ProQ-KG: Integrated Cyber-Physical Production System Knowledge Graph for Quality Issue Analysis

Analyzing complex quality issues in the context of modern production systems is a challenging task that requires the coordination of multiple stakeholders and the integration of their knowledge across multiple domains. This heterogeneous knowledge necessary for quality issue analysis can be organized based on products, processes, and production system resources (PPR) on the one hand, and potential failure modes and effects (FMEA) on the other hand. In this paper, we introduce an integrated ontology that unifies Product, Process, Resource (PPR) and Failure Mode and Effect Analysis (FMEA) modeling approaches, and to provide a foundation for (i) the construction of integrated FMEA-PPR knowledge graphs, and (ii) the coordination of quality issue analysis among heterogeneous domain experts using knowledge-graph-based methods. Furthermore, we propose ProQ-KG, a knowledge-graphbased framework that facilitates quality issue analysis and multi-view coordination across diverse stakeholders. We implement our KG-based quality issue analysis approach and demonstrate its feasibility by means of three real-world use-case applications in collaboration with an industrial partner from the automotive industry. Our evaluation shows that the integrated approach makes it easier for stakeholders to structure and share their mental models as well as to navigate the integrated knowledge to identify root causes of quality issues.