Research On Commuting Trip By Rail Transit Based On Mobile Phone Signaling Data

At present,economic and social development put forward new requirements for transportation,intelligent transportation system is the inevitable choice for the development of transportation.As the backbone of the public passenger transport system,rail transit has gradually become one of the main modes of transportation used for commuting.Researching the characteristics of rail transit commuting is helpful to improve the efficiency of the rail transit commuting trip,enhance the attractiveness of rail transit commuting and improve the traffic situation and it is an important supplement to the intelligent transportation system.The analysis of the commuting trip characteristics of the rail transit relies on accurate and integral data.Traditional data acquisition and research scheme has some limitations.Mobile phone signaling data has the advantages of large sample range,long-term continuous monitoring and it has many application scenarios in the field of traffic.This thesis relies on the "Public Data Model Development Service based on Big Data of Chongqing Mobile Communication Company" project,using mobile phone signaling data to research the characteristics of rail transit commuting.First of all,this thesis analyses the volunteer signaling data provided by the operator,constructs a map grid system and on this basis chooses three characteristics of whether the service range of the base station where the signaling is located can cover the rail line,whether the trajectory of the travel signaling data is linearly related to the rail line and whether the travel speed and the orbital speed match to construct the identification model of rail transit trip based on the mobile phone signaling data.The accuracy of the model is verified by identifying the trip mode of the volunteer.The recall rate is 94.1% and the precision is 87.3%,and the results show that this model can effectively identify the passengers of the rail transit and has advantages compared with other algorithm models.Then based on the identification model of rail transit trip,this thesis uses the time characteristics to identify the Home-Work location of the rail transit passengers,constructs the commuting trip identification model of rail transit and identifies the passengers who use rail transit for commuting.On this basis,this thesis researches the characteristics of rail transit commuting: First,the thesis calculates commuting passenger flow of the target station during the peak hours of commuting and compares with the Automatic Fare Collection data.The result shows that the two errors are small and data are highly correlated after conversion,and further proves the correctness of the commuting trip identification model of rail transit.Then the thesis calculates the commuting passenger flow proportion in the overall passenger flow and analyses the distribution of source station of commuting passenger flow.Finally based on the commuting trip identification model of rail transit,the thesis researches the radiation distribution characteristics and the distance-time characteristics of the commuting connection trip,analyses the connection status of the target station.The results show that the target station’s connection is in good condition,but because the bus station distributes unevenly,there are some areas of low efficiency connection.This thesis researches the characteristics of rail transit commuting through mobile phone signaling data.It provides a new idea for modern traffic information data acquisition and application,and provides data support for rail transit planning,which has important significance and practical value.At present,the results of the thesis have been applied to a data service platform in an operator and have obtained good application effect.