Research Into Dispatch Decision System Of Public Traffic Vehicle By Modern Non-linear Mathematics Method

With the high speed growth of both social economy and urban population, the demand of urban citizenry for travel in the same city is increasingly growing, which usually exceeds the supply of public traffic's capacity, and results in severe traffic crowding. How to solve this problem is important to each country, especially to the developing country, whose fiscal condition is so poor that it could not afford enough funds to construct new public traffic facility in the short run. The fact in our country is that both lacks of public traffic facility and poor management exist together. So, at present, the first task of our country, which tries improving the level of public traffic management, is more wise and realistic way.This essay starts from the above purpose, concentrating on researching on the bus dispatch system of public traffic, trying to make good use of current resources, and establish automatic dispatch system, based on intelligent technology. This essay can be divided into four major sections.The first section of the essay, reviews the history of development of urban public traffic both in China and in other countries, analyzes the tendency of development, illustrates the important role of urban public traffic, puts forward problems of current dispatch system of urban public traffic that exists in our country today, explains the reason that I select to solve this problem in order to prepare for my doctor degree, and states the main content, logic and methods in the research process, in addition to the significance of the research.The second section of the essay, refers to the modern nonlinear theory including neutral network and genetic algorithm, discusses the impact of structure parameters of neutral network hidden layers on the neutral network, and the engineering and technological applications of both neutral network andgenetic algorithm, gives a simple and practicable and effective method to supplement lost variables.The third section of the essay, analyzes the distributing law of passenger flow of public traffic, brings forward a new prediction method which could keep accordance to the practical distributing law of passenger flow by and large, furthermore, could meet the demand of optimizing bus dispatch system. This method based on intelligent technology breaks through the traditional way.The fourth section of the essay, takes into account the object of both traditional service function of public traffic and demand for profit, gives a new dispatch model. This section, based on the prediction method in the third section, researches on the passenger flow law of each station, and predicts possible passenger flow of each regular runs, which number amounts to the sum of passengers of each station who come to station in a interval on the given line, and finds out the congregation of optimized dispatch plans by using genetic algorithm, then, selects the best dispatch plan from the above congregation by using FAHP method to evaluate. In addition, this section also discusses the assistant countermeasures of dispatch system in public traffic.