北京交通大学学报2025，Vol.49Issue(1)：180-190,11.DOI:10.11860/j.issn.1673-0291.20240051
基于减碳效益的共享单车骑行网络社区识别与影响因素分析
Identification and analysis of influencing factors in shared cycling network communities based on carbon reduction benefits
摘要
Abstract
To explore the relationship between community identification and the carbon reductionben-efits of shared bicycles,this study examines the identification of carbon reduction benefit communities and their influencing factors in Xi'an,China.First,based on the Hello Bike order data in Xi'an in 2020,the spatiotemporal distribution characteristics of shared bicycle trips are analyzed.Second,the carbon reduction benefits of shared cycling are quantified,and their temporal variation characteristics are analyzed.Then,the Louvain algorithm is employed to identify communities in the central urban area of Xi'an based on carbon reduction benefits,followed by classification using the K-means cluster-ing algorithm.Finally,the Gradient Boosting Decision Tree(GBDT)model is applied to explore the impact of the built environment on carbon reduction benefits.The results indicate that shared bicycle usage exhibits distinct morning and evening peak periods,with hotspots concentrated along subway lines and subway transfer stations.Shared bicycles significantly contribute to carbon reduction,with evident peak-hour effects.A total of 16 communities are identified based on carbon reduction benefits,with minimal overlap between these active communities and administrative divisions.The identified communities exhibit a"low-coupling,high-cohesion"structural pattern,where the city center con-tains a greater number of smaller communities,while peripheral areas have fewer but larger communi-ties.Central communities demonstrate more significant carbon reduction benefits.Based on the weighted average degree,graph density,and average clustering coefficient of the community,the 16 communities are categorized into three categories:low,medium,and high carbon reduction areas.All built environmental factors positively influence carbon reduction benefits,though to varying extents.The findings provide valuable insights for the management and policy formulation of shared bicycle car-bon reduction initiatives in Xi'an.关键词
城市交通/减碳效益/建成环境/社区识别/梯度提升决策树Key words
urban transportation/carbon reduction benefits/built environment/community identifica-tion/Gradient Boosting Decision Tree(GBDT)分类
交通工程引用本文复制引用
马书红,朱敏,杨磊,段超杰,董治宇..基于减碳效益的共享单车骑行网络社区识别与影响因素分析[J].北京交通大学学报,2025,49(1):180-190,11.基金项目
国家自然科学基金(52272316) （52272316）
教育部人文社科基金(23YJAZH097) （23YJAZH097）
陕西省社会科学基金(2023R030) National Natural Science Foundation of China(52272316) （2023R030）
Humanities and Social Research Foundation of Education of China(23YJAZH097) （23YJAZH097）
Social Science Foundation of Shaanxi Province(2023R030) （2023R030）
