Emergency management systems (such as forest fire fighting, traffic accidents,
earthquakes, pollution accidents) and military systems, have been classified by
Brehmer (1991d) and Hutchins (1990) as social dynamic systems, containing co-operating actors.
These systems can be seen as a special case of dynamic environment management systems,
in which the decision makers’ goal is to define and organise a set of co-ordinated
actions to reach a goal state and to limit the negative consequences on humans,
material and economic environment as far as possible. One characteristic of these
dynamic systems is that they are often so complex that the systems need some type
of hierarchical organisation (Brehmer 1991d).
An abstract view of the world can in it simplest form be a
world model consisting of a target system,
and a controlling system, based on the staff and its subordinate. See figure 1.
Figure 1. A complex dynamic system containing;
target system and a controlling system based on the staff and their subordinate units.
Example: A forest fire fighting domain.
The Target System
The target system is the system that is the target of the emergency organisation’s operations.
Examples of target systems include the fire in forest fire extinguish
operations or the enemy forces in military operations.
When civilians are in danger or when they can influence the target system,
they can also be classified as a part of the target system.
The target system can be classified as a complex dynamic system,
which changes both autonomously and as a consequence of actions performed on the system (Brehmer 1995).
The Controlling System
In an emergency organisation, the staff’s subordinates are the staff’s
tool in their task of controlling the target system.
Examples of a subordinate unit include fire fighting units,
ambulance units, and military units.
In large hierarchical organisations, such as a military brigade,
the emergency organisation can consist of several levels - companies, platoons, and so forth.
The controlling system is only semi-controlled by the staff and can as,
the target system, be viewed as a complex dynamic system.
It changes both as a consequence of the commands and on the subordinate’s own decisions.
The staff has to treat the emergency organisation as a complex dynamic system that should be controlled.
The commanders’ task in a staff is to command and control1 and control2 the organisation.
This means that they should collect information from their subordinates
so that they acquire a shared situation awareness.
On this basis they should plan and transmit orders to their subordinates
in order to direct and co-ordinate actions between the subordinate units.
The staff functions as decision maker only and does not operate directly on the target system.
The staff’s work can be described as tactical reasoning process,
as described for instance by Rogalski & Samurçay (1993 a, b).
Examples of tasks in their work are:
Gathering and sorting out relevant and consistent information.
Making hypotheses about the system.
Identifying the current state risks and possible evolution risks.
Defining goals and planning one or several appropriate strategies.
Distributing work and resources to the subordinate units.
Co-ordinating actions between different units.
The size of the staff can vary between a few persons in a small emergency
organisation to a large number of persons in a staff at some high level in a military hierarchy.
Complex Dynamic Systems
Emergency management systems and military systems have two important properties:
they are both complex dynamic systems and they are both controlled by distributed decision making.
The complexity and the dynamic and autonomous behaviour of the target
system and the emergency organisation make them difficult to predict and control.
The states of the system are changed both autonomously and as a function of
the actions that the decision maker makes.
The control of the systems has to occur in real-time and assessing a
complete current state of the system is difficult because the dynamics of
and the relations within the system are not visible.
Brehmer (1994) defines a complex dynamic system by the following criteria:
Gathering and sorting out relevant and consistent information.
Making hypotheses about the system.
The states of the system are changed both autonomously and as a function of the actions that the decision maker makes.
The decision makers have limited time to make their decisions.
It is difficult to see the current state of the system.
It is difficult to see what relations exist in the system.
The difficulties in understanding the current state of the system,
situation awareness, is an important problem and an important task
to be trained in in emergency management and military command and control training.
Emergency Management Training
An experienced commander’s or staff member’s knowledge about their task
and the system can be described as theoretical knowledge or experience-based knowledge.
The theoretical knowledge can be viewed as knowledge that can be learned through theoretical studies.
The experience-based knowledge is knowledge about how it is to work
in real situations and cannot be learned without experience the behaviour
patterns of the system and work involved.
Berkum & Jong (1991) point out that this kind of knowledge can be a good
target for simulation-based learning environments.
They refer to this type of knowledge as compiled conceptual knowledge,
and view it like ‘getting a feel for the underlying model’
(without needing to being able to articulate its internal workings).
Another way to describe such knowledge is in terms of Rasmussen’s
three levels of human action control; knowledge-based behaviour, rule-based behaviour,
and skill-based behaviour (Rasmussen 1983). See Figure 2.
Figure 2. A Three level description of human action control.
The skill-based behaviour represents sensorimotor performance during acts or activities that,
after a statement of an intention, take place without conscious control as smooth, automated,
and highly integrated patterns of behaviour.
The rule-based behaviour represents conscious activities controlled by stored
rules or procedures that may have been derived empirically on previous occasions.
Performance is goal-oriented, structured by feed-forward control through stored rules.
Often the goal is not explicitly formulated,
but is found implicitly in the situation releasing the stored rules.
The knowledge-based behaviour represents activities that are activated
by conscious plans delivered from explicit goals and mental models of the problem.
