Analysis Of Subway Travel Characteristics And Its Influencing Factors Based On Spatio-temporal Data

With the rapid development of social economy,the degree of urbanization and motorization is further deepened,more and more urban residents,residents car ownership rate is also getting higher and higher,followed by serious environmental pollution,traffic congestion problem is becoming more and more prominent.As a kind of public transportation,urban rail transit can bear a great amount of passenger traffic,effectively improve the efficiency of residents’ travel,objectively understand the characteristics of urban rail transit passenger flow,and analyze the mechanism of influencing factors of passenger flow and travel time.It is of positive significance to solve the problem of urban traffic congestion and rationally use resources to plan the layout of urban rail transit.It can also further improve the travel efficiency of urban rail transit,guide the rational adjustment of urban spatial layout,and promote the harmonious development of the city.Taking Nanchang Metro Line 1 and Line 2 as an example,based on multi-source big data such as passenger traffic data,subway card data,subway station spatial location data,POI data,built environment data and economic attribute data,Arc GIS10.8,Python3.9,Tableau and other software were used for data processing.The spatial and temporal characteristics of Nanchang subway travel were analyzed from two dimensions of time and space,and the influencing factors of station passenger flow and travel time were analyzed.The limit gradient boosting model(XGBoost)was further used.The influence of 19 factors on the number of station facilities,built environment and economic attributes on the passenger flow and travel time of the station is analyzed.The results of the research and analysis are as follows:(1)The spatial pattern of the average travel time of the passenger flow at the Nanchang subway station is mainly characterized by a U-shaped distribution with low middle and high sides.The probability density distribution of subway travel time and passenger flow conforms to the characteristics of the Gamma curve,and the average travel time of each station is24 minutes.(2)The passenger flow distribution of the station is mainly concentrated in the central urban area of Honggutan,the old central urban area and the east area of the city,which is generally characterized by high in the middle and low in the surrounding areas.The average daily passenger flow of each station is 11424.(3)The number of facilities has the highest degree of influence on passenger flow.Multiple factors such as commercial housing and transportation facilities have a multi-stage increasing nonlinear positive correlation with passenger flow.Shopping consumption and distance from CBD are negatively correlated with passenger flow.House prices show convex nonlinear characteristics.(4)The number of facilities and the built environment have a high degree of influence on travel time.Shopping consumption and distance from CBD have a non-linear positive correlation with travel time.Traffic facilities,road network density,housing prices and average post wages have a non-linear negative correlation with travel time.Based on the above research conclusions,considering the current situation of Nanchang Metro,relevant optimization suggestions are put forward from multiple perspectives.