Ontologies in a Multi-Agent System for Automated Scheduling

Abstract

Multi-Agent Systems (MAS) have been successfully used in a wide range of applications such as robotics and e-commerce, and in particular in planning and scheduling. The aim of this paper is to present the interesting features that the use of ontologies in MAS offers. As an example, the development of a MAS for automated planning and scheduling in a University Research Group Scenario is shown in this paper. In this scenario, researchers are frequently proposed to attend internal meetings about several subjects such as lessons planning or research evaluations. Scheduling and negotiating meeting details such as time and location becomes highly complicated as the number of intended attendees increases. Moreover, there are usually conflicts about the use of some common resources such as portable computers or projectors. As can be seen, the scheduling problem that the MAS solves is very easy. So having solved it is not what is important about this paper. In contrast, what is important is the potential which a scheduler can schedule for the items whose description, for example, is on the web, and can read on it (without knowing a~priori) the logic of how the scheduling can be done.