Building multi-agent systems for workflow enactment and exception handling

6Citations
Citations of this article
12Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Workflows represent the coordination requirements of various distributed operations in an organisation. Typical workflow management systems are centralised and rigid; they cannot cope with the unexpected flexibly. Multi-agent systems offer the possibility of enacting workflows in a distributed manner, via software agents which are intelligent and autonomous, and respect the constraints in a norm-governed organisation. Agents should bring flexibility and robustness to the workflow enactment process. In this paper, we describe a method for building a norm-governed multi-agent system which can enact a set of workflows and cope with exceptions. We do this by providing agents with knowledge of the organisation, the domain, and the tasks and capabilities of agents. This knowledge is represented with Semantic Web languages, and agents can reason with it to handle exceptions autonomously. © Springer-Verlag Berlin Heidelberg 2010.

Cite

CITATION STYLE

APA

Lam, J., Guerin, F., Vasconcelos, W., & Norman, T. J. (2010). Building multi-agent systems for workflow enactment and exception handling. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6069 LNAI, pp. 53–69). https://doi.org/10.1007/978-3-642-14962-7_4

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free