Study On The Scheduling Problem Of Public Transit With Stochastic Demands

With the development of social economy and high speed growth of urban population, the travel demands of urban citizen is rapidly growing, which usually results in severe traffic congestion. As urban public transit has the advantages of larger carry capacity, relatively small investment and lower road occupancy rate, local governments and transport management departments have gradually realized that the key to ease the problem on urban transit is to give priority to the development of urban public transportation. The bus scheduling problem is the core content of urban public transit. Using a scientific and rational optimization management method of bus scheduling can effectively ease the severe traffic congestion, balance the benefits of the passengers and the bus companies, reduce passenger's travel cost and alleviate the urban traffic congestion. It also could increase the operating efficiency of public vehicles and improve the service level of public transport system.The bus scheduling problem contains two objectives, which are conflict. One is to minimize the number of trips, which concerns the operational cost of a bus company. Another is to meet the passenger's demands and improve the ride comfort. Acquiring the accurate and reasonable data of travel demand is the premise and base of bus scheduling. But the actual passenger data is constantly changing, which is difficult to obtain. To study the problem, a model of vehicle scheduling for public transit with stochastic demands is established, and some examples are given to test the model.First of all, this paper reviews the history of development of urban public traffic both in and out of China, analyzes the tendency of development, illustrates the important role of urban public traffic, research problems of current bus scheduling system that exist in our country today, and put forward the main content, methods and the significance of the research. Second, we analyze and simplify the bus scheduling problem, and give the reasonable assumption. Then a model of vehicle scheduling for public transit with stochastic demands is established. Finally, the model has been verified by some instances.