An Urgent Traffic Dispersion and Assignment Model for Urban Road Flooding

Urban road flooding often causes road capacity reduction, traffic congestion and inconvenience in citizens daily travel. This paper proposes an urgent traffic dispersion and assignment model for the case of urban road flooding, with the aim of maximizing the level of service in the road network by controlling the road capacity. First, based on prospect theory and taking the historical travel time average as a reference, the model adopts the BPR (Bureau of Public Roads) function to represent the cost-flow relationship to calculate the prospect value of each route. Second, the travelers route choice behavior is described in logitmodel with the prospect value as the utility. Third, based on the route choice results, the traffic flow on the congested road sections is dispersed by controlling the road capacity, so the traffic flow to the flooded roads can be adaptively assigned to other roads. Finally, a direct iterative method and genetic algorithm are used to solve the proposed model. The former attempts to implement the traffic assignment based on the travelers route choice behavior, and the latter is used to find the satisfying solution through selection, crossover and mutation. The proposed model is applied to a given road network with an assumption of some road capacity reduction due to road flooding. The results show that when the proposed model is applied, the saturation ratio (or level of service) of the roads in the entire network is more uniform and the distribution of the saturation ratio of main roads is reduced, so the traffic flow in the whole network can remain smooth and the level of service can remain high.