Improved Situational Awareness in Emergency Management through Automated Data Analysis and Modeling

In this paper, we address the problem of situational awareness in the context of emergency management. Due to the widespread application of information technology, emergency managers face a new challenge in information overload, rather than the lack of information as in the past. These new tools deliver large volumes of data to the decision makers, often in real time. However, during the crisis, the task of identifying the key relevant information poses a real challenge and failure to do so can have disastrous consequences. To address this problem, we propose a model for the fusion of data from diverse sources, both real-time and static, to derive numerical measures that represent the status of a jurisdiction. In our approach, we combine the data available from various sources with human expertise to build customized models based on threats characteristic for particular jurisdictions. We developed a prototype based on data collected for several municipalities and implemented it as a computer model.