Multi - Level Conventional Bus Regional Coordination Timetable

Under the background of multi-hierarchy transit network optimization, in order to make the public transit meeting the diverse transit demand better, a bus schedule optimization model was established. Which’s building was according to passenger flow space-time distribution characteristics after analyzing the passenger trip distribution characteristics. The within hierarchy timetable scheduling based on passenger demand forecasting and regional synergistic timetable for multi-hierarchy transit network optimization mode were constructed.Firstly, the main factors affecting the establishment of multi-hierarchy transit regional synergistic timetable was analyzed. Public transit passenger flow distribution characteristics and bus network structure were studied. The operation dispatching objectives and modes of multi-hierarchy transit were determined.Secondly, the methods of collecting and auditing bus passenger flow data were analyzed. The bus passenger demand was predicted by BP neural network algorithm and RBF neural network algorithm. The optimize process and constrain condition of single line within hierarchy timetable based on flow prediction were designed. What’s more, evaluation model was constrcted. Combined with the example, the result shows that comparing to the former timetable scheme, the total cost of timetable scheme based on BP neural network algorithm and RBF neural network algorithm saves 2.44% and 4.80% respectively.Thirdly, considering public traffic line network optimization convergence pattern, the departure interval and vehicle scheduling form were used as decision variable. From the concepts of the passenger trivel time cost and operation profit, a multi-objective optimization model was established, which considered the passenger comfort degree, collaborative constraint of headway and enterprise capacity etc. Integrated varieties of optimization algorithm characteristics, the genetic algorithm, the particle swarm optimization algorithm and the genetic particle swarm optimization algorithm were used to solve the model. On the practical problems the algorithm steps to solve the model were designed in MATLAB software.In the end, getting through the analysis of the example, in the algorithm aspect, model target can be effective convergenced. The accuracy and convergence efficiency of genetic particle swarm optimization algorithm were significantly higher than that of genetic algorithm and particle swarm optimization algorithm in this paper. In the respect of solving result, the multi-hierarchy public transit timetable solved by the genetic particle swarm optimization algorithm reduces 3.48%、5.47% and 8.42% total cost than the timetable based on the RBF passenger demand forecast that choose three kinds weight value. In practice, the decision of timetable scheme should be made depending on the weights representing the relative importance of the operation profit and the passenger time cost.