Analysis And Prediction Of The Impact Of Short-term Large-scale Activities On The Passenger Flow Of Surrounding Rail Stations

With the improvement of the national economic level,more and more large-scale activities are held in major cities.The activities lead to the concentration of people which may cause great pressure on the surrounding infrastructure.Relevant departments urgently need to solve this problem.Therefore,this paper studied the passenger flow of surrounding subway stations during the short-term large-scale activities.This research use the IC card data and event data,analysis of various types of short-term events surrounding rail site characteristic,the influence of the stadiums were affected orbit around a fixed site and 3 hours before the event,passenger flow peak flow of passengers for 1 hour prior to the start of the activity within 1 hour after the end of the activity site traffic surge,15 minutes to 30 minutes at the end of the activity peak passenger flow.In the prediction of passenger flow of railway stations,there are many influencing factors and a large amount of data,and the decision tree method has the advantages of fast prediction speed and high accuracy,and is often used for prediction.Therefore,this paper chooses it as the prediction model.Then,a gradient descent decision tree(GBDT)prediction model is constructed based on 12 influencing factors,including track station,date attribute,month,activity attribute,venue location,time period before and after the activity,week,weather,organizer,activity type,start time and number of participants.Through a short-term large-scale activity,the passenger flow of the surrounding railway stations was predicted.The comparison and verification analysis between the actual results and the predicted results of the model showed that the average prediction accuracy of the 15-minute passenger flow of the east bridge,east section 40 and tuanjie lake stations was 93.67%,90.76% and 89.61%,respectively,with a high prediction accuracy.The passenger flow into the station is affected by the small passenger flow base in some periods,and the forecast accuracy is respectively 80.68%,78.96% and 78.11%.The passenger flow of rail stations in the period affected by large-scale activities varies significantly from time to time,so it is necessary to develop corresponding emergency solutions for passenger flow of different intensities,so as to better cope with large passenger flow and reduce the investment of manpower and material resources.So the thesis finally by K-means algorithm will affect the time 15 minutes each rail site traffic into Ⅲ,grade Ⅱ or Ⅰ sudden large passenger With the improvement of the national economic level.