Responding to large-scale disasters such as hurricanes and to the smaller-scale events that can accompany them (fires, structure collapses, and hazardous material leaks) involves a significant commitment of personnel and equipment resources, usually requiring the coordination of multiple agencies. Emergency managers--who control, coordinate, and manage the personnel and equipment allocated--need appropriate training in order to muster an effective response.
But what training is appropriate? To answer this question, we need an emergency management training system that will provide the automated tools to define and prioritize tasks to be trained, simulate the disaster in a realistic manner to drive a training exercise, provide results based on interactions among the trainees, and assess the results of decisions made in response to simulated events. These are the goals of the Plowshares Project, which we offer as a model for technology transfer from military to civilian use. he Plowshares Project the name of the technology transfer program that encompasses Janus and TERRA. It is a constructive war game simulation showing a map of a 60 x 60-km area along with a number of overlays (e.g., fire stations) and several hundred entities (e.g., emergency vehicles).
The response to a large-scale disaster in the United States is typically managed at the county level in the following way: The field units (e.g., fire trucks and fire fighters, ambulances and paramedics, police officers and their vehicles) communicate with their respective agency headquarters from the field via radio and telephone, providing information regarding emergency situations and receiving response instructions. The agency headquarters, or agency emergency operations centers (EOCs), may be located throughout the county or in the county EOC, a single facility devoted to and equipped for emergency management response.
It takes an extensive commitment of field units, personnel, and time to perform a training exercise based on this arrangement without using simulation. Because county governments are experiencing budget constraints, as are other government entities, less expensive and more efficient means for conducting training exercises are necessary.
HOW SIMULATION HELPS
The U.S. Army's Janus combat simulation model has been enhanced to support emergency management events, functions, and scenarios. The emergency management version of Janus, known as TERRA (Training for Emergency Rapid Resource Allocation), has been used initially to train officials responsible for controlling and coordinating a single jurisdictional response to a countywide emergency.
TERRA simulates an emergency, such as a hurricane or chemical spill, and models its effects on population and property. TERRA also simulates the individual field units (e.g., fire trucks and bulldozers) and the actions those units can take in response to the emergency.
In August 1995, the Plowshares Project staff conducted a proof of principle demonstration in order to exhibit the applicability of the TERRA software to a civilian emergency management training environment. This demonstration took place at the Orange County Emergency Operations Center in Orlando, Florida, with personnel from Orange County Fire and Rescue, Sheriff's Office, Public Works, Public Utilities, and Health and Human Services participating in the exercise.
At the beginning of the demonstration, the participants were briefed on the objectives and rules of the exercise. The agency EOC managers (the primary training audience) took their positions near the TERRA workstations, with the person designated as the TERRA interactor at the computer ready to interpret what was seen on the screen and relay this information to the agency EOC managers. Their cues as to the extent of the damage, which come from field units in an actual emergency, would come from the TERRA workstations during the exercise. The emergency support function personnel took their positions in the operations room, where they had access to messages by radio, operating in a manner similar to that in an actual emergency.
At the start of the exercise, participants received a description of the current situation, including weather information and location of the hurricane, which had passed by this time. As the exercise began, signs of damage were apparent on the TERRA workstations. The rubble, damaged buildings, and fires created by the hurricane took the form of icons on the map displayed on the computer screen. Once the incidents were relayed to the agency EOC managers, decisions were made regarding where and how to allocate resources to best manage the problems at hand. As more information came in during the exercise, it became necessary for ESFs to make decisions and coordinate responses with other agencies. The information was then passed down to the agency EOC managers, who had the TERRA interactor respond to the decisions by reallocating resources within TERRA as specified. Interagency cooperation is a vital training objective and was facilitated by the simulation. For instance, the participants from Public Works and Public Utilities had to coordinate to decide which agency would respond to a certain blocked road, based on the available response units in the area.
The scenario events were carefully planned by training experts on the project team using the STEP tool. The final scenario was designed to replicate the kinds of events with which the participating county agencies would most likely be faced during and following a hurricane. Those events were automatically generated by the TERRA software. Their sequence and location were chosen to tax the agencies' resources, thus forcing communication with other agencies and thereby maximizing the training benefit. The events included the presence of street debris; disturbance at a shopping center; shelter collapse from a tornado; house fire, tornado, or looter; power loss at a sewage treatment plant with sewage spill; train derailment with chlorine leak; chlorine fumes at shelter; escaped prisoners; sewage spill with road washout; tree down; auto accident with entrapment; truck accident with fuel spill; and helicopter crash on the interstate.
Like all simulations, TERRA generates only a small portion of the stress of the real situation (an actual emergency) or even a realistic field exercise. It does offer a number of beneficial training elements, including realistic communication (it uses actual communication channels such as radios and telephone); practice in allocating resources, such as vehicles, units, and equipment; and, most of all, emphasis on teamwork skills. The tempo of the exercise can be increased by both speeding up the simulation and increasing the number of events the participants must encounter. This does place additional stress on participants, though not of the same type as danger encountered in a real emergency. In general, participants indicated that several hours' involvement in the exercise was tiring.
The proof of principle demonstration validated the basic precepts of the project: first, that combat simulation is a feasible starting point for developing an emergency management exercise driver; and, second, that a computer simulation offers benefits over traditional scripted manual or paper exercises.
In order to determine whether the training objectives have been met, an evaluation must be performed. Typically, two types of evaluation are used to determine training effectiveness when using simulation: an after-action review and quantitative analysis.
In an after-action review, participants review selected components of the exercise, and a critique is performed, led by a senior member of the agency leading the exercise. This review should include a visual aid to show the replay to all exercise participants.
A spokesperson for each training group selects an event that is critical to meeting the training goals. The spokesperson defends the actions taken during the training exercise or uses the opportunity to describe the lessons learned for the benefit of all the trainees. Constructive criticism of the tactics may be given by any member of the group, but particularly the subject matter expert, and suggestions are made as to how they might perform better in the future.
Following our proof of principle demonstration, an after- action review took place with the exercise director presiding. The agency EOC managers reported on an event that they felt was significant for their group. These events, which had been recorded by the simulation during the exercise, were rerun using TERRA Replay and projected on a screen for general viewing. Other trainees had the opportunity to discuss the reasons that certain tactics were used and make suggestions about how the agencies' resources could be better allocated in future training exercises.
The other form of evaluation is of a more quantitative nature and ties performance to training objectives. This is done by using measures of effectiveness that are related to the training tasks and incorporated into the simulation. For example, speed of response is a general measure of effectiveness. A specific, related measure is speed to dispatch resources after the start of an event, such as dispatching fire trucks to a fire.
From this experience it is clear that valuable emergency management training can be accomplished in a command post exercise format and that existing military simulations can be adapted for that purpose.
One goal of any training exercise is that the medium that delivers the training be as "transparent" as possible to the training audience. In other words, the exercise should possess the necessary realism so that the training delivery system enhances rather than interferes with the training experience. This was achieved through close coordination with subject matter experts, high-quality training exercise design, and software modifications that enabled the most important events (as defined by the training audience) to be simulated. in future training exercises.
[Extracted from "From Battlefield to Emergency Management," by Mary P.
Slepow, Mikel D. Petty, & J. Peter Kincaid, ERGONOMICS IN DESIGN, October
1997, Vol. 5, No. 4. Copyright 1997 by Human Factors and Ergonomics
Society, P.O. Box 1369, Santa Monica, CA 90406-1369; 310/394-1811, fax