AgentSteel: an agent-based online system for the planning and observation of steel production

Sven Jacobi, Cristián Madrigal Mora, Esteban León Soto, Klaus Fischer

In: AAMAS Industrial Applications. International Joint Conference on Autonomous Agents and Multiagent Systems (AAMAS-2005) July 25-29 Utrecht Netherlands Seiten 114-119 ISBN 1-59593-093-0 ACM 2005.


The steel production of the German company Saarstahl AG, a global respected steel manufacturer, represents a Supply Chain which comprehends several time-critical and highly interference susceptible processes. Therefore, flexibility, robustness and fast reorganisation are indispensable requirements on a system responsible for the planning of production inside this Supply Chain to ensure acceptable costs and retain their respected position on this market. In this paper we present an agent-based generic solution for the planning and observation of the complete production process inside the steelwork Völklingen of Saarstahl AG. The presented system calculates solutions for given daily target schedules - based on the concrete orders by the customers - by a combined distributed online planning and scheduling algorithm, and, additionally, performs the observation of its realisation. Moreover, the presented real-time multiagent system supports the reorganisation after operational faults in this high dynamic context by suggesting new solutions to the responsible conductor.

Weitere Links

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence