Design And Implementation On Passenger Volume Statistical Analysis And Forecasting System Of Rail Transport

In recent years, with the development of the economy of our country and theaccelating of the urbanization process, urban traffic congestion conditions and thedeterioration of urban environment issues are becoming increasingly serious and graduallyleading to some kind of social problems. Conventional ground transportation has beenincreasingly unable to satisfy the increasing traffic demand, so the development of railtransport which has the name of “green transportation” has become an inevitable choicefor city transport. China’s urban rail transit starts late. By analyzing the rail trafficoperating conditions of several domestic cities constructing rail traffic earlier, we find thatthere is a common problem among these cities, which is the inconsistency of actualpassenger and forecast passenger——mainly performed as shortage of passenger volume.Because of the large investment scale and the high operating cost of the contruction of railtransit, the inaccuracy of passenger volume forecast has led to serious economic problems.For the problem of passenger volume shortage commonly found in rail transportoperation, this paper takes Wuhan Rail Transit Line One as an example and works on thecollection and clearing of the IC card transaction data and the entry and exit staticstics andanalysis of different types and different properties for the passenger data. Statistical typesinclude year, month, week, day and hour; Statistical propertities include maximum value,minimum value, average value and total value. Finally, on the basis of enough statisticaland analytical results, the paper used the Gray Forecast Model to predict the futurepassenger volume of Light Rail Line One. This article first does statistics and analysis onthe passenger volume data of actual operations of light rail. Then the Gray Forecast Modelused the actual passenger volume information to construct the forecast model, do residualtest, correlation test and back-residual test and compare the inspection error to thestandard value to decide whether residual correction is needed. At last, when the error iswithin the normal range, use the model to predict the passenger volume in the future. Compared to traditional passenger flow prediction algorithm, the Gray ForecastModel eliminates the time and energy needed for extensive survey operation.Theconstruction of the model is based on real statistical and analytical results of passengervolume. Forecast model is used to predict the annual passenger volume of the line, thetotal passenger volume of each site, the total monthly passenger volume of each site andthe early peak hour of each site. The forecast results play an important role in realizing thereasonable allocation of resources and rational urban transport planning. Due to timeconstraints, the achievement of system function is not perfect and there still exists somedeficiency, so further research and implementation is still needed.