The Research On The Rules Of Public Transport Passenger Flow Based On Data Mining

Public transport passenger flow is the basic data of urban transit planning, operation scheduling and organization. If the rule of public transport passenger flow could be acquired accurately on time, the public transport company can take appropriate measures to meet the demands of passengers, improve transit capability, cut down the economic cost and social cost. The bus IC cards generate massive operation data, so using these data can reflect the characteristic of public transport passenger flow accurately on time. On this account, this paper uses bus IC card data to make a research on the rule of public transport passenger flow, based on data mining technology.This paper focuses on the research of the rule of public transport passenger flow. Although the existing research method of transport passenger flow data is relatively mature and being used widely, there is still some disadvantage. The transport passenger flow data is collected by manual investigation. It costs a lot of money and time, and its validity can't be guaranteed. When deal with massive data, the existing mathematical statistical methods is helpless. This paper will make a research on bus IC card,combine the traditional mathematical statistical methods with state-of-the-art data mining method, and get the rule of public transport passenger flow and passenger flow index needed in public transport operation through the analysis and treatment on bus IC card data.According to successive and relativity of the data mining result, this paper firstly makes a research on the division of passenger flow peak interval, calculates the site and time of getting on and off the bus based on the result of the division; Secondly, based on the data mining result this paper already have, makes prediction on the number of people getting on and off the bus; At last uses mathematical statistical methods to solve the indexes of passenger flow, and summaries the rule of public transport passenger flow. After building the model of data mining based on bus IC card data, this paper chooses the real IC card data from public transport group, and makes a research on the model and algorithm. The experiment proves the applicability of data miming model base on bus IC card.The research on the rule of public transport passenger flow in this paper fuscous in there aspects:Firstly, the trip characteristics of passenger, including riding time and trip distance;Secondly, the timeliness rules of public transport passenger flow, including passenger flow periodicity and timeliness imbalances; Thirdly, the spatiality rules of public transport passenger flow, including passenger flow in the station, sectional passenger flow and spatiality imbalances.