Coordinated multi-view graph analysis and improvement with applications in cyber-physical production systems engineering

Abstract

Graphs are a general class of models that consist of nodes and edges. In technical domains, such as discrete manufacturing, these graphs model real-world artefacts. The field of Cyber-Physical Production Systems (CPPS) Engineering, concerned with interconnected production plants, uses multi-view graphs to represent contributions coming from several engineering disciplines, such as plans for a motor in mechanical, electrical, and software/automation engineering. Multi-disciplinary engineering with dependencies between the models, that represent multiple views on the same real-world artefact, require coordination between engineers for analysing and improving linked multi-view graphs. This leads to challenges, founding the need for coordination. These challenges include the representation and linking of shared nodes, e.g., motors, in multiple engineering disciplines with shared properties, e.g., motor power. Furthermore, challenges exist in the coordination of tasks in a graph in CPPS Engineering, due to missing support for integrated coordination on graphs. This thesis proposes the Coordinated Model Analysis and Improvement (CMAI) approach to improve coordination between engineers in the field of CPPS Engineering. To represent models in CPPS Engineering, this thesis introduces the CMAI-Domain-Specific Language(DSL). Based on the CMAI-DSL, this thesis introduces the CMAI method. This method leverages the proposed approaches to improve coordination between engineers by providing an automated way of keeping data in sync between models and introducing coordination markers. These markers provide an efficient and integrated coordination tool on graphs. Furthermore, this thesis introduces a technical prototype, the Generic CPPS Modeling Editor. This thesis uses a qualitative evaluation to show how selected use cases in the field of CPPS Engineering can be conducted and profit off the introduced CMAI method.