Research On Statistical Learning Methods For Subway Passenger Flow Forecasting

In recent years,with the acceleration of the urbanization process,the contradiction between the growing urban transportation demand and the limited transportation supply capacity has become increasingly prominent,resulting in major cities facing severe traffic congestion problems,the peak caused by residents 'travel behavior The phenomenon of congestion during the period was very common.Traffic congestion will not only bring great inconvenience to residents' travel,but also cause huge social and economic losses,which seriously restrict the development of the city.In recent years,urban rail transit is playing an increasingly important role in urban transportation.While effectively alleviating the pressure of road traffic,it is gradually facing the challenge of congestion.Therefore,for the interval passenger flow estimation and congestion prediction of the subway network,targeted research is needed.Reasonably planning and scheduling traffic by predicting congestion is an effective measure to alleviate urban traffic congestion.In response to the above urgent problems to be solved,this paper intends to use machine learning research methods to carry out the total incoming traffic in hours for the inbound passenger traffic data of a city subway from January 1,2018 to September 30,2018.Statistics,analysis and research of station passenger flow.First,based on the existing subway passenger flow data of a city,a descriptive analysis of several factors affecting the subway passenger flow of the city is carried out,and charts and other data are used to express the data more intuitively.Secondly,use statistical software to carry out modeling analysis based on time series analysis and neural network based modeling analysis of existing subway subway inbound passenger flow,and further carry out passenger flow prediction analysis,comparing the analysis of the two models After the results,select the model that is most suitable for the study of the actual situation of this article.The results of the study are of great significance to the optimization of individual travel processes and the planning and management of urban transportation.