Controlling a steel mill with BOXES

We describe an application of the BOXES learning algorithm of Michie and Chambers (1968) to a large-scale, real-world problem, namely, learning to control a steel mill. By applying BOXES to a model of a skinpass mill (a type of steel mill), we find that the BOXES algorithm can be made to produce a robust controller relatively quickly. Various aspects of the BOXES algorithm are adapted for the to higher dimensionality and noise present in the skinpass mill. These changes are critically examined to find those which give a better controller.

• Type: Conference
• Year: 1993
• Where: MI
• Authors: Michael McGarity, Claude Sammut, David P. Clements
• Keywords: Artificial Intelligence, BOXES Algorithm, BOXES Learning Algorithm, MI 1993, Steel Mill