| Public transportation plays a supporting and leading role in the sustainable development of cities.Promoting the priority development of public transportation is an effective way to mitigate environmental pollution and traffic congestion,as well as improve the efficiency of urban transportation operation.However,with the development of society and changes in travel rules,urban bus routes cannot effectively meet the daily travel demands of passengers,and their efficiency still needs to be improved.Bus route planning needs to consider the common influence of multiple factors from a global perspective to seek the convergence of urban development and actual demands,so as to balance the costs of travelers and the benefits of operators.However,it faces huge challenges to accurately perceive passenger travel demand,formulate bus route planning schemes that meet multiple goals,and quantitatively analyze the influencing factors and their weights.On the one hand,since the public transport demands shows complex spatial and temporal heterogeneity,traditional statistical analysis methods cannot adapt to the frequently changing road structure and traffic demands,and the method of perceiving passenger travel demand and laying out bus stations based on trajectory data lacks consideration of the overall distribution of urban bus stations,the accuracy of mining candidate bus stops still needs to be improved.On the other hand,the possible bus route planning schemes constitute a huge solution space,and it is extremely difficult to improve the efficiency and accuracy of finding the optimal route set.In addition,the quantitative analysis of the influencing factors of the route planning and the resulting weights are extremely important to compare the differences between the candidate routes and formulate a reasonable route planning scheme.In order to solve the above problems,this study proposes a visual analytics framework for bus route planning that can perceive passengers’ travel demands,organically integrates intelligent processing,visual representation and interactive analysis of urban traffic data,identifies hotspots for passenger travel,and improves the accuracy of bus station mining and route design.Quantitatively analyzing the influence of attribute weights on the formulation of route planning schemes can provide scientific decision-making basis and practical means for transportation departments to carry out bus route planning,thus promoting the development of smart cities.The main research contents of this paper include:1.Propose a method for mining candidate bus stops based on GPS data perceiving passenger travel demands.It comprehensively considers the connectivity of passenger travel hotspots and the service area of bus stations,which can improve the rationality and accuracy of mining candidate bus stops,so as to better realize the overall deployment of bus stops in urban areas.2.Propose a method for finding the optimal bus route set based on the multi-objective optimization and an interactive method of quantifying route attribute weights.A multi-objective model of bus route planning is constructed and solved by the NSGA-II multi-objective optimization method,which can improve the accuracy of the generated routes and the efficiency of finding the optimal set.On this basis,a method of quantifying route attribute weights is proposed,and the factors affecting the bus route planning schemes are quantitatively analyzed from multiple perspectives.Moreover,an interactive feedback mechanism is designed to deeply integrate the experience of planners,aiming to improve the accuracy and credibility of the optimal route recommendation.3.Design and develop a visual analytics system for bus route planning that can perceive passengers’ travel needs.This system supports the planner’s visual analytics process from the global overview of the route schemes,comparative analysis of routes,and to the detailed exploration.Through multiple collaborative views and interactive designs,the multi-level comparison of bus routes is realized,and the optimal routes can be explored in depth from the perspective of multiple attributes.This paper conducts experiments based on real urban transportation datasets,and verifies the effectiveness of the proposed method and system for urban bus route planning through multiple case studies and the user evaluation,which can provide important support for the formulation of urban bus route planning schemes. |