A common view on the differences between a novice and an expert is that the
expert can select and use the relevant information cues,
and works with the signs and symbols in an efficient way.
It is also well-known that an expert has a more compiled knowledge about the task and the system.
This means that the expert works more on the rule- and skill-based behaviour level.
What is hard to understand or to get a good feeling for through theoretical studies
is the dynamic changing behaviour of a complex dynamic system.
The goal of the training in an emergency management training system is often to let
the trainees experience the system so that they develop more compiled knowledge about their
tasks and the system.
Another important reason for using training system in emergency management systems
(such as forest fire fighting and military systems) is that it is
often too expensive or humanly impossible to practise in real-life situations.
In these cases we can use emergency management training systems to bridge the gap between
the theoretical studies and work in the real world.
Which means that the system is used to train people in acting and understanding
the situations that are common in their work.
These kinds of training systems are sometimes classified as tactical training systems.
The knowledge gained in typical emergency training systems can be in the
form of experiencing some dynamic behaviour of the target system or the
emergency organisation, and learning how to perform planning and co-ordination of actions.
Examples of common training goals for the commander and staff follow:
To know the concepts and the relation between the concepts,
and to experience the dynamic behaviour of the target system.
To be able to assess the current state of the target system and predict the effects of alternative actions.
Understanding the architecture of the emergency organisation.
Knowing how to use the organisation to get information and how to use and
co-ordinate the resources so that they regulate the target system.
Understanding other persons’ needs and goals,
and understanding the importance of shared frameworks and goals.
Understanding the doctrine and the goals of their work,
common strategies to solve the tasks, work procedures,
how to interpret and understand the current situation i.e. ‘situation awareness’,
identifying future critical situations, and suchlike.
To experience work under time pressure, with a high information load,
or to work with inconsistent or missing information.
The teaching strategies in this type of system are usually based on briefing,
debriefing, and selection of pedagogical training situations.
Briefing means that the trained staff is informed in advance on what they are supposed to learn.
After a training session comes debriefing where the teacher and the staff
discuss the activities performed in the training session.
Briefing and debriefing can be performed in many ways.
Basic steps in a briefing and debriefing session can be to explain the intended
training situations, pointing to critical information to be observed by the trainee,
(optional) suggesting alternatives actions, and providing feedback on performed actions.
A specialised debriefing format, sometimes used in military training situations,
is after-action review (AAR) which has the goal of maximising group interaction
to investigate what they learned and what it implicates for future engagements.
(Rankin, Centner & Crissey. 1995)
One important problem with this type of training system is the simulation of the surrounding world.
The simulation of the surrounding world should be so realistic that
it generates the same behaviour pattern as the real world.
The decision maker in the staff uses the behaviour pattern to generate a mental model of the world.
It is therefore important that this behaviour pattern describes
the dynamic behaviour that exists in the real world system.
The demands on the behaviour pattern are not so important for the
items that can be separated from the training goals,
but is very important for the behaviour that is connected to the training goals.
There is always a risk for the students to create a mental model of the system
that does not correspond to the real world, by learning some
simulation-specific behaviour (Gestrelius 1993, 1998).
The problem with the simulation is that the real-world systems
often are based on co-operating actors in a complex system.
Some important problems are:
Natural Language Interaction
The interaction between the staff and the co-operating actors in the emergency
organisation should in many cases mirror real situations,
which means complex natural language sentences.
A common way to solve this simulation problem is to simplify and restrict the communication,
or to use role-playing human-actors that play
the role of the different co-operating actors in the environment.
The use of role-playing actors makes it possible for the staff to express themselves in natural language.
The activity descriptions of the actors in the emergency organisation and
the target system are often complex and are therefore difficult to simulate.
The modelling of an actor can be simplified for a certain training goal or
the actor can be simulated by a human role-playing actor.
One goal with the activity simulation is that the combination of
all activities should generate the same dynamic behaviour patterns as
the real world system. Common strategies for this simulation problem
are to simplify the modelling of the actors, related some goal,
or to use role-playing people that simulate the actors.
Training Session Control
The simulation should have a control functionality that can be
adapted to support the pedagogical strategy.
The simulation of the activities and the communication should vary
depending on training goals, the knowledge of the students and the
commander’s and the staff’s operations.
A common simulation strategy in this type of training system is to compose
a training organisation that consists of training manager,
role-playing training assistants and some computer simulation support.
See figure 3. The tasks in this type of training system can be described thus:
Typical problems in this type of training organisation are the actors’
activity simulation and the pedagogical control of the training session.
The trained staff should work in their normal environment.
They should use the same tools in a training session as in a real situation.
The similarities with reality make the training more realistic.
The trained staff should mainly communicate with the training assistants;
in some situations they can also communicate with the training manager or the computer simulation system.
Their task is to realistically play the role of actors that exist in the simulation.
The training assistant’s task is to follow an abstract scenario,
communicate with the staff, react to the commands from the staff and to
the information they get from the computer simulation,
and follow the commands from the training manager.
The idea is that when some event occurs they should use the
computer simulation or start a mental simulation of the activity.
Mental simulation means that the training assistants simulate in
their heads the activities that take place in a normal real-world situation.
This means that they must build up the activity in their minds and reflect on what probably happens.
The main task for them is to keep the training session on track.
This means that they communicate with the training assistants and
describe important activities that are going to happen.
The training manager should also follow the activity of staff and
direct the session so that it generates proper training for the staff.
The training assistants and the training manager should have a computer system
that helps them to simulate physical phenomena and store data about the world.
Typical objects in the physical environment that can be simulated are target
systems such as fire or some unit in the controlling system.
In these situations the simulation can keep a state description of
the units and, for example, count down the fuel supplies as they are used up.
In more advanced simulations the computer system can also have
simple models of human actors so that it can generate a simple
simulation of actors that exist in the world.
As an example, in a forest fire domain a training assistant can be
responsible for simulating a fire-fighting unit, its chief,
and all the items and persons that are connected to the unit.
In this case one way to use a computer support tool is to let it
simulate the physical fire-fighting unit and some of the
activities carried out by the fireman connected to this unit.
The activity simulations are often performed by the training assistants’
mental simulation and by the physical simulation of the computer simulation.
The main problem for the training assistants is to perform a good mental
simulation of their units in the world. One problem is to keep all the
processes in mind and not to forget any important response from these activities.
There is always a risk of the training assistant becoming overloaded when he
or she is responsible for simulating a large part of an organisation.
It is important that the training assistants have a good understanding
of how the entities that he or she is supposed to simulate work.
The role-playing of the actors is based on the knowledge and
experience that the training assistants have.
The pedagogical control of the simulation is also important
for the success of the training sessions.
The training manager must in some way define the goal of the session,
in the form of work task, training situations or desired dynamic behaviour.
Here it is important that the training assistants understand the goals and
follow the directives of the training manager. One problem for the training
assistants is to co-ordinate and synchronise the training assistants´
activity so that desired training situations are generated.
The following references are used on this www page.
Berkum, J.J.A., & van, Jong T. de (1991).
Instructional Environments for simulations. In Education & Computing,
No. 6 (1991), pp. 305-358.
Brehmer, B. (1991d).
Organisation for Decision Making in Complex Systems. In Distributed Decision Making:
cognitive models for co-operative work, J. Rasmussen, B. Brehmer & J. Leplat (eds.) pp. 335-347.
John Wiley & Sons, New York. ISBN 0-471-92828-3.
Brehmer, B. (1994).
Verbal communication at seminar: Distributed Decision Making,
4 Feb. 1994, in the course, Higher psychology, at Linköping University, Sweden.
Gestrelius, K. (1993).
Pedagogik i simuleringsspel : Erfarenhets-baserad utbildning med överinlärningsmöjligheter.
Pedagogisk Orientering och Debatt 100. Lund University, Sweden.
Gestrelius, K. (1998).
Simulation and training games : Experiential Learning Saab Training Systems AB.
Hutchins, E. (1990).
The Technology of Team Navigation. In Intellectual Teamwork:
Social and Technical bases of Collaborative Work J. Galengher, R.E. Kraut, & C. Egido (eds.), 1990.
Rankin W.J., Centner F.C., & Crissey, M.J. (1995).
After Action Review and Debriefing Methods: Technique and Technology.
In Proceedings of the 17th Interservice /Industry Training Systems and Education Conference,
Albuguerque, New Mexico. Pp. 252-261, 1995.
Rasmussen, J. (1983).
Skills, Rules, and Knowledge; Signals, Signs, and Symbols, and Other Distinction in Human Performance Models.
In IEEE Transactions on System, Man, and Cybernetics, Vol smc-13, No. 3, May / June 1983.
Rasmussen, J. (1993).
Deciding and Doing: Decision Making in Natural Context.
In Decision Making in Action: Models and Methods,
G. A. Klein, J. Orasanu, R. Calderwood, and C. E. Zsambok (eds.),
ISBN 0-89391-794-X, pp. 158 - 171, 1993.
Reigeluth C.M. & Curtis R.V. (1987).
Learning situations and instruction models. In Instructional Technology: Foundations.
R.M. Gagne & R. Glaster, (eds.), pp. 175-207. Hillsdale, NJ: Lawrence Erlbaum. 1987.
Rogalski, J.,& Samurçay, R. (1993a).
A Method for Tactical Reasoning (MTR) in Emergency Managment:
Analysis of Individual Acquisition and Collective Implementation.
In the book Distributed Decision Making: cognitive models for co-operative work.
J. Rasmussen, B. Brehmer & J. Leplat. (eds.), ISBN 0-471-92828-3.
John Wiley & Sons, New York. pp. 287-298. 1993.
Rogalski, J.,& Samurçay, R. (1993b).
Analysing communication in complex distributed decision-making.
Ergnomics, 36, pp. 1329-1343. 1993.