Research Of Public Transportation Passenger Volume Prediction Model Based On Public Transportation IC Card Data

In recent years,with the continuous improvement of people's living standard,the number of cars in China gradually shows excessive growth.Traffic congestion,energy consumption increase,environmental pollution aggravation,travel inconvenience and other phenomena become more and more popular.And then it causes a series of social problems,such as frequent traffic accidents and increased travel costs.At the same time,the proportion of public transportation in most cities of our country can not reach the normal level.The development speed of urban public transportation is difficult to keep pace with social progress and to satisfy the demand of people's travel.In order to alleviate the aforementioned phenomena and solve the correlative social problems,it is imperative to develop intelligent public transportation from the strategy of building a resource-saving and environment-friendly society and from the angle of humanized management.This thesis relies on "Xinxiang urban public transportation intelligent application demonstration project" which comes from Ministry of Transportation,and makes an intensive study of the public transporation IC card records for2015 of this city.The purpose is to establish public transportation passenger volume prediction model.Consequently,it can provide auxiliary guidance for public transportation company to carry on a series of management and arrangement such as making a route plan which is supported by theory,improving the accuracy of intelligent scheduling application,arranging transportation capacity reasonably and saving operating costs.And it also can offer a high-quality transit service to citizens.The main research contents of this thesis are as follows:(1)Do research about the background,working mechanism,and future development of the public transportation IC card.The public transportation IC card records for 2015 of Xinxiang City,Henan Province are taken as an example to analyze public transportation IC card records.Four steps(extraction,screening,statistics,and integration)of data mining for public transportation passenger volume are proposed,and two public transportation passenger volume time series are integrated.(2)Conduct research on the basic theory of chaos.The chaotic characteristics of the two public transportation passenger volume time series are determined respectively.Via calculation,both are chaos time series.The phase-space reconstructions are conducted to the two chaos time series separately.(3)Establish the chaos-wavelet neural network prediction model and implement the prediction to the public transportation passenger volume of Bus 6 of Xinxiang City,Henan Province.