Modeling Emergency Managers’ Hurricane Evacuation Decisions
Modeling Emergency Managers’ Hurricane Evacuation Decisions
复制标题
对应急管理人员的飓风疏散决策进行建模
DOI：
10.3141/2604-10
复制
发表时间：
2017
影响因子：
1.7
通讯作者：
C. Wilmot
中科院分区：
文献类型：
--
作者：
Ravindra Gudishala;C. Wilmot
应急管理和决策支持系统(EMDSS)工具在帮助应急管理人员在一个司法管辖区受到风暴威胁时就人员转移到安全地带作出重要决策方面发挥了重要作用。EMDSS的重要组成部分之一是疏散需求模型，该模型预测家庭在受到风暴威胁时是否以及何时疏散。该模型的一个关键输入是应急经理发布疏散通知的决定。现有的数学模型没有根据飓风预报和其他背景因素来预测应急管理者是否以及何时会发布疏散通知。为了填补这一空白，这项研究试图开发一种模型，用于预测当一个司法管辖区受到风暴威胁时，应急经理是否以及何时会发布疏散通知。来自风暴后评估调查和报纸档案的数据被用来检索过去疏散经理对45个沿海县或教区的五场风暴所做的决定。然后，这些数据被用来通过使用时间依赖的序贯Logit范式来开发离散选择模型。五个独立的预测变量-风暴潮、清除时间、登陆时间、飓风类别和一天中的时间-被发现是应急管理人员决策的良好预测变量。该模型可用于应急管理人员估计其他应急管理人员在面临疏散决策时如何决定疏散一个地区。该模型还可以帮助从事飓风疏散行为建模和理解的研究人员和从业者。
Emergency management and decision support system (EMDSS) tools play an important role in assisting emergency managers with making important decisions about the movement of people to safety when a jurisdiction is threatened by a storm. One of the important components of an EMDSS is an evacuation demand model that predicts whether and when households will evacuate when they are threatened by a storm. A critical input to that model is an emergency manager's decision to issue an evacuation notice. No existing mathematical models predict whether and when an emergency manager will issue an evacuation notice on the basis of a hurricane forecast and other contextual factors. To fill this gap, this research study sought to develop a model that would predict if and when an emergency manager would issue an evacuation notice when a jurisdiction was threatened by a storm. Data from poststorm assessment surveys and newspaper archives were used to retrieve past decisions made by evacuation managers for five storms in 45 coastal counties or parishes. The data were then used to develop a discrete choice model by use of the time-dependent sequential logit paradigm. Five independent predictor variables—storm surge, clearance time, time to landfall, hurricane category, and time of day—were found to be good predictors of the decisions made by emergency managers. This model could be useful to emergency managers to estimate how other emergency managers decide to evacuate an area when they are faced with an evacuation decision. The model could also benefit researchers and practitioners engaged in modeling and understanding hurricane evacuation behavior.