A pattern-recognition-based method for controlling a multi-loop production system

The control of an equipment-intensive production system in the electronics industry is a very complicated task due to the complex process flow, especially the necessity for several loop sequences of the same processes. This paper describes the development of a pattern-recognition-oriented control-decision support system (CDSS) in order to set the best control strategy and an application for a circuit board fabrication system. The CDSS has been designed to be flexible to adapt to a change of market demands. This procedure is based on the state table concept. A combined discrete-event/process-interaction simulation model for the production system has been developed in order to provide basic data for the CDSS. The simulation analysis evaluates the alternatives of the control strategy on the efficiency performances, such as throughput and WIP.