人类行为时空分布特征及恐慌人群行为中的统计物理问题研究

The second ten years of this century is the great era that the complexity science goes deep into each domain, has the new breakthrough unceasingly, and faces a bigger challenge. .Many important complex systems (life, economy and social systems; Internet, human behaviors, traffic flow in the cities, social relation network, etc. ) are waiting for us urgently to carry on complete new understanding and exploration using the statistical physics research technique, unifying both of the real diagnosis statistics and theoretical model analysis. This project will unearth and understand deeply the basic physical mechanism and the statistical physics rules determinning the basic evolutionary process of the complex systems for human community groups, revolving round the statistical physics questions for the spacial and temporal distribution characteristics of the humanity repeating behaviors and the panic crowd behaviors in emergency, focusing on the complexity phenomenon for the characteristics of the human own behaviors and the social interactions among the individuals, and constructing the statistical mechanics models for crowd dynamics and social physics systems. The research emphasis will be put on finding the dynamical rules for the panic crowd behaviors in thunderbolt emergency and investigating the effects of all kinds of artificial intervention strategies for crowd dispersal. Based on the non-equilibrium statistical mechanics, the nonlinear stability analyses and the feedback control method, This project will study the determinism and random rules for the panic behaviors of the large-scale crowd, construct the evacuation behavioral models for the panic crowd in emergency, select the best dispersal strategy, provide the scientific theory basis for the panic group behavior analysis and the strategy design for emergency evacuation.

本世纪的第二个十年是复杂性科学深入各个领域，不断有新的突破而面临更大挑战的时代。许多重要复杂系统（生命、经济和社会体系：互联网、人类行为、城市交通流、社交网络）正迫切等待我们运用统计物理研究方法，通过实证统计与理论模型分析相结合的手段进行全新的认识和探索。本项目将围绕人类重复行为的时空分布特征与突发事件下恐慌人群行为中的统计物理问题，针对人类自身行为的规律特性及社会个体之间的相互作用等复杂现象，构建人群动力学和社会物理系统的统计力学模型，深入挖掘和理解各种决定人类群体这一复杂系统演化过程的基本物理机制与统计物理规律。着重研究突发事件下恐慌人群行为的动力学规律和各种人为干预疏散策略的效果。拟基于非平衡统计力学、非线性稳定性分析和反馈控制手段，研究大规模人群恐慌行为的确定性与随机性规律，构建人群紧急疏散模型，甄选最佳疏散方案，为恐慌群体行为分析与疏导策略设计提供科学理论依据。

本项目研究了突发事件下恐慌人群行为的动力学规律和人为干预疏散策略。基于非平衡统计力学、非线性稳定性分析和反馈控制手段，及人类行为动力学，研究了大规模人群恐慌行为的确定性与随机性规律，构建人群紧急疏散行为模型，甄选出最佳疏散方案，为恐慌群体行为分析与疏导策略设计提供科学理论依据，对于实现非常规突发事件下降低伤亡事故发生率及减少经济损失的国家重大战略目标具有重要意义。在研究过程中，发挥了本项目组在自然群集动态分析、人类行为动力学、复杂系统和复杂网络方面的研究优势。本项目的研究，基本达到原设的科学目标：深刻提炼恐慌群体行为的动力学机理；探索和研究个体动态与个体间信息交互对于恐慌群体“涌现”行为的作用、人群在危机传播中的竞争与协作机制、疏散群体构架优化、建筑物逃生通道和辅助设施合理设计，并对于疏散引导、疏散资源合理配置、危险趋势迅速遏制、拥挤踩踏伤亡有效缓解等应急管理实际应用提出科学的建议。2013-2016 年间共发表论文74篇