Tuesday, April 23, 2013

Intelligent Visualization and Information Presentation for Civil Crisis Management

Adrienko & Adrienko's paper "Intelligent Visualisation [sic] and Information Presentation for Civil Crisis Management" describes research conducted as part of the EU-funded project OASIS to develop methods for effective visualization support for situation analysis and management during times of crises, namely through the "Situation Manager" software module. The authors cite the major goals of the research are "to reduce the information load of the analyst, decision maker, or information recipient without omission of anything important and to ensure quick and accurate comprehending of the information" (Adrienko & Adrienko, 2007, p.889).

The research behind "Situation Manager" aims to create a generic crisis management system to support response and rescue operations in the event of large-scale disasters. Like intelligence products, visual analytic products should be deliverable in a timely fashion, while the information is relevant, and presented in a way that it is easily understood and therefore can be used by decision-makers. The authors suggest intelligence visualization requires reducing the information load on the recipient, display choices and designs that ensure quick and accurate recognition of meaning, and account for the characteristics of the medium used to view the information.

The authors suggest that intelligent visualization is used for two different purposes: to support the work of an analyst, planner, or decision maker, or to build an information presentation to send to a specific recipient.  It is unclear if the authors see these purposes as mutually exclusive, though it is unlikely. In the development of intelligent visualization, expert knowledge is needed in the emergency management domain ontology, generic roles involved in emergency situations and their information needs, and techniques and methods to manipulate and organize different types of data. The research uses literature on crisis management as a source of domain-specific knowledge and literature on data analysis, graphic design, and geographic visualization as domain-independent knowledge to create an expert system.

"Situation Manager," still in its early stages at the time of this article's publication, is a scalable vector graphics presentation that is interactive and includes the emergency management expert information and a problem-specific user interface. The system takes information about the crisis situation and the territory affected by the situation, essentially answering the what and where questions. The where aspect includes an impact zone which can be supplied or self-created. The software module then identifies the hazardous agents involved in the situation based on preexisting knowledge and potential secondary hazardous events caused by the first crisis event. This helps to locate objects or resources that need to be saved or protected in the impact zone to prevent further damage, including an estimate of the number of people living in the zone based on census data. If the object is people, the Situation Manager detects if people in the impact zone can escape danger without outside intervention. The module displays objects and dangers by degree of how critical each is through the sizes of symbols used on the maps, larger symbols representing a more critical object at the current time.

This research is particularly applicable to the intelligence field considering the blatant similarities between what the authors conclude are requirements for intelligent visualizations and the requirements we at Mercyhurst accept as necessary for an analytic/intelligence product. When reading the article I could specifically relate it to the work a classmate is doing to identify the potential for nuclear exposure caused by earthquakes in Iran.

The article itself is only a description of research being conducted, rather than methodologically reporting the findings and methods of the research. While the article suggests different sources of information used to create the expert system for visualization for crisis management, it would strengthen the presentation of the research to formally address all areas of data collection. Still, the authors do a good job of explaining the research in developing the "Situation Manger" in general terms and use good examples to communicate the use of such a tool. Though the module was not complete at the time of the article's publication, the authors provide a roadmap of implementation upon completion and suggest likely real-world applications.

It is interesting to me that the article does not address why visualization is a useful communication tool for decision-makers in general, not even referring to the extant literature on the subject which is extensive. Further, because the module was not complete at the time of publication, there is no record of successful implementation or feedback by users. Further analysis of this type of visual analysis and data visualization is necessary to strengthen the argument for the Situation Manager module.

Andrienko, N., and Andrienko, G. (2007). Intelligent Visualisation [sic] and Information Presentation for Civil Crisis Management. Transactions in GIS 11(6), 889-909. doi:10.1111/j.1467-9671.2007.01078.x


  1. I really liked this article and its approach to helping with crisis situations. One other thing I got from this article is that with the simplification of the huge data overload this makes projects more manageable and capable of having people walk into a crisis product and be able to understand it within a short amount of time. I also liked that added touch of increasing the icon sizing with importance. One other thing I picked up on is the morbid idea that if people are within a no-escape area after a crisis this tool can be used to engage authorities to effectively move the people it can save. Lastly, with your last comment paragraph I think the authors may have just assumed visualization is beneficial for decision makers.

  2. MK -- I agree with your assessment. This article and study is certainly relevant to the intelligence community, particularly when looking at potential disaster plans. While it is key to have plans solidified in writing, it is important to include visual elements for the decision makers, so that a point is further taken and more easily demonstrated. Particularly, when different factors can be changed, and the resulting visualization is altered accordingly.