Research On Urban Public Transport Visual Analysis Of Spatio-Temporal Trajectory Data

Public urban transport network has the characteristic of wide coverage,large passenger capacity,high efficiency,low transportation cost and low energy consumption.Therefore,the study of urban bus has been widely concerned.Buses have the features of stable running routes and static parking places,which is helpful to study road congestion patterns,the congestion patterns and importance of bus station.As a complex system,the punctuality rate of public transport system is affected by traffic jam,varying depature interval,etc.How to describe such uncertaionty is an important research subject.The sparse trajectory data is adopted in this paper,namely the pertinent records of the buses arrival and departure from the relevant bus stations.On the basis of bus data preprocessing,an enhanced node importance evaluation algorithm is presented,namely the dynamic DN evaluation algorithm.A visualization model of the knotted-line is designed for uncertainty visualization.The main work of this paper is summarized in the following three aspects:First,the visual analysis of urban traffic situation includes road congestion pattern analysis,bus station congestion pattern analysis and bus station importance analysis.In order to quantify the importance of bus station in different time periods,the dynamic DN evaluation algorithm is proposed,which combines the dynamic properties of bus station,including parking time and traffic volume,to evaluate the importance of bus station dynamically.Compared with the traditional node importance evaluation algorithm base on network topology,the experiments show that the dynamic DN evaluation algorithm is more real-time and effective.Second,in the visual analysis of bus uncertainty,the knotted-line model is designed to visualize bus uncertainty by modeling passenger waiting time and bus arrival time.It is composed of three parts,violin plot,multi-later ring and risk indicator.The violin plot shows the uncertainty of passenger waiting time.The multi-layer ring describes the uncertainty of bus arrival time.The risk indicator expresses the possibility of passenger spending time on the trip.The user evaluation shows that the knotted-line model can effectively help passagers understand the uncertainty of bus system and make decisions.Third,based on the large-scale buses trajectory dataset,the dynamic DN evaluation algorithm and the knotted-line visualization model,the city bus visual analysis system is designed and implemented,namely BVis.Many spatio-temporal trajectory data visualization technologies and uncertainty visualization technologies are integrated in this system to analyze urban traffic situation and bus system uncertainty.Finally,the practicability and effectiveness of BVis are demonstrated by three case studies and a user study.