Demand Responsive Connector Route Planning Under Time-Varying Road Network

Trunk public transport can effectively solve the long-distance travel problem of residents,but at the same time,it will lead to too long walking distance of residents,and there is the "last kilometer" problem of residents’ travel.The Demand Responsive Connector is a new type of public transport that provides "door to door" personalized service for passengers according to their travel needs in low-density travel areas,and is an effective way to solve the "last kilometer" problem of residents’ travel.At present,most researches on the route planning of Demand Responsive Connector are carried out based on the assumption of constant vehicle speed.However,road network is time-varying and vehicle speed is dynamically changing.Therefore,it is very necessary to consider the time-varying road network factor when studying the route planning of Demand Responsive Connector,which can better meet the current situation and make the route plan more reliable.Firstly,in this thesis,based on the current research situation,the relevant basic theories of the research problem are analyzed.The definition and characteristics of Demand Responsive Connector are introduced,and the service mode,system components and operation process of Demand Responsive Connector system are determined.The definition and characteristics of time-varying road network are described,and the timedependent function based on travel speed is selected to describe the time-varying nature of the road network through the analysis of time-dependent functions.After a basic study of Demand Responsive Connector and time-varying road networks,the factors affecting the path planning of Demand Responsive Connector under time-varying road networks are analyzed in a targeted manner,and the path planning process is determined.Then,two major elements of the research problem,real-time demand and timevarying road network,are addressed.For the real-time demand problem,the periodic update strategy and path update strategy are determined;for the time-varying road network problem,the time-varying travel time calculation method of the road section is summarized with the time-dependent function,and considering the uncertainty of the vehicle travel time under the time-varying road network,the method of using the improved Dijkstra algorithm to find the shortest path of travel time between two demand points is proposed.On this basis,the time-varying travel time and time-varying travel distance parameters between demand points are introduced,and the initial path planning model and dynamic path optimization model with the objective function of minimizing the total cost including bus departure cost,travel cost,passenger ride time cost and time window penalty cost are constructed from the perspectives of both bus operating companies and passengers,and a genetic algorithm invoking Dijkstra’s algorithm is designed to solve the problem.Finally,the Qiujiawan rail station of Line 6 in Nanan District,Chongqing was selected as the interchange point for the case study.The validity of the model and solution algorithm is verified by combining the actual road network with Python programming for model solving.By comparing the results with those of the Demand Responsive Connector planning under the static road network,it is concluded that the Demand Responsive Connector route planning considering the time-varying road network is more in line with the actual situation,and can avoid vehicles passing through the congested section and improve the operation efficiency.