Research On Distributed Calculation Of Spatio-temporal Characteristics Of Public Transport Big Data And Comprehensive Evaluation Of Lines

Since the 21st century,China’s economy has been developing continuously and rapidly,and the growth momentum of motor vehicle ownership has also remained strong for many years.According to the research of the National Bureau of Statistics,the number of motor vehicles in China has maintained an increasing trend year by year.By 2021,the number of motor vehicles in China has reached 302 million,a year-on-year increase of 7.5%.Behind the continuous growth of motor,vehicle ownership is the increasing travel demand of residents.Compared with other capital cities,the private cars travel rate in Beijing is the highest,but the density of the road network is the lowest.The contradiction between the increasing travel demand of this kind of residents and the inadequate public transport supply is the main reason for the formation of traffic congestion.Therefore,it is particularly important to have a clearer and intuitive understanding of the Spatio-temporal travel characteristics and Spatio-temporal demand distribution of passengers and to establish a public transport evaluation system that meets the needs of China’s urban public transport development and provides guiding suggestions for public transport development.Based on the public transport IC card data,combined with GPS data,public transport station data and station space vector data,this paper uses Hadoop distributed computing to clean and process the public transport IC card data,and uses overlay analysis,information visualization,spatial correlation analysis and other research methods from the whole day of weekdays and holidays,morning and evening peak Under the different time and space constraints of the city-Administrative Region-line-bus stop,the temporal and spatial characteristics,distribution law and temporal and spatial demand distribution of bus passenger flow in Beijing are excavated.The areas with typical commuting characteristics and stations with poor performance are summarized,which lays a research foundation for the selection,evaluation and calculation of bus line research objects.According to the analysis results,this paper establishes the evaluation index system and evaluation model for bus lines and puts forward a comprehensive evaluation model based on improved super-efficiency DEA from three aspects: basic layout,operation and service level of bus lines.The model restricts the established bus line evaluation system from three aspects:passenger flow time characteristics,spatial characteristics and passenger flow intensity,and solves the efficiency,it avoids the problem that the evaluation results are too subjective due to the simple use of the AHP model,and meets the needs of dynamic changes of the urban public transport system.Taking Beijing public transport as an example,this paper uses the comprehensive evaluation method close to the ideal value based on AHM attribute hierarchy model,ordinary DEA model and improved super-efficiency DEA comprehensive evaluation model to test the constructed evaluation system and evaluation model respectively.Some lines in Beijing are selected as the research object for verification,and the bus lines with poor performance are screened.Combined with the analysis results of passenger flow temporal and spatial characteristics,the key factors affecting the performance of bus lines are found,and targeted optimisation strategies of lines and stations are put forward,to provide a scientific basis for the improvement and optimisation of bus lines.Based on the Spatio-temporal analysis of bus passenger flow and comprehensive evaluation of bus lines,this paper develops and constructs a visualization platform for Spatiotemporal characteristics of bus big data and comprehensive evaluation of bus lines based on spatial information visualization technology.With a more intuitive and visual way and form,this paper transforms a large amount of bus IC card data into the operation status and operation efficiency of buses and bus lines and carries out real-time visual rendering and analysis at the fore-end;it provides a visual tool platform for bus route optimization.