An intelligent supervision system for open loop controlled processes

Abstract

On-line manufacturing systems for continuous production are traditionally checked by constant time-based inspection and statistical process control processes. In the monitoring operation of large scale process plants, it is important to detect and locate process variables as they can significantly modify the operation and quality of the products involved. The present paper explores an alternative for monitoring these systems that uses the open loop control paradigm. The proposed method is defined and presented as an application for one particular industrial process--rubber die extrusion. The advantages of this method for this particular application, including its implementation throughout a multi-agent system, compatible with the HoloMAS paradigm for managing, modeling and supporting complex systems, are also discussed.