Railway Station Traffic Flow Volume Production Based On SARIMA Model

With the development of contemporary economy,traffic modes of all kinds have made unprecedented progress,and people are more willing to travel.Among them,railway is the most important way for long distance travel of the residents in our country because of its convenience and low price.In recent years,the railway sector has taken various measures to improve railway passenger transportation and enhance the passengers riding experience.In particular,since the advent of EMU trains and 12306 online ticketing system,the experiences of ticket purchase,service,punctuality rate and other aspects have been considerably improved.In such a benign development trend,the railway sector is still facing the biggest problem in railway transport of passengers: the influencing factors of changes in passenger traffic volume are complicated,the fluctuation range is big,and it is therefore difficult to predict accurately via traditional methods.To deal with the above problems,put forward a more accurate prediction method for railway passenger flow volume,we select the whole passengers riding 435 days data of a railway to analysis and modeling to research a more accurate prediction method.In this paper,the original data are cleaned and standardized,and the valuable information such as the number of passengers on the train,the departure time,and the train number is extracted,which is then sorted into structured data to facilitate the analysis firstly.Secondly,the data are divided into two types: non-holidays and holidays.For periodic non-holidays data,periodic ARIMA model is added into traditional SARIMA model to predict,and for holidays data,fluctuation coefficient model is used to predict.Finally,the prediction accuracy of the model is evaluated by calculating the relative error between the predicted value and the real value.It has been verified that,the prediction on railway passenger flow achieves good results.During the non-holiday period,the average relative error of passenger flow volume is only 0.0758,and the one during the holidays is only 0.0914.Thus,it proves that the combination of SARIMA model and fluctuation coefficient model can predict railway passenger flow volume more accurately.We make an accurate prediction on changes in railway traffic flow volume in a short period of time,which can provide a reference for the railway sector to arrange reasonable dispatch and make full use of material and human resources,so as to avoid resources wastes or station chaos caused by insufficient preparation.