The Bus Scheduling Model Research Based On Bus Arrival Time Prediction

With the continuous development of urban economy and society and the increase of urban traffic volume, traffic problems, especially traffic congestion in some large cities have become increasingly serious. People gradually realize that it is important of developing urban public traffic to solve urban traffic problems. Some cities have begun a large-scale subway construction, but the bus remains to be the main stakeholders of passengers. The management level of bus scheduling plays a decisive role in the quality of public transport operation. The development of intelligent public traffic system optimization has become the focus in today's society.Bus scheduling studied in this paper mainly aims to the research on planning method of the grid interval of public transport vehicles. In this paper, the method to predict changes of passenger flow based on the quasi-real-time passenger information and Kalman Filter-based bus arrival time prediction method are discussed. Namely, the paper applies public traffic IC cards and other related quasi-real-time information to predict short-term traffic flow, establishes bus scheduling optimization model based on multi-objective programming, and uses genetic-simulated annealing algorithm to optimize the operational scheduling parameters.The first chapter introduces the research background and significance, states the status quo at home and abroad, highlights the main research content and goals according to the recent status combined with the research direction, and gives the research the technical route.The second chapter describes the traditional bus scheduling and intelligent bus scheduling methods, according to the latter, proposes bus scheduling system based on real time passenger flow and bus arrival time prediction, and studies the key technologies.The third chapter analyzes the valid data processing in the database of public transport IC card, introduces the judgments of the get-off site based on the get-on site, and applies exponential smoothing method to make a short-term traffic forecast in a case and derives scientific evaluation.The fourth chapter introduces single-site bus arrival time prediction theory and multi-site bus arrival time prediction theory, predicts multi-site bus arrival times based on support vector machine, calculates bus arrival time by Matlab, and evaluate the results.The fifth chapter discusses the establishment of bus scheduling model based on multi-objective programming and solution steps of parameters in the model applied genetic-simulated annealing algorithm and checks the model in an example. It can be known from the results that bus operational scheduling optimization model based on multi-objective planning and genetic-simulated annealing algorithm are suitable for bus scheduling optimization system.