| With the development of economy and diversification of travel demand,online car-hailing plays an increasingly important role in people's daily travel activities.A large amount of positioning data have been generated during the travel process.These positioning data reflect a wide range of residents' movement in space and are recorded for a long time,which can provide a new basis for the study of people's movement in urban areas and its changes in space and time.On the basis of summarizing the relevant researches of domestic and foreign scholars on the temporal and spatial features of residents,this paper combines the order data generated by the online car-hailing platform and related geospatial data to carry out the following research:(1)Analyze the changes in the residents' online car-hailing travel activities over time,and the difference in the regularity of the travels between workdays and weekends.According to the characteristics reflected by the amount of travels in the unit time,a day is divided into four time periods,and the travel time,travel distance,and average speed of the residents' online car-hailing travel are compared.(2)The density-based OPTICS algorithm was used to cluster the pick-up and drop-off locations to identify the hot areas of the online car-hailing travel activities.This algorithm can reduce the sensitivity of the clustering process to the input parameters,making the clustering results more objective and accurate.Based on the identification of hot areas,the spatial distribution of such areas and their changes over time were studied.The travel hot areas are generally distributed in the vicinity of the city center,and the density of online car-hailing travel around the large traffic station is always maintained at a high level.There is a certain imbalance in the travel hot areas in the space,and it's not necessary that there are more drop-off activities in the hot areas of the pick-up activities.(3)Study the core area of the online car-hailing travel activities and its directionality in the urban space through the standard deviation ellipse of the travel activity spatial distribution.The results show that the online car-hailing travel are balanced in all directions,and do not produce obvious bias in a certain direction,which also reflects the relatively reasonable spatial layout of the city.Create the Thiessen Polygon to divide the research area and determine the service area of a single subway station,calculate the concentration of online car-hailing travel activities in each service area,visualize and analyze the concentration of travel activities in each area,and study the distribution of areas with different concentration in urban space.Extending from the center of the city to the periphery,the degree of concentration of travel activities around the subway station is generally weakening,but the degree of spatial concentration of travel activities near large traffic stations is significantly higher than that of the adjacent areas.(4)The Ordered Logistic model was used to study the influence of the type and quantity of points of interest(POI)on the intensity of travel activity.The study found that different types of POI have different effects on travel activities,and even the impact of same POI on pick-up and drop-off activities may vary.The traffic facilities have the most significant impact on the intensity of the online car-hailing travel activities.The impact of dinning and medical facilities on the travel activities is more obvious.The living service facilities and sports facilities have no significant impact on the drop-off activities,but have a significant impact on pick-up activities.The study of the temporal and spatial features of residents' travel by online car-hailing helps to deepen our understanding of residents' travel behaviors and patterns,explains the travel needs of potential passengers in different time periods and different regions from the perspective of traffic management,facilitates the dispatching management of online cars,and also serves as a reference for order prediction and vehicle allocation on online car-hailing platform,so as to realize the effective allocation of limited resources.The research results in this paper can also provide help for location-based services,such as consumer recommendation,store location selection,etc,and have important reference value for urban planning. |
PDF Full Text Request