Research On The Forecasting Method Of Passenger Flow In Airport Land-side Connection Transport

In recent years,with the rapid development of the national economy,the demand for air transportation in my country has increased sharply,and the air transportation industry has also made great progress in the past ten years.The average annual growth rate of passenger transportation throughput exceeds 10%.While putting forward higher requirements for capacity expansion and operation management,it also brings more severe pressure to the airport’s internal passenger flow organization and landside transportation.At present,airports are generally faced with the fact that the landing side transportation system cannot fully meet the growing traffic demand,resulting in the problem of traffic congestion inside and outside the airport terminal and the flow of people.Knowing and predicting the passenger flow characteristics of each landside transportation mode of the airport in a timely manner can better organize and dispatch various landside transportation modes of the airport and improve the overall service level of the airport landside transportation system.Based on the analysis of passenger flow characteristics,this paper makes a short-term forecast of the continuous passenger flow of various landside transportation modes in Beijing Daxing International Airport,and based on this,simulates the landside transportation system of Beijing Daxing International Airport.The service level of the transportation mode is evaluated and targeted optimization suggestions are put forward.The main work is as follows:The paper firstly collects the data of arrival passenger flow,multi-transportation continuous passenger flow,etc.,and analyzes the overall change trend,time period change characteristics,passenger flow sharing rate of each landside transportation mode,and passenger flow under the influence of external conditions.The characteristic change law is analyzed,which provides a basis for establishing a short-term forecast model of passenger flow.Secondly,the principles of the ARIMA model and the LSTM model are expounded,and then the algorithms of the two models are implemented based on the airport express passenger flow data,SPSS software and Python language.The prediction accuracy of the two models was compared,and it was found that the prediction accuracy of the LSTM model was higher.On this basis,comprehensively considering the influencing factors such as holidays,time periods,weather,etc.,a LSTM-based landside transportation mode continuous passenger flow prediction model is established,which improves the accuracy of the prediction model.Thirdly,based on the data of on-the-spot investigation and passenger flow prediction,the micro-simulation of the operation state of the landside traffic system of Beijing Daxing International Airport was carried out by using the micro-traffic simulation software TESS NG.Through the simulation results,it is found that the current Beijing Daxing International Airport has problems such as serious queues in the taxi connection area and high load pressure at night during the peak passenger flow period,and identified the bottleneck sections in the road network.Finally,this paper selects 6 evaluation indicators such as travel time and travel cost to evaluate the service level of each landside transportation mode of Beijing Daxing International Airport,and puts forward targeted optimization suggestions based on the evaluation results.