Applying intelligent agents technology in a collaborative work environment Nikos Karacapilidis, George Vouros a and John Darzentas b Industrial Management Lab, MEAD, University of Patras, 26500 Rion Patras, Greece. E-mail: nikos@mech.upatras.gr a Dept. of Information and Communication Systems, University of the Aegean, 83200, Karlovasi, Samos, Greece. E-mail: georgev@aegean.gr b Dept. of Mathematics, University of the Aegean, 83200, Karlovasi, Samos, Greece. E-mail: idarz@aegean.gr Received 15 June 1999; received in revised form 12 January 2002; accepted 1 September 2002 Abstract The current state of the art in computer-mediated collaborative work is mainly limited by the fairly primitive discourse structures that most asynchronous conference systems implemented so far provide and because they lack consensus seeking and problem solving abilities. To address these issues, we first propose a refined model of the discourse acts met in a collaborative work environment. In the sequel, using this model in the context of an already implemented system, we apply intelligent agents technology in order to enhance and facilitate collaborative work. The overall approach combines concepts from the AI and CSCW disciplines, while aims at giving a highly active role to the participants involved in such environments. Keywords: computer-supported collaborative work, human-computer interaction, intelligent agents, knowledge representa- tion, argumentation, discourse acts Introduction The current state of the art in computer-mediated collaborative work is mainly limited by the fairly primitive discourse structures that most asynchronous conference systems implemented so far provide, where the underlying discourse acts are basically of the comment or reply type (Turoff et al, 1999, Vouros 2000b). In addition, most such systems lack consensus seeking abilities and decision making methods (Hurwitz and Mallery, 1995). From an OR point of view, multiple criteria decision making (MCDM) methods and Game Theory have been the most used approaches in the development of collaborative work environments (Jarke, 1986). MCDM methods provide a means of integrating multiple views of a problem and support both quantitative and qualitative criteria. Most of these methods can be interactive, allowing for easy Intl. Trans. in Op. Res. 10 (2003) 13–31 # 2003 International Federation of Operational Research Societies. Published by Blackwell Publishing Ltd.