Modern production systems face enormous challenges due to rising customer requirements resulting in complex production systems. The operational efficiency in the competitive industry is ensured by an adequate production control system that manages all operations in order to optimize key performance indicators. Currently, control systems are mostly based on static and model-based heuristics, requiring significant human domain knowledge and, hence, do not match the dynamic environment of manufacturing companies. Data-driven reinforcement learning (RL) showed compelling results in applications such as board and computer games as well as first production applications. This paper addresses the design of RL to create an adaptive production control system by the real-world example of order dispatching in a complex job shop. As RL algorithms are "black box" approaches, they inherently prohibit a comprehensive understanding. Furthermore, the experience with advanced RL algorithms is still limited to single successful applications, which limits the transferability of results. In this paper, we examine the performance of the state, action, and reward function RL design. When analyzing the results, we identify robust RL designs. This makes RL an advantageous control system for highly dynamic and complex production systems, mainly when domain knowledge is limited.

No abstract

Structured light scanners for three-dimensional surface acquisition (SL scanners) are increasingly used for dimensional metrology. The optical configuration of SL scanners (distance to object and baseline width) influences the triangulation process, on which the scanners’ measurement principle relies. So far, only a limited number of studies has investigated the optical configuration’s influence on the accuracy of a SL scanner. To close this gap, this work presents a design of experiment in which the optical configuration of a SL scanner is systematically varied and its influence on the accuracy evaluated. Further, tactile reference measurements allow to separate random from systematical errors, while a special test specimen is used in two different configurations to ensure general applicability of the findings. Thus, this work provides support when designing a SL scanner by highlighting which optical configuration maximizes accuracy.

Inspektionsprozesse werden im Remanufacturing auch heute noch vorwiegend manuell durchgeführt, da die Einschätzung des Qualitätszustands von rückläufigen Gebrauchtprodukten komplex und damit schwer zu automatisieren ist. Dies ist darauf zurückzuführen, dass Abnutzungsgrade, Deformationen und Schädigungen eine individuelle Bewertung des Gebrauchtprodukts nach sich ziehen und somit schwer standardisierbar sind. In diesem Beitrag werden die Anforderungen an ein System für die Bewältigung der Herausforderung der automatisierten Inspektion im Remanufacturing abgeleitet. Darauf aufbauend wird das Konzept einer Befundungsstation, welches diese Anforderungen erfüllt, präsentiert und Anwendungsfälle im Rahmen des von der Carl-Zeiss-Stiftung geförderten Forschungsprojekts „AgiProbot - Agiles Produktionssystem mittels mobiler, lernender Roboter mit Multisensorik bei ungewissen Produktspezifikationen“ vorgestellt.

No abstract