考虑客流引导和小群体行为的地铁车站疏散模型

地铁车站疏散过程客流引导及小群体行为均对人员疏散行为及疏散结果产生较大影响. 为更加有效地模拟真实疏散情况，研究了考虑客流引导和小群体行为的社会力模型. 该模型通过分析小群体运动特征及客流引导策略对行人期望速度的影响，对既有社会力模型进行修正；针对地铁车站站台人员疏散过程，设计更贴近实际情况的多智能体感知及决策流程，构建基于多智能体技术的疏散仿真模型. 以北京地铁西直门2号线站台为研究对象展开研究，结果表明客流引导及小群体效应对疏散时间、疏散效率、瓶颈区域及绕行距离均有显著影响:客流引导可提高疏散效率18%～45%，小群体行为则会增加绕行距离17%. 

Modelling Passenger Evacuation from Metro Platforms Considering Passenger Flow Guidance and Small Group Behaviour

Passenger flow guidance and small group behaviour during the actual evacuation process in metro stations greatly affect the behaviours of individual passengers and the evacuation result. To make a better simulation on the actual evacuation situation, the social force model was explored, which can take into account passenger flow guidance and small group behaviour. The model was modified by analysing the influence of passenger flow guidance and small group behaviour on the expected velocity of passengers. A multi-agent technique that involves realistic perception and decision processes was employed to develop a model for simulating passenger evacuation from a metro platform. Xizhimen metro station in Beijing was used as a case study object to verify the proposed model. Results indicate that both passenger flow guidance and small group behaviour have obvious effects on the evacuation time and efficiency, bottleneck areas, and detour distances. Passenger flow guidance increases the evacuation efficiency by 18%-45% and small group behaviour leads to an increase of 17% in the detour distance.