Adaptive self-learning distributed and centralized control approaches for smart factories

Procedia CIRP, vol. 104, pp. 1577–1582

Abstract

The increasing application of cyber-physical systems creates a manufacturing environment in which the technical requirements for distributed control approaches, self-learning systems and analytics of previously untapped data are given. While distributed control approaches are capable to evaluate this information locally and react immediately, centralized approaches react inertly to analyzed machine performance data. In this paper, we study the performance and ability to address the ever increasing challenges in industry of both types of control approaches within an established multi-agent based discrete event simulation.

Authors 2

1. Oliver Antons corresponding Aachen Chair of Management Science

   RWTH Aachen University

   Affiliation as printed

   Chair of Management Science, RWTH Aachen University, Kackertstraße 7, 52072 Aachen, Germany

2. Julia C. Arlinghaus

   Otto-von-Guericke-Universität Magdeburg

   Affiliation as printed

   Otto-von-Guericke University Magdeburg, Universitätsplatz 2, 31904 Magdeburg, Germany