Research On The Optimization Of Combinational Scheduling BRT Based On Real-Time Passenger Flow

BRT(Bus Rapid Transit) system is a new kind of public transport mode with high quality, high efficiency, low energy consumption, low pollution and low cost. It has been widely used in China and many other countries. The operational and scheduling methods are important factors of service capacity for those BRT systems that have been built. In order to attract more passengers and improve its competitiveness, convenient and comfortable services must be provided in BRT.The thesis started with the current research status of operation and dispatching in BRT and pointed out that the existing daily vehicle departure schedule was fixed and dispatching methods were simple. Thus, it failed to meet the passengers’travelling demands and limited BRT’s advantages. Therefore, after careful analysis of passenger distribution and real-time passenger information, the thesis proposed an idea to establish a combinational scheduling model for BRT vehicle scheduling so as to optimize the rational dispatching intervals and mode.Firstly, the factors influencing BRT’s combinational dispatching were deeply analyzed. Then, based on time and space, the distribution characteristics of passenger flow in BRT were studied. Then the steps to determine the scheduling methods of BRT were figured out Secondly, the structure and function of BRT tolling system were also probed into. The information collection process for BRT passengers was analyzed. A method based on BP Neural Network to forecast short-time passenger OD distribution was given. Then, a model was established here so as to lay a foundation for the following combinational dispatching model.Thirdly, the dispatching intervals and vehicle dispatching mode of regular buses, sectional buses and express buses were taken as decision variables. The object is to minimize the total cost of passengers’travelling and operational vehicles. Based on the constraint conditions such as the passengers’comfortablity, departure frequency that operators could bear and so on, a combinational dispatching model of BRT was established. Then a program with genetic algorithm was developed to solve the model, and the optimal vehicle dispatching time and dispatching schedule was obtained by solving the model.Finally, the real case of BRT line1of Jinan city was analyzed, obtaining the optimal combination of scheduling scheme based on Real-time passenger flow. In the example, the total cost from the model and the current dispatching modes were compared. The result indicated that7.4%of system’s total cost was saved by using the combinational dispatching modes. Optimal effect of the model was remarkable. The example is not only showing that the result of this model can meet the requirement,but also showing that the algorithm is feasible and effective.