Cooperative decisions and actions in multi-agent worlds

Analysis, understanding and simulation of complex systems involving men and machines, entail the use of a set of interacting models more or less simple and software engineering tools strongly based on approaches such as distributed artificial intelligence and multi-agent systems. In this paper the authors propose to see such complex systems as cooperative distributed problem solvers and propose a two-stages model for decision making. This model allows the cooperative agents to structure a group-decision before any discussion work in order to minimize the communication and to dispose of a representation of their decision potential