基于时空莫兰指数的轨道交通多类型异常客流特征分析

Analysis of multi-type abnormal passenger flow in urban rail transit based on spatiotemporal Moran's I

摘要

异常客流是城市轨道交通运营安全面临的重要挑战，深刻把握其时空演化规律，是实现精准预警与高效响应的基础。然而，现有研究仍缺乏针对不同类型事件下异常客流特征的系统性对比分析。为此，本文以沈阳轨道交通网络为研究对象，构建基于客流波动的时空莫兰指数模型，对大型活动、节假日与恶劣天气三类事件引发的异常客流时空特征进行刻画分析。研究发现，三类异常客流均表现出时空正相关性，但演化模式具有显著差异：大型活动客流呈现“局部聚集、阶段突出”的非扩散性特征，时空冲击强、影响范围集中；节假日客流则具有“范围广、历时长”的多中心扩散特征，其演化主要源于站点功能替代，而非单纯需求增长；恶劣天气通过强化刚性通勤、抑制弹性出行，进一步推动客流向通勤骨干站点集聚；在复合事件情景下，客流结构呈现主导事件与背景事件相互叠加、协同作用的复合格局。本研究结论可为不同类型异常客流下的差异化运营管控与动态运力调度提供决策支持，并在特征构建与数据增强层面，为破解异常客流预测中“样本稀缺、难以学习”的瓶颈提供理论依据。

Abstract

Abnormal passenger flow poses a significant challenge to the operational safety of urban rail transit. Accurate early warning and effective response depend on a deep understanding of the spatiotemporal patterns of such flows. However, existing studies still lack a systematic comparative analysis of the characteristics of abnormal passenger flow under different types of events. To address this gap, this study took the Shenyang rail transit network as the research object and constructed a spatiotemporal Moran’s I model based on passenger flow fluctuations to compare and analyze the spatiotemporal evolution characteristics of abnormal passenger flows under three types of events: large-scale activities, holidays, and severe weather. The findings reveal that all three types of abnormal passenger flows exhibit significant spatiotemporal positive correlations, yet their evolution patterns differ markedly: large-scale activity passenger flows show a non-diffusive characteristic of “localized aggregation with strong periodicity”, featuring high spatiotemporal impact intensity but limited spatial scope; holiday passenger flows demonstrate a multi-center diffusion pattern with “wide influence range and long duration”, driven primarily by the functional substitution of stations rather than mere demand increase; severe weather intensifies rigid commuting demand while suppressing flexible travel, leading to further concentration of passenger flow at key commuting stations. Under composite event scenarios, passenger flow structures present a composite pattern resulting from the synergistic interaction between dominant and background events. The findings of this study can provide a decision-making basis for differentiated operational management and scheduling strategies under different types of abnormal passenger flow events, while also offering theoretical support at the feature construction and data augmentation levels to address the bottleneck of “scarce samples and difficult learning” in abnormal passenger flow prediction.

关键词

轨道交通 / 异常客流 / 时空特征 / 莫兰指数

Key words

rail transit / abnormal passenger flow / spatiotemporal characteristics / Moran's I

参考文献

[1] wang Haiyang, Li Longyuan, Pan Pingjun, et al. Online detection of abnormal passenger out-flow in urban metro system[J]. Neurocomputing, 2019, 359: 327-340.

[2] 李兆丰, 倪少权, 孙克洋, 等. 基于多特征融合的城市轨道交通短时客流预测[J]. 交通运输工程与信息学报,2020,18(4):93-102.

[3] Li Zhaofeng, Ni Shaoquan, Sun Keyang, et al. Urban rail transit short-term passenger flow prediction based on multi-feature fusion[J]. Journal of Transportation Engineering and Information, 2020, 18(4): 93-102. (in Chinese)

[4] Jin Guangyin, Cui Yan, Zeng Liang, et al. Urban ride-hailing demand prediction with multiple spatio-temporal information fusion network[J]. Transportation Research Part C: Emerging Technologies, 2020, 117: 102665.

[5] Gan Zouxian, Feng Tao, wu Yunteng, et al. Station-based average travel distance and its relationship with urban form and land use: an analysis of smart card data in Nanjing city, China[J]. Transport Policy, 2019, 79: 137-154.

[6] Li Yuan, Yang Linchuan, Shen Han, et al. Modeling intra-destination travel behavior of tourists through spatiotemporal analysis[J]. Journal of Destination Marketing & Management, 2019, 11: 260-269.

[7] 游世凯, 张聪, 邹歆. 基于多源数据的无锡轨道交通客流特征分析[J]. 都市快轨交通,2020,33(5):94-100.

[8] You Shikai, Zhang Cong, Zou Xin. Passenger flow characteristics of wuxi urban railway based on multi-source data[J]. Urban Rapid Rail Transit, 2020, 33(5): 94-100. (in Chinese)

[9] 吴祥国, 翟长旭, 张建嵩, 等. 重庆市郊铁路江跳线客流时空特征分析[J]. 铁道运输与经济,2024,46(5):192-200.

[10] wu Xiangguo, Zhai Changxun, Zhang Jiansong, et al. Analysis of spatial-temporal characteristics of passenger flow in Chongqing suburban railway Jiangtiao line[J]. Railway Transport and Economy, 2024, 46(5): 192-200. (in Chinese)

[11] 叶倪碹, 杨超, 闫茜. 深圳市轨道交通周五大客流特征及成因分析[J]. 城市轨道交通研究,2021(5):24.

