A Knowledge‐Based Approach for Understanding and Managing Additive Manufacturing Data

Additive manufacturing (AM) is an innovative production approach that has gained significant attention due to its ability to overcome many limitations associated with traditional manufacturing techniques. As a consequence of efforts to optimize various AM processes, especially across different methods, a vast amount of data is either utilized (e.g., material properties, printer specifications, and process settings) or generated (e.g., monitoring data during printing, slicing strategies, and parameter configurations). Effectively managing, understanding, and retrieving information from this data remains a major challenge. The data often exhibits complex interrelationships and is distributed across heterogeneous sources, making it difficult for researchers and industry professionals to extract meaningful insights or make informed decisions. To address these challenges, we propose a knowledge‐based approach designed to support the structured management of AM data. The core of this approach is a modular ontology ( PBF‐AMP‐Onto ), which serves as a semantic foundation for integrating diverse data sources, enabling semantic querying, and supporting decision‐making systems. This ontology facilitates semantics‐aware data management, enhances the interpretability of AM processes, and contributes to the optimization of manufacturing outcomes. In this paper, we focus on one of the most advanced AM techniques, powder bed fusion (PBF), with a particular emphasis on electron beam (EB‐PBF). To validate the feasibility and practical utility of our approach, we constructed a knowledge graph using a workbench ( PBF‐AMP‐KG Workbench ) and based on our ontology using data from real‐world EB‐PBF use cases. We then demonstrate how domain‐relevant queries, such as those concerning process parameters, material behavior, and machine settings, can be answered efficiently using this knowledge graph, showcasing its potential to support researchers in navigating and leveraging AM data more effectively.