Two-Mode Threshold Graph Dynamical Systems for Modeling Evacuation Decision-Making During Disaster Events.
Two-Mode Threshold Graph Dynamical Systems for Modeling Evacuation Decision-Making During Disaster Events.
复制标题
用于灾害事件期间疏散决策建模的双模式阈值图动态系统。
DOI：
10.1007/978-3-030-36687-2_43
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Halim N, Kuhlman C
中科院分区：
文献类型：
--
作者：
Halim N, Kuhlman C
社会科学的最新研究结果表明，邻里效应在家庭疏散决策中起着重要作用。邻居疏散可以激励一个家庭疏散。另一方面，如果许多邻居撤离，那么个人或家庭决定撤离的可能性就会降低，因为他们害怕抢劫。使用网络上传播的标准模型，例如阈值模型，不能捕获这种行为。在这里，我们提出了一种新的图动力系统模型，2模式-阈值，它捕捉到了这种行为。我们研究了两种模式阈值在不同网络中的动力学特性，发现与标准阈值模型有显著差异。我们通过在弗吉尼亚州弗吉尼亚海滩的小世界网络上进行基于代理的仿真来演示我们的模型的实用性。我们使用它来了解该地区的疏散率，并评估该模型和不同初始条件对疏散决策动态的影响。
Recent results from social science have indicated that neighborhood effects have an important role in an evacuation decision by a family. Neighbors evacuating can motivate a family to evacuate. On the other hand, if a lot of neighbors evacuate, then the likelihood of an individual or family deciding to evacuate decreases, for fear of looting. Such behavior cannot be captured using standard models of contagion spread on networks, e.g., threshold models. Here, we propose a new graph dynamical system model,2mode-threshold, which captures such behaviors. We study the dynamical properties of2mode-thresholdin different networks, and find significant differences from a standard threshold model. We demonstrate the utility of our model through agent based simulations on small world networks of Virginia Beach, VA. We use it to understand evacuation rates in this region, and to evaluate the effects of the model and of different initial conditions on evacuation decision dynamics.