[12] Ye Nixuan, Yang Chao, Yan Qian. Characteristics and cause analysis of Shenzhen rail transit large passenger flow on Friday[J]. Urban Mass Transit, 2021(5): 24. (in Chinese)

[13] 郑俊锋, 方旭峰, 王晨阳. 基于时间序列分解方法的城市轨道交通大客流预测方法[J]. 城市轨道交通研究,2023,26(8):163-170.

[14] Zheng Junfeng, Fang Xufeng, wang Chenyang. Prediction method for urban rail transit large passenger flow based on time series decomposition[J]. Urban Mass Transit, 2023, 26(8): 163-170. (in Chinese)

[15] Jiang Shixiong, Lin Yuchen. Ridership and human mobility of metro system under the typhoon weather event: a case study in Fuzhou, China[J]. Urban Rail Transit, 2022, 8(1): 32-44.

[16] Guo Yongqing, wang Xiaoyuan, Xu Qing, et al. weather impact on passenger flow of rail transit lines[J]. Civil Engineering Journal, 2020, 6(2): 276-284.

[17] 禹丹丹, 陈文, 韩庆龙, 等. 北京地铁节假日客流特点和运营组织方法分析[J]. 现代城市轨道交通,2019(4):58-66.

[18] Yu Dandan, Chen wen, Han Qinglong, et al. Analysis of Beijing subway passenger flow characteristics and operation organization method during holidays[J]. Modern Urban Transit, 2019(4): 58-66. (in Chinese)

[19] 马飞, 蒋金凤, 敖誉芸, 等. 非均衡大客流冲击下城市轨道交通网络抗毁性建模及演化特征[J]. 清华大学学报(自然科学版),2024,64(10):1717-1733.

[20] Ma Fei, Jiang Jinfeng, Ao Yuyun, et al. Modeling and evolution characteristics of urban rail transit network resistance under large passenger flows[J]. Journal of Tsinghua University (Science and Technology), 2024, 64(10): 1717-1733. (in Chinese)

[21] 钱慧敏, 李静, 张琳, 等. 持续型大型活动影响因素及客流分布特征分析[J]. 交通工程,2020,20(1):60-65.

[22] Qian Huimin, Li Jing, Zhang Lin, et al. Analysis of influence factors and passenger flow distribution characteristics about lasting large-scale events[J]. Traffic Engineering, 2020, 20(1): 60-65. (in Chinese)

[23] Jiang Shixiong, Cai Canhuang. Unraveling the dynamic impacts of COVID-19 on metro ridership: an empirical analysis of Beijing and Shanghai, China[J]. Transport Policy, 2022, 127: 158-170.

[24] Fernandes P, Santos L, Bandeira JM, et al. Multiple linear regression of metro ridership in the context of football events: a case study of Metro de Lisboa[J]. Transportation Research Procedia, 2025, 86: 151-158.

[25] Yao Enjian, Hong Junyi, Pan Long, et al. Forecasting passenger flow distribution on holidays for urban rail transit based on destination choice behavior analysis[J]. Journal of Advanced Transportation, 2021(1): 9922660.

[26] 肖逸影, 王东, 吴宁宁, 等. 冻雨极端天气下武汉市轨道交通客流特征分析[J]. 现代城市轨道交通,2025(7):107-113.

[27] Xiao Yiying, wang Dong, wu Ningning, et al. Analysis of passenger flow characteristics of wuhan municipal rail transit under extreme weather condition of freezing rain[J]. Modern Urban Transit, 2025(7): 107-113. (in Chinese)

[28] Alghushairy O, Alsini R, Soule T, et al. A review of local outlier factor algorithms for outlier detection in big data streams[J]. Big Data and Cognitive Computing, 2021, 5(1): 1.

[29] 杨静, 代盛旭, 张红亮, 等. 大型活动散场期间地铁车站短时进站客流预测[J]. 科学技术与工程,2021,21(5):2042-2048.

[30] Yang Jing, Dai Shengxu, Zhang Hongliang, et al. Prediction of short-term passenger flow of metro station in the period of planned special events[J]. Science Technology and Engineering, 2021, 21(5): 2042-2048. (in Chinese)

[31] Liu Jianxiao, Shi wenzhong, Chen Pengfei. Exploring travel patterns during the holiday season-a case study of Shenzhen metro system during the Chinese spring festival[J]. ISPRS International Journal of Geo-Information, 2020, 9(11): 651.

[32] Kalkstein AJ, Kuby M, Gerrity D, et al. An analysis of air mass effects on rail ridership in three US cities[J]. Journal of Transport Geography, 2009, 17(3): 198-207.

[33] 陈绍宽, 韦伟, 毛保华, 等. 基于改进时空 Moran’sI指数的道路交通状态特征分析[J]. 物理学报,2013,62(14):527-533.

[34] Chen Shaokuan, wei wei , Mao Baohua, et al. Characterization of road traffic state based on improved spatiotemporal Moran’s I index[J]. Acta Physica Sinica, 2013, 62(14): 527-533. (in Chinese)

基金资助

国家自然科学基金资助项目(52302403)

辽宁省社会科学规划基金项目(L24BGL005)

中央高校基本科研业务费专项资金资助项目(3132026183)