43rd Congress of the European Regional Science Association: "Peripheries, Centres, and Spatial Development in the New Europe", 27th - 30th August 2003, Jyväskylä, Finland
This paper analyses the multi-agents systems that are now considered the best tool to simulate and study real world. We review the main characteristics of a multi-agents system, namely interactions and cooperations of agents, communications and behaviours between them and finally the schedule of actions and jobs assignment to agents. The multi-agents system approach is increasingly applied in social and economic sciences; so we study mainly the territorial applications. In these applications new characteristics arise from the consideration of territory (land and space where the agents live or territory as an agent in itself, that evolves in the time). We study possible new applications of multi-agents applied to the territory (for instance, to define town planning policies or to locate dangerous facilities). Furthermore we study new tools to make operational multi-agents systems (mainly Swarm, the toolkit of Santa Fe Institute). With Swarm we present two kind of territorial applications: with located agents (fixed in space) and with not located agents (moving in the space). Finally we show the results of these applications.