Cognitive capabilities for the CAAI in cyber-physical production systems

This paper presents the cognitive module of architecture for artificial intelligence (CAAI) in cyber-physical production systems (CPPS). The goal this is to reduce implementation effort (AI) algorithms CPPS. Declarative user goals and provided algorithm-knowledge base allow dynamic pipeline orchestration configuration. A big data platform (BDP) instantiates pipelines monitors CPPS performance further evaluation through module. Thus, able select feasible robust configurations process varying use cases. Furthermore, it automatically adapts models based on model quality resource consumption. also additional test from different classes. CAAI relies well-defined interfaces enable integration modules effort. Finally, an Docker, Kubernetes, Kafka virtualization individual as messaging-technology communication used evaluate a real-world case.