Narrow Results

Space–time correlation analysis has become a basic and critical work in the research on road traffic congestion. It plays an important role in improving traffic management quality. The aim of this research is to examine the space–time correlation of road networks to determine likely requirements for building a suitable space–time traffic model. In this paper, it is carried out using traffic flow data collected on Beijing’s road network. In the framework, the space–time autocorrelation function (ST-ACF) is introduced as global measure, and cross-correlation function (CCF) as local measure to reveal the change mechanism of space–time correlation. Through the use of both measures, the correlation is found to be dynamic and heterogeneous in space and time. The finding of seasonal pattern present in space–time correlation provides a theoretical assumption for traffic forecasting. Besides, combined with Simpson’s rule, the CCF is also applied to finding the critical sections in the road network, and the experiments prove that it is feasible in computability, rationality and practicality.

Traffic congestion has become a major problem restricting the development of major cities. ITS (Intelligent Transportation System) can record the state of traffic and predict the future traffic state, then reasonably optimize the travel scheme, so as to achieve the purpose of alleviating traffic congestion. Meanwhile, traffic flow prediction can provide data support for ITS, so many researchers have done a lot of research on traffic flow prediction. Many researchers take the traffic network as an undirected graph, and use the GCN (Graph Convolution Network) model to study the traffic flow prediction, and have achieved good prediction results. However, the traffic network is directed, and the traffic network is regarded as an undirected graph, which loses the direction information of the road network. Therefore, this inspires us to propose a graph convolution operator DGCN (Directed GCN), which can make full use of the in degree and out degree information of each station in the traffic network. The experimental results show that the graph convolution neural network based on this operator has better prediction accuracy than the state-of-the-art models.

Considering both the high complexity of urban traffic flow systems and the bounded rationality of travelers, providing traffic information to all travelers is an effective method to induce each individual to make a more rational route-choice decision. Within Advanced Traveler Information System (ATIS) working environment, temporal and spatial evolution processes of traffic flow in urban road networks are closely related to strategies of providing traffic information and contents of information. In view of the day-to-day route-choice situations, this study constructs original updating models of the cognitive travel time of travelers under four conditions, including not providing any route travel time, only providing the most rapid route travel time, only providing the most congested route travel time, and providing all the routes travel times. The disaggregate route-choice approach is adopted for simulation to reveal the relationship between the evolution process of network traffic flow and the strategy of providing traffic information. The simulation shows that providing traffic information to all travelers cannot improve the operational efficiency of road networks. It is noteworthy that an inappropriate information feedback strategy would lead to intense variation in various routes traffic flow. Compared with incomplete information feedback strategies, it is inefficient and superfluous to provide complete traffic information to all travelers.