KIT | KIT-Bibliothek | Impressum | Datenschutz

Instance-Specific Production Control in Circular Manufacturing Using Gaussian Mixture Models

Bail, Finn ORCID iD icon 1; Darijani, Ali ORCID iD icon 2,3,4; Hoffmann, Luisa ORCID iD icon 5; Engels, Maurice ORCID iD icon 1; Becker, Jens; Benfer, Martin ORCID iD icon 1; Stricker, Nicole; Heizmann, Michael 5; Beyerer, Jürgen 2,3,4; Lanza, Gisela 1
1 Institut für Produktionstechnik (WBK), Karlsruher Institut für Technologie (KIT)
2 Fakultät für Informatik – Lehrstuhl IES Beyerer: Interaktive Echtzeitsysteme (Lehrstuhl IES Beyerer: Interaktive Echtzeitsysteme), Karlsruher Institut für Technologie (KIT)
3 Institut für Anthropomatik und Robotik (IAR), Karlsruher Institut für Technologie (KIT)
4 Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung (IOSB)
5 Institut für Industrielle Informationstechnik (IIIT), Karlsruher Institut für Technologie (KIT)

Abstract:

Circular manufacturing systems are characterized by high uncertainty, as the condition of each returned product is individual, resulting in stochastic process behavior that directly affects production control decisions. Deterministic routing rules and fixed process parameters, which are common in linear production, are therefore insufficient. This paper proposes a probabilistic production control framework for circular factories based on Gaussian Mixture Models (GMMs). Products and their components are represented as probabilistic entities, with their process related characteristics modeled using GMMs trained on historical shopfloor data, prior process steps for the specific instance, and inspection results. For each product and component instance, the models predict the process duration, the probability of successful execution, and the risk of failure or damage. These probabilistic predictions enable instance specific decisions on process and station selection, as well as the adaptive choice between direct processing and intermediate inspection steps. By explicitly capturing multimodal condition states and uncertainty, the framework enables risk aware scheduling and routing decisions and improves robustness and responsiveness compared to deterministic control approaches. ... mehr


Zugehörige Institution(en) am KIT Institut für Anthropomatik und Robotik (IAR)
Institut für Industrielle Informationstechnik (IIIT)
Institut für Produktionstechnik (WBK)
Fakultät für Informatik – Lehrstuhl IES Beyerer: Interaktive Echtzeitsysteme (Lehrstuhl IES Beyerer: Interaktive Echtzeitsysteme)
Publikationstyp Proceedingsbeitrag
Publikationsjahr 2026
Sprache Englisch
Identifikator KITopen-ID: 1000198414
Erschienen in Conference on Circular Production Systems and Technologies (CCPT 2026)
Veranstaltung Conference on Circular Production Systems and Technologies (CCPT 2026), Karlsruhe, Deutschland, 08.06.2026 – 10.06.2026
Verlag Springer
Projektinformation SFB 1574/1 KLF, 471687386 (DFG, DFG KOORD, SFB 1574/1)
Bemerkung zur Veröffentlichung Accepted to "2026 Conference on Circular Production Systems and Technologies (CCPT)"
Schlagwörter Circular Manufacturing, Remanufacturing, Production Control
KIT – Die Universität in der Helmholtz-Gemeinschaft
KITopen Landing Page