交通运输系统工程与信息 ›› 2025, Vol. 25 ›› Issue (6): 197-208.DOI: 10.16097/j.cnki.1009-6744.2025.06.018

• 系统工程理论与方法 • 上一篇    下一篇

地铁突发事件中行李遗弃与结伴行为对疏散效能的耦合影响

张立业1 ,马月熙1 ,马志成2 ,赵丰磊1 ,焦春硕*1   

  1. 1. 山东科技大学,交通学院,山东青岛266590;2.中国石油大学(华东),机电工程学院,山东青岛266590
  • 收稿日期:2025-08-06 修回日期:2025-09-22 接受日期:2025-09-28 出版日期:2025-12-25 发布日期:2025-12-24
  • 作者简介:张立业(1981—),男,山东临沂人,教授,博士。
  • 基金资助:
    泰山学者项目基金(taqn202103077)。

Coupled Effects of Luggage Abandonment and Companion Behavior on Evacuation Efficiency in Subway Emergencies

ZHANG Liye1, MA Yuexi1, MA Zhicheng2, ZHAO Fenglei1, JIAO Chunshuo*1   

  1. 1. College of Transportation, Shandong University of Science and Technology, Qingdao 266590, Shandong, China; 2. College of Mechanical and Electronic Engineering, China University of Petroleum, Qingdao 266590, Shandong, China
  • Received:2025-08-06 Revised:2025-09-22 Accepted:2025-09-28 Online:2025-12-25 Published:2025-12-24
  • Supported by:
    Shandong Taishan Scholars Program(taqn202103077)。

摘要: 地铁大规模疏散效能优化是轨道交通安全的关键,但现有模型因均质化假设和单一人群类型划分,缺乏复杂人群研究。为此,本文提出改进的社会力模型,构建结伴行为-行李携带-遗弃行为三维耦合动态仿真框架。基于青岛站1081份问卷数据校准参数,通过AnyLogic对廊道转角与车厢出入口关键瓶颈区域进行仿真。首先,依据实证数据设定行李丢弃率和结伴行人占比,建立结伴群体、携带行李箱行人、普通行人及遗弃行李箱4类异质主体交互规则,改进单因素模型对复杂人群行为的简化处理;然后,基于地铁转角几何特征与列车尺寸构建模拟疏散场景;最终,通过仿真对比验证表明,在转角与车厢区域耦合模型较单因素模型疏散时间差10s以上,在多耦合疏散模拟中,结伴行人疏散受影响最大,延误时间达携带行李者的1.5倍(转角区)和1.66倍(车厢区)。

关键词: 城市交通, 社会力模型改进, 仿真模拟, 地铁行人疏散

Abstract: The optimization of large-scale evacuation efficiency in subways is crucial to rail transit safety. However, existing models lack research on complex crowds due to their homogenization assumptions and single-type classification of crowds. To address this, this paper proposes an improved social force model and constructs a three-dimensional coupled dynamic simulation framework involving companion behavior, luggage carrying, and luggage abandonment behavior. Parameters are calibrated based on 1081 questionnaire data from Qingdao Railway Station, and simulations are conducted on key bottleneck areas such as corridor corners and carriage entrances/exits using AnyLogic. The luggage abandonment rates and the proportion of pedestrians traveling in groups are defined based on the empirical data. The interaction rules for four types of heterogeneous subjects, namely companion groups, pedestrians carrying suitcases, ordinary pedestrians, and abandoned suitcases, are established to improve the simplified handling of complex crowd behaviors in single-factor models. A simulated evacuation scenario is established based on the geometric characteristics of subway corners and train sizes. The simulation comparison and verification show that in corner and carriage areas, the coupled model has an evacuation time difference of more than 10 seconds compared with the single-factor model. In multi-coupling evacuation simulations, the evacuation of pedestrians traveling in groups is the most affected, with the delay time reaching 1.5 times (in corner areas) and 1.66 times (in carriage areas) of those carrying luggage.

Key words: urban traffic, social forces model improvements, simulation and emulation, subway pedestrian evacuation

中图分类号: