| Taxi is an important part of the urban public transportation system,which accounts for a high proportion of urban traffic,but also has the characteristics of private transportation.By increasing the waiting time for taxis and the number of taxis to solve the problems of uneven distribution and high empty driving rate in the taxi industry,traffic congestion will be aggravated and a vicious circle will be created.However,the relationship between urban traffic development and urban spatial structure is inseparable.The two affect and permeate each other.Taxi trajectory data can reflect the temporal and spatial law of taxi passengers to a certain extent.It is of great significance to traffic demand management to clarify the relationship between taxi travel characteristics and urban spatial structure and to understand the inherent mechanism of the temporal and spatial heterogeneity of taxi passengers.In order to improve analytical precision,the taxi GPS data of Xi’an is needed to be pretreated.And then,from the time dimension,by statistical method,the frequency of taxi boarding and disembarking in each period of November 2019 is counted by week attribute,according to the results,redefine the working day and nonworking day.And randomly select 20 working days and non-working days to make statistics on the boarding and alighting frequency of taxis in each period of time.It is found that the boarding and alighting time of taxis in Xi ’an is nonstationary and goes through four ups and downs in one day.The peak time of each ups and downs is defined as the peak time of taxi boarding and alighting.Secondly,based on the nuclear density analysis method,a thermal diagram of the occurrence frequency of taxi passenger-carrying events in each peak period of working days and nonworking days is generated.And then taking the high-value density point as the passenger hot spot,defining the density value as the heat value,and classify the hot spots,and drawing the spatial distribution map of hot spots classification in each peak period of working days and nonworking days.The study found some of the first-class hot spots at each time period are distribution centers for urban residents.The calorific value of hot spots in these areas does not fluctuate greatly due to changes in working days,non-working days and time periods,while most of the hot spots for passengers have temporal and spatial non-stationarity.Finally,by introducing the urban POI data,the hot spots of taxi carrying passengers are matched to the nearest urban facilities,and the internal influence mechanism of the spatialtemporal differentiation characteristics of taxi carrying passengers is discussed by constructing regression model.In order to ensure the rationality of the model construction,the optional in-dependent variables included in the model are screened.Three variables related to official business are included in the screening results,which shows that the travel service provided by taxi is often induced by official business factors.The linear regression model and the geographical weighted regression model are constructed,and the results show that the geographical weighted regression model is better.From the perspective of time and space,it shows that there is a significant correlation between the characteristics of taxi carrying passengers and the distribution of urban spatial facilities,that is,from the perspective of urban society and geographical environment,to further understand the travel characteristics of urban taxi passengers,and to provide reference opinions for urban taxi dispatching,driver search and urban traffic infrastructure construction. |
PDF Full Text Request