Study On The Relationship Between The Fluctuation Of Nanjing Rail Transit Flow And The Rainfall Based On Operational Data

With the urban road traffic congestion problem has become increasingly prominent, the construction and reasonable operation of rail transport are particularly important. Since rail transit passenger flow is the basis of rail transit investment, and is also the basis for the decision of rail operation scheduling. In order to explore the influence of rainfall weather factors of rail transit passenger flow, and more accurately forecast and control the rail transit passenger flow, this paper, based on Nanjing rail transit passenger flow statistics, track credit card data and rainfall data study on rail transit throughout the day, time fluctuation rules of passenger flow and the change of different travel characteristics under the condition of rainfall weather.Fistly, through the analysis of the data structure and characteristics of rail transit operation data and weather data, data cleaning are conducted on the raw data considering data application requirements and storage criteria and rail transit passenger flow data are fitted using ARIMA model, the flow benchmark values were calculated when rainfall weather occurs, furthermore the rail transit full-time, time-sharing traffic fluctuations rate are obtained under the condition of rainfall weather.Secondly, the rainfall grade division based on passenger diurnal fluctuation rate is studied considering the traditional diurnal rainfall classification. On the basis of the analysis of the relationship between diurnal transit fluctuations and rainfall, rainfall weather with the diurnal fluctuation of passenger flow relation model is fitted, and then the laws of diurnal transit fluctuations in rainfall weather conditions are concluded.Thirdly, the rainfall classification based on time-sharing passenger flow volatility is studied considering the traditional time-sharing rainfall classification, then the paper analyses the passenger fluctuation in different periods when under different grades of rainfall, and summarizes the rail transit passenger flow fluctuation regularity, furthermore the rail transit time-sharing passenger flow concept model under the condition of rainfall weather is constructed.Finally, on the basis of the establishment of the Hadoop data platform, based on massive credit card data to identify "home-station" and "work-station"of traveler, and then determine travel purpose of credit card record, each trip purpose distribution characteristics are analysed and the influence of rainfall weather to each trip purpose corresponding rail transit full-time and time-sharing passenger flow are summarized.