An Ontology Driven Multi-agent System for Client Assignment in a Bank Queue

This paper presents an ontology driven multi-agent system that uses a negotiation process for decisionsupport in a Bank Queue. The system assists queue client assignment based on the client profile and the cashiers’ workload in order to guarantee a minimum time response in client attention. The multi-agent system has a direct positive impact in the quality of service. Simulations of service providers’ management are presented in order to optimize the use of the resources. Our ontological user profiling and multi-agent system approach can be easily extended and adapted to other domains by adding client profile characteristics and adapting agent behaviours. The ontology proved to be useful when sharing content between agents and performing semantic checks.