CAAI—a cognitive architecture to introduce artificial intelligence in cyber-physical production systems

Abstract

This paper introduces CAAI, a novel cognitive architecture for artificial intelligence in cyber-physical production systems. The goal of the architecture is to reduce the implementation effort for the usage of artificial intelligence algorithms. The core of the CAAI is a cognitive module that processes the user’s declarative goals, selects suitable models and algorithms, and creates a configuration for the execution of a processing pipeline on a big data platform. Constant observation and evaluation against performance criteria assess the performance of pipelines for many and different use cases. Based on these evaluations, the pipelines are automatically adapted if necessary. The modular design with well-defined interfaces enables the reusability and extensibility of pipeline components. A big data platform implements this modular design supported by technologies such as Docker, Kubernetes, and Kafka for virtualization and orchestration of the individual components and their communication. The implementation of the architecture is evaluated using a real-world use case. The prototypic implementation is accessible on GitHub and contains a demonstration.

Cognitive capabilities for the CAAI in cyber-physical production systems

“…The result is a blueprint for a domain-specific class of systems and facilitates communication and knowledge transfer. We evaluated proposed reference architectures for the application of AI in CPPS from the field of automation and cognitive sciences in previous work [7] and proposed CAAI [15] as an alternative that was more focused on software development for the application of AI in CPPS. Thus, we build on previous work, e.g., [46] and [19], to develop a three-tier architecture that enables an adaptable system with rapidly changing configurations.…”

“…Artificial intelligence (AI) in cyber-physical production systems (CPPS) can help to significantly reduce costs [22], but its implementation requires expert knowledge, and thus might be cost-intensive [15]. To tackle these challenges, a modular and extendable Cognitive Architecture for Artificial Intelligence (CAAI) in CPPS was introduced in previous work [15]. The proposed architecture can be used by companies of all sizes, even SMEs, as the implementation is possible on an existing computing cluster, smaller edge devices, and also the cloud.…”

“…The work at hand describes the cognitive module and its functionality, which is further referred to as Cognition, in more detail than [15]. The task of the Cognition is to propose candidate pipelines with proper parameters, compute several pipelines in parallel, evaluate the pipeline quality, and switch to promising pipelines during the operational phase, e.g., if they are likely more efficient w. r. t. accuracy or resource consumption.…”

A Modular Federated Learning Architecture for Integration of AI-enhanced Assistance in Industrial Maintenance - A novel architecture for enhancing industrial maintenance management systems in the automotive and semiconductor industry.

“…The objective of human-centered cognitive systems is to automate manufacturing processes further and assist workers in their cognitive tasks (Fischbach et al 2020). This can be achieved by using the untapped potential of combining unstructured and structured data in order to extract hidden knowledge.…”

“…AI systems enable the improvement of various Key Performance Indicators (KPI), ultimately leading to a reduction in costs and optimizing plant management in smart factories (Fink et al 2021;Passath et al 2021). However, their development, implementation and deployment require a high level of expert knowledge and cost-intensive (computational) resources as well as reliable infrastructure (Fischbach et al 2020). In addition to expert and prior knowledge, the available data is also an essential criterion for successful implementation of an AI system.…”