Graph-enabled cognitive digital twins for causal inference in maintenance processes

Abstract

AbstractThe increasing complexity of industrial systems demands more effective and intelligent maintenance approaches to address manufacturing defects arising from faults in multiple asset modules. Traditional digital twin (DT) systems, however, face limitations in interoperability, knowledge sharing, and causal inference. As such, cognitive digital twins (CDTs) can add value by managing a collaborative web of interconnected systems, facilitating advanced cross-domain analysis and dynamic context considerations. This paper introduces a CDT system that leverages industrial knowledge graphs (iKGs) to support maintenance planning and operations. By employing a design structure matrix (DSM) to model dependencies and relationships, a semantic translation approach maps the knowledge into a graph-based representation for reasoning and analysis. An automatic solution generation mechanism, utilising graph sequencing with Louvain and PageRank algorithms, derives feasible solutions, which can be validated via simulation to minimise production disruption impacts. The CDT system can also identify potential disruptions in new product designs, thus enabling preventive actions to be taken. A case study featuring a print production manufacturing line illustrates the CDT system's capabilities in causal inference and solution explainability. The study concludes with a discussion of limitations and future directions, providing valuable guidelines for manufacturers aiming to enhance reactive and predictive maintenance strategies.KEYWORDS: Cognitive digital twinsindustrial knowledge graphscausal inferencesemantic modellingquality assurancemaintenance AcknowledgementsThe authors would like to acknowledge the professional advice of Teo Man Ru and Tiong Je Min from Tetra Pak Jurong Pte Ltd.Disclosure statementNo potential conflict of interest was reported by the author(s).Data availability statementData is not available due to commercial restrictions. Due to the sensitive nature of this study, the participants of this study did not consent to public sharing of their data, so support data is not available.Additional informationNotes on contributorsKendrik Yan Hong LimKendrik Yan Hong LIM is a Ph.D. candidate at the School of Mechanical and Aerospace Engineering, Nanyang Technological University (NTU), Singapore and a senior research engineer at Singapore’s Agency of Science and Technology (A*STAR). He holds a bachelor’s degree in mechanical engineering from NTU, and a master’s degree in Industry Engineering from Chiba University, Japan. His research interests include engineering informatics, digital twins, and smart product-service systems.Theresia Stefanny YosalTheresia Stefanny Yosal is currently working as an equipment engineer at a manufacturing company. She holds a bachelor’s degree in mechanical engineering from Nanyang Technological University (NTU), Singapore. Her research interests are digital twins, product design and development, and manufacturing.Chun-Hsien ChenChun-Hsien Chen is a Professor at the School of Mechanical & Aerospace Engineering of Nanyang Technological University, Singapore. His research interests are product design and development, engineering/design informatics for managing/supporting digital design and manufacturing, and human factors and management of human performance. He has more than 280 publications in these areas. He is Co-Editor-in-Chief of Advanced Engineering Informatics (ADVEI).Pai ZhengPai Zheng is currently an Assistant Professor, Wong Tit-Shing Endowed Young Scholar in Smart Robotics, and Lab-in-Charge of Digitalised Service Laboratory in the Department of Industrial and Systems Engineering, at The Hong Kong Polytechnic University. He received the Dual bachelor’s degrees in mechanical engineering (Major) and Computer Science and Engineering (Minor) from Huazhong University of Science and Technology, Wuhan, China, in 2010, the master’s degree in mechanical engineering from Beihang University, Beijing, China in 2013, and the Ph.D. degree in Mechanical Engineering at The University of Auckland, Auckland, New Zealand in 2017. His research interest includes human-robot collaboration, smart product-service systems, and smart manufacturing systems. He serves as the Associate Editor of Journal of Intelligent Manufacturing and Journal of Cleaner Production, Editorial Board Member of Journal of Manufacturing Systems, Advanced Engineering Informatics and Journal of Engineering Design, and Guest Editor/Reviewer for several high impact international journals in the manufacturing and industrial engineering field.Lihui WangLihui Wang is a Chair Professor at KTH Royal Institute of Technology, Sweden. His research interests are focused on cyber-physical systems, human-robot collaboration, and brain robotics. He is the Editor-in-Chief of International Journal of Manufacturing Research, Journal of Manufacturing Systems, and Robotics and Computer-Integrated Manufacturing. In 2020, he was elected one of the 20 Most Influential Professors in Smart Manufacturing by SME.Xun XuXun Xu is a professor of Smart Manufacturing at the Department of Mechanical and Mechatronics Engineering, The University of Auckland. He has been working in the field of intelligent manufacturing solutions for over 30 years. Dr. Xu is the Director of the Laboratory for Industry 4.0 Smart Manufacturing Systems (LISMS). His current research focus is on Industry 4.0 technologies, e.g. smart factories, digital twins, and cloud manufacturing. Dr. Xu is a Fellow of ASME. He was recognised by the Web of Science as a Clarivate™ Highly Cited Researcher in 2020. In the same year, he was named among of the ‘20 Most Influential Professors in Smart Manufacturing’ by the Society of Manufacturing Engineers (SME).