| With the proposed ”carbon peaking” and ”carbon neutral” goals in China,the transportation sector,as the second largest consumer of oil and a major producer of greenhouse gases,is a critical area for energy efficiency and emission reduction actions.Traffic congestion is the main reason for the dramatic increase in carbon emission from the transportation sector and the main challenge of daily travel for residents.Reducing the impact of traffic congestion on residents’ commuting while reducing carbon emissions from personal travel is of great practical importance to the sustainable development of cities and enhancing residents’ travel experience.Current research in this area focuses on two main points: traffic congestion prediction and mixed traffic travel.Traffic congestion prediction can reduce the impact of congestion on users by notifying them in advance to bypass the predicted congestion area,which is very practical for occasional congestion during non-peak travel hours,but very limited for users who travel during fixed hours of the commuting peak.The improvement of urban public transportation makes it possible to the research on mixed traffic travel.By selectively using public transportation instead of small cars during the user’s travel,mixed traffic travel can reduce the impact of traffic congestion on the user’s trip without altering the user’s original travel path,making it ideal for solving commuter peak travel challenges and reducing personal travel carbon emissions.However,existing studies focus on route recommendation for mixed traffic travel,lacking the exploration and analysis of travel context.Therefore,this paper proposes a multi-layer complex traffic network model containing both taxi and bike-sharing travel information,as well as a mixed traffic route recommendation method based on this model to take into account urban traffic pattern analysis and mixed traffic route recommendation.The specific tasks are as follows.(1)Multi-layer complex traffic network model.To address the problems that the traditional road network model cannot display urban traffic flow information and the singlelayer traffic network cannot describe the urban mixed traffic travel structure,a traffic feature network based on the extraction of multiple travel mode trajectory data is proposed and fused into an urban mixed traffic feature network to realize the extraction and visualization of urban mixed traffic flow information.The traffic feature network is clustered based on a data field clustering algorithm to generate a traffic community network that provides an overview of urban traffic conditions.Finally,several networks are combined into a multi-layer complex traffic network model in specific steps to achieve a progressive exploration and analysis of urban traffic patterns.(2)Low-carbon mixed traffic route recommendation.A low-carbon mixed route recommendation method based on a multi-layer complex traffic network model is proposed to address the problem that traditional single-mode route recommendation and congestion prediction cannot solve the travel problems during the peak commuting hours and reduce the carbon emissions of individual travel.For the challenges that traditional route planning algorithms are hardly adapted to route recommendation on traffic feature networks and the inefficiency of algorithm operation under large-scale networks,an A* algorithm using genetic algorithm improvement is proposed to complete real-time mixed route calculation on urban mixed traffic feature networks.And a trip chain-level carbon emission assessment method is proposed to realize the calculation of carbon emission distribution in urban streets and the assessment of carbon emission of mixed traffic routes.(3)Visual recommendation system for urban mixed traffic based on multilayer complex networks.Based on the above proposed method,we designed and completed the mixed traffic visual recommendation system.Based on the concept of "whole to part" visualization and combined with a variety of interactive technologies,it enables progressive exploration and analysis of urban hotspots,street carbon emission distribution,traffic patterns and mixed traffic route recommendations.Finally,the usability and effectiveness of this paper are verified through several case studies of the system with different analysis purposes. |
