Federated multi-agent scheduling in Virtual enterprises

This paper presents an approach on how the cooperation among autonomous agents can be properly supported and enhanced by a federated database when supporting a distributed multi-agent scheduling scenario in a Virtual Enterprise, is envisaged. The innovative concept investigated in this paper considers that the data is not sent from one agent to the other via a high-level protocol, as is the case in the traditional push strategy, but rather through the pull strategy, via accessing the respective agents ' federated database. Thus, the high-level protocol is applied for coordination and activity control purposes only, separating the control from data, and preserving the privacy of each of them by means of the federated database-based approach. This work is partially developed in the context of the! NCO MASSYVE project and also uses some results produced in the ESPRIT PRODNET II project.