Calculation Model And Evaluation Of Bus Stops Information Based On The Bus Data And Road Network Data

Urban public transport is the main component of the transportation industry,is the inevitable product of urban traffic development,but also the important infrastructure for residents to survive,so "smart bus" is also an "intelligent city" an integral part.Bus traffic is the basis for deep excavation of traffic travel data and research of passenger travel patterns.Bus time is the traveler is most concerned about one of the traffic information.Therefore,the geographic information system and data analysis as a means to start the bus travel analysis and excavation work,combined with the bus data structure,to explore the passenger on and off station inference and bus arrival time prediction method,the urban traffic problem is active significance.Based on the comprehensive analysis of the applicability of the domestic and foreign research methods on passenger flow and travel chain,the advantages and disadvantages of the forecasting speed of the bus arrival time,combined with the characteristics of the data source and the human and financial resources,this paper puts forward a single trip chain as the research object,The establishment of multiple linear regression model to calculate the average traffic speed of the bus,the integrated instantaneous speed and the distance from the station to calculate the average speed of the forecast to predict the bus arrival time.Based on the AFC and GPS data of Shenzhen,this paper uses time matching and density clustering method to determine the passenger station.The concept of passenger travel behavior and regularity is introduced.Bus travel section is continuous,according to the next ride on the car to determine the location of the passengers off the station.Bus travel break the passengers,combined with the passenger frequency of the frequency of high frequency stations and bus routes downstream of the site to attract the right to determine the break when the passengers off the location of the possibility of passengers off the coordinates and design deduction passengers off the station algorithm.Furthermore,the road is segmented by the method of K nearest neighbor,and the multiple regression velocity model is used to estimate the average velocity of each section.According to the bus passenger getting off the station inference algorithm,the examples are analyzed and the results are calculated and the passengers in the downstream stations are calculated to get off the frequency and the passengers are analyzed at high frequency.The feasibility of the algorithm is analyzed according to the passengers’ The distance between the vehicle stations and the weight of each prediction point determines the accuracy of the prediction and is verified to indicate that the method is valid.According to the arrival time forecasting model to calculate the actual bus arrival time,through the comparison with the true value of the assessment,indicating that the error within a reasonable range.Using the calculation results of the average speed of the road section to establish the database,and the road patency level division and real-time visual expression,the conclusion is in line with the actual state.