Bus Dynamic Network Simulation And State Identification

The growing of the public transit system has brought complex operation and management problems, such as coordinated scheduling of large networks, transport capacity deployment, operational evaluation and other issues. Due to technical limitations, the previous research can just rely on the static bus line network for modeling and research, whether it is on the network structure or a variety of complex scheduling problems. The maturity and application of communication technology and internet technology make it possible to master the distribution of public transport vehicles on the network in real time. Based on this background, we abstracts and builds the dynamic network which reflects the dynamic distribution of public transport vehicles on the network in this thesis, and the distribution of the vehicles on the network is supported by the data. We apply complex network analysis to study the state recognition of public transit dynamic network through dynamic simulation. This paper gets the support of Chinese National Nature Science Fund:The complexity and Empirical Study of bus dynamic network for vehicle scheduling. (No.51308176). Specifically, our studies mainly cover the following aspects:Firstly, a dynamic network model of public transportation is build. In this paper, we construct a dynamic network model with taking the bus station as the network node. Besides that, the different position information of the vehicle at all times is considered as the connection edge elements in the model. Along with the change of the position of the vehicle on the network, the public transportation network presents different topology structures.Secondly, a public transportation network simulation system is constructed. We load a certain amount of road traffic and set signal timing and other traffic parameters to build the road network and public transportation network on the basis of Vissim and Matlab platform.Finally, we research homogeneity of public transit dynamic network with static traffic network and passenger flow network. The paper features parameter of complex network such as degree distribution, structure entropy. What’s more, we identify the state of public transit network through analyzing the distribution of buses on the network.We construct a more perfect simulation system and identify the distribution status of public transit vehicles on the network based on the bus dynamic network model in this paper. In addition, we study the change law of network degree distribution and structure entropy, as well fit the degree distribution of dynamic network static line network and passenger flow network at different time, to judge transit buses’state in terms of global network. Moreover, we analyze the degree variation for all nodes and the match-degree in the network, as well judge transit buses’ state from view of local station. In short, we present a new method of public traffic network model construction and the mechanism of state identification. This can shed light on a new research method for public transportation network identification and scheduling.