MOBEDIC has a generic architecture for modelling decisions. It uses scripts to define typical actions or procedures associated with particular conditions: people adopt particular scripts according to the situations in which they find themselves. Scripts are organised into sets, each set applying to a specific situation. Each script in the set has an associated probability which determines the likelihood of that script being selected from the set. Hence, the non-deterministic behaviour of people can be modelled.
MOBEDIC represents the "brain'' of the automata in the Movement Model. It is responsible for making decisions and for instructing an automaton to carry out some action. Making a decision is a two stage process that involves selecting a script and then executing it. Role, current activity, perception of the environment and training constrain the number of possible actions and hence the applicable script sets. Of course there may be more than one script set defined for a given situation. In such a case, all competing script sets are evaluated to check their degree of applicability to the current situation, and the most suitable one is selected. A single script is then chosen from this set based on its probability.
The Movement Model broadcasts information such as the presence of smoke or fire, the occurrence of alarms and an automaton's current position. Just as humans have to block out irrelevant information, MOBEDIC filters this information before analysing it and modifying an individual's perception of the world.