Temporary network of enterprises: Managing complexity by a multi-agent architecture

Abstract Present evolution of manufacturing systems is characterized by multi-firm enterprises, often organized in multi-site networks where each firm can be managed with large autonomy. As the system complexity and the number of local decision makers grow, the uncertainty of the production management seems to grow also. This paper first aims to verify whether the increasing complexity of actual manufacturing systems is going to allow an increasingly greater autonomy to local decision makers (also called ‘agents’). The scope is to show when and how local decentralized decision makers could apply efficient controls without degrading global efficiency. Then, taking into account that the final aim of an enterprise is profit, a positive relation between global system efficiency and local agents autonomy is proved.