首页|基于随机森林的人群疏散预测及影响因素分析

基于随机森林的人群疏散预测及影响因素分析

扫码查看
为对人群疏散预期目标快速预测,分析预期目标与多重影响因素之间的关系,提出基于疏散仿真模拟和随机森林相结合的预测模型.首先,构建人群疏散三维模型;其次,分析影响疏散的关键因素;最后,基于随机森林构建预测模型,开展预测变量的重要性分析.研究结果表明:该方法能对预期目标进行快速预测,预测准确度达到 94%;基于变量重要性分析发现,人员数量对疏散的影响显著,且小群体的存在对疏散也有重要影响.研究结果可为人群安全疏散评估提供一定参考.
Crowd evacuation prediction and influencing factors analysis based on random forest
In order to quickly predict the expected goals of crowd evacuation and analyze the relationship between the expec-ted goals and various influencing factors,a prediction model based on the combination of evacuation simulation and random forest was proposed.Firstly,a three-dimensional model of crowd evacuation was constructed.Secondly,the key factors affecting evacuation were analyzed.Finally,a prediction model based on random forest was constructed,and the importance analysis of the prediction variables was carried out.The results show that this method can quickly predict the expected goals with a pre-diction accuracy of 94%.Based on the importance analysis of variables,it is found that the number of personnel has a signifi-cant impact on evacuation,and the presence of small groups also has significant influence on evacuation.The research results can provide some reference for the safety evacuation assessment of crowd.

crowd evacuationrandom forestinfluencing factorimportance analysis

张贺丹、丁元春、刘祖文

展开 >

江西理工大学 赣州市工业安全与应急技术重点实验室,江西 赣州 341000

江西理工大学 应急管理与安全工程学院,江西 赣州 341000

南昌工程学院 鄱阳湖流域水工程安全与资源高效利用国家地方联合工程实验室,江西 南昌 330099

人群疏散 随机森林 影响因素 重要性分析

国家自然科学基金

72164016

2024

中国安全生产科学技术
中国安全生产科学研究院

中国安全生产科学技术

CSTPCD北大核心
影响因子:1.119
ISSN:1673-193X
年,卷(期):2024.20(3)
  • 15