In recent years,the market for bike sharing in China has gradually tend to be stable after fierce competition.Its rapid development has brought a great impact on other transportation modes in the city,especially on buses and subways which have large volume and service intensity.Existing studies show that there is a compound relationship between bike sharing with public transportation modes,which is both competition and cooperation.In order to promote the sound and healthy development of bike sharing in the transportation system,this study subdivides and quantifies the composite relationship from the perspective of public travel,explores the interaction mechanism between bike sharing and public transportation,and confirms the development strategy and direction of bike sharing in the future.Therefore,based on the riding data of Mobike in Chengdu,bus and subway stations,road network,points of interest and demographic and economic data,this paper divides the research area into grids as statistical analysis units to carry out the related work of relationship research.First of all,the relevant studies on the mechanism and influencing factors of the interaction between bike sharing and public transport in the existing literature are summarized,and the relationship studied in this paper is subdivided into three types: competition,connection and complement.Then,after processing the abnormal and missing data in the original data,according to the location of public transport stations,the research area is divided into three space types: within 50 m of the stations,from 50 m to 400 M of the stations,and beyond the400 m of the stations.The relationship between each bike riding and public transportation can be identified by judging the type of public transportation demarcated range of the starting and ending locations of bike sharing riding.Next,the usage characteristics,such as age and gender,as well as the distribution feature in time and space of each relationship type of bike riding is analyzed.Finally,in order to analyze the influence mechanism of different factors on the three relationship types of riding,four types of explanatory variables,including user attributes,cycling characteristics,socio-economic factors and built environment factors,are constructed,and the spatial autoregressive statistical model and random forest regression machine learning model were used to explore.Riding characteristics analysis results show that the competitive relations of the bike riding accounted for more than half of all riding,up to 57% on weekdays.The connection relation bike riding ratio is about 42%.There are also significant differences among three different bike riding relationship types during weekday and weekend,users age,riding duration and distance.The calculation results of Moran’s I showed that that the three relationship types of riding are spatially correlated,among which,the aggregation degree of complementary relationship was the highest,and the competition and the connection relationship decrease successively.The results of spatial autoregressive model show that the effect of user age,gender,riding duration,population density,housing price and other factors on different relationship bike riding are different.For example,the average riding duration is positively correlated with the competition and supplement relation riding,but negatively correlated with the connection relation,The degree of influence of various factors on the three relation is also different.It can be seen from the results of random forest regression model that the distance from the starting position to the bus station has the greatest influence on the three relation ridings,and the top five variables of importance are: average distance to bus station,distance to nearest subway entrance,male proportion,riding distance and riding duration,and the importance of each variable to the different relation ridings are different.Based on the above research results,the corresponding bike sharing development strategy is proposed,so that it can coordinate and sustainable development with public transportation. |