Artificial neural networks as tools for controlling production systems and ensuring their stability

Abstract

Models of artificial neural networks can be used to control a production system, and thus to ensure its stability. Such models are very useful tools, because they can be built quickly and easily. The only issue is a large amount of data needed in the neural network training process. However, in the era of common availability of IT systems, the parameterization and standardization of production processes is not a problem anymore. Contemporary production systems are mostly automated and metered. This paper presents a method for building a model of an artificial neural network for controlling a wire harness production system and determining its stability. © 2013 IFIP International Federation for Information Processing.