Research On The Optimization Of Bus Dispatching Based On Passenger Flow Forecasting

In recent years,the speed of road construction are wildly mismatched with the growth rate of urban motor vehicle quantities,as a result,the traffic congestion and environmental pollution have become important issues that restrict the healthy and orderly development of the city.Speeding up the construction and development of public transport system is one of the effective ways to improve the efficiency of road resource utilization and alleviate the pressure of urban traffic congestion.The public transport dispatching is the key to improve the operational efficiency and management level of the public transport enterprises.Scientific public transportation dispatching can improve the utilization efficiency of public transport resources,improve the quality of public transport service and improve the attractiveness of public transport to the public.The bus passenger flow is the basis and prerequisite for bus dispatching optimization.Based on the distribution characteristics of passenger flow in time and space,in this paper the periodic variation of passenger flow is studied on the basis of deep analysis of a large number of passenger flow data,and the passenger flow forecasting model is established.Firstly,the distribution of the bus passenger flow in time is regarded as the ordinal sample,and the ordinal clustering algorithm which is raised by Fisher is used to segment the passenger flow time.Then,on the basis of the time division,combined with the grey forecast theory,the bus passenger flow forecasting model is established to forecast the distribution of passenger traffic on one operation day and provide data support for bus dispatching optimization.The departure time of the bus is the key to justify whether the dispatching work is scientific and reasonable.In this paper a public transportation dispatching optimization model is established based on the minimum weight sum,which considers the operation cost of bus enterprise and the waiting cost of passengers as the objective function,the starting interval,the vehicle full load rate and the enterprise economic benefit as the constraint condition and the departure time interval as the decision variable.Based on the data of passenger flow,this model can be used to calculate the departure time interval for each period of the bus operation day.The dispatching optimization model is constrained,multivariable and nonlinear,which can be solved by the particle swarm optimization algorithm,and the concrete method of processing constraints by introduced penalty function in the process of solving is given.In this paper,the forecasting model is applied to the bus 45 road of Zhengzhou,and the optimal scheduling is carried out on the basis of the forecast data,and the scheduling scheme with different interests is obtained.The results show that focusing on different stakeholders will have different effects on the departure interval,the total social cost of the operation day and the waiting time of the passengers.The model can provides a reference plan for the dispatching person to set the starting timetable and adjust the traffic plan in the actual operation schedule of the bus.