model is efficient in mitigating traffic congestion according to the less total system travel time than the other ways compared in this paper. The tradable credit scheme offers the better combination of cost-effectiveness, administrative flexibility and distributional fairness comparing with congestion pricing. Moreover, this tradable credit scheme is revenue neutral.