Visual Analysis Of Route Selection Of Passenger Air-Rail Integration Transportation

With the development of railway transportation technology,the high-speed railway has been gradually favored by passengers due to its advantages of safety,comfort and quickness.However,the rapid development of high-speed railways has also caused strong impacts on the civil aviation market.In order to improve people’s travel efficiency and ensure the coordinated development of the railway and civil aviation industries,researchers have carried out studies on the air-rail integration mode.However,it still lacks the tools to analyze how to make personalized air-rail integration travel plans for passengers.Therefore,this thesis studies the visual analytics technology for the selection of air-rail integration routes from the perspective of passengers,which includes:1.The web crawling technology is studied.A method based on Python and combines urllib and Selenium,is used to crawl railway and airline-related data webpages.The multithreading module "threading" is used to improve the crawler efficiency.The crawled data is ultimately processed and stored in MySQL database.2.According to the characteristics of air-rail integration route data,a visualization method of bundling air-rail integration candidate routes is proposed.Its main idea is to bundle the routes whose transfer cities are close to each other:First,the transfer cities are clustered according to their locations.Then,the routes are segmented,and control points are set on each sub-route and the connection line between the departure city(or destination city)and the cluster center of transfer cities.Finally,a Bezier curve through the control points is drawn and the route diagram with edge-bundling effect is obtained.The experimental results show that this method can effectively reduce the visual clutter of air-rail candidate route diagram,and provide users with clear and intuitive route analysis.3.An edge-bundling method of Parallel Coordinates is proposed,which aims to reduce the visual clutter in the middle and both sides of curves between axes.The method first calculates the center position of data of different classes on each coordinate axis.It then calculates the offset distance of the data from the center of the class according to the offset strength set by the users.The next step is to set offset control points in the middle of the two adjacent axes,and on their sides.The final step is to draw a spline curve through the control points to obtain the Parallel Coordinates diagram with edge-bundling effect.The experimental results show that the method can reduce the visual clutter of Parallel Coordinates both in the middle of axes,and on their sides.Meanwhile,it can improve the resolution of the curves on the sides of the axis.4.For the high complexity of accurately calculating the number of line crossings of Parallel Coordinates,a method based on an approximate measure of the number of line crossings is proposed to fast rearrange the dimensions of Parallel Coordinates.In this method,the number of line crossings is firstly used as the visual clutter measure of the Parallel Coordinates.Then,the representative points in each dimension are selected,and used to form representative lines.Next,the number of line crossings of the original Parallel Coordinates is estimated based on the crossings of the representative lines.Finally,the dimensional arrangement is optimized so that the overall number of line crossings is minimized.The experimental results show that the method can effectively reduce the number of line crossings between different classes of lines,and the dimensions with similar distribution characteristics are arranged in the close position to make the relationship between data clearer.Besides,this method also has efficient time performance.5.A visualization analytics tool for the selection of air-rail integration routes is designed and implemented based on the crawled data.There are four modules in the tool:user input module,candidate route visualization module,timetable visualization module and travel data visualization module.Meanwhile,interactions are designed in and between each module to meet the users’ requirements of the analysis and selection of air-rail integration routes.