Major Disasters:
Major disasters, be they natural or man-made, are met by aggressive relief efforts on the part of a variety of organizations both public and private. The organizational response takes the form of integrated crises management units (ICMUs). ICMUs arise due to a pre-laid plan, negotiation, law or chance. The structure of these ICMUs is important because of the rapidly changing nature of disasters and their aftermath. Individual "first responders" (e.g. firefighting companies, police departments and emergency medical services) must be autonomous, yet still be able to take direction from the organization that has taken shape as the coordinating entity.
The objective of this research is to analyze the formation and dissolution of ICMUs so that disaster planners can use the information in determining the optimal response for their areas for a given emergency. Of particular interest was the determination of the quantity, content and quality of the information that flowed between the organizations. Individual rescue units must have enough information to follow the overall plan yet not be overloaded with information they do not need. In turn, the units must relay the status of their particular tasks to the emergent leading organization. The two way flow of this information will be shaped by the network form of the ICMU.
The events of September 11, 2001 left behind a large body of information regarding the activities of the integrated crises management units and first-responders. The pool of emergency responders included volunteers, charitable organizations and commercial companies, as well as numerous government agencies. Many of the organizations kept logs (e.g. the Federal Emergency Management Agency (FEMA)), formal reports and informal notes. In addition, key individuals can supply eye witness accounts and provide in-depth interviews. An important source of data is local newspaper reports which were the data used for this research.
September 12, 2001 to September 21 ,2001 editions of the Pittsburgh Post Gazette were scanned for articles relating to the crash of Flight 93. A list was made of key organizations and individuals (e.g. FBI, governor). When an organization or prominent individual on this list is mentioned in an article, the article is further scanned for other instances of entities on the list (e.g. FAA, Red Cross). The key assumption is made that mutual mentions indicate that the entities are in contact with each other.
The organizations and entities are entered into a matrix organized with newspaper articles as the rows and entities as the columns. The matrix, A, is transposed and multiplied by itself (ATxA) to give an organization by organization matrix. The intersections contain the number of times an entity X, is mentioned in the same article as entity Y. The diagonal gives the total number of times an organization is mentioned. This information is used to produce social network graphs. Using the graphs along with the original newspaper articles, a picture can be formed of who speaks to whom and who knows what. Having this knowledge is a first step in gaining a coherent picture of the total rescue organization's communication structure. By analyzing the graphs and the graph measures it will be possible to trace the communication paths which will explain in part how and why the IMCU took the form it did.
The graphs represent contacts between 70 organizations targeted as key entities in the clean-up in the aftermath of Flight 93's crash. The graph for September 11, for instance, clearly shows the FBI in charge with strong presences of the State Police and the Red Cross. By September 18th, the graph shows that the number of entities involved and the primary effort shifted from the crash site to returning Pittsburgh's transportation infrastructure to its pre-crisis state.