Human knowledge centered maintenance decision support in digital twin environment

International audienceThe transition to Industry 4.0 has improved factories by improvingthe manufacturing process. With increasing automation, awareness ofthe role of humans in industrial maintenance management is alsoimportant for the realization of the Industrial Internet of Things(IIoT). Today’s smart factories use data from various sources forextraction of valuable insights to improve manufacturing processesand avoid failures. These improvements also add to the complexityof resolving different maintenance issues faced during themanufacturing process. There is a need to leverage untapped humanknowledge in Maintenance Work Orders (MWOs) to handle these complexchallenges using state-of-the-art Natural Language Processing (