Abstract
Decision Support Systems provide assistance to humans involved in complex decision-making processes. They are particularly relevant in domains where human operators have to take operational decisions regarding the management of complex industrial or environmental processes. In these applications, Decision-support agents are responsible for parts of the decision-making process in a (semi-)autonomous (individually) rational fashion: they collect and facilitate decision relevant data, but also provide advanced reasoning services to analyse the meaning of this information, and make it available to the decision-maker in the course of flexible, semantically rich dialogues. Next-generation agent-based DSS aim at an open architecture to allow for a dynamic integration of more and better decision support services. In this article, we present a multiagent service architecture as a first step towards this goal. Our approach draws upon the latest FIPA specifications for agent and service interoperability. By mapping FIPA key abstractions – its agent communication language, directory service and, in particular, its service description model – to the requirements of agent-based Decision Support Systems, we identify potential shortcomings, and show how they can be overcome. We illustrate our approach in the domain of bus fleet management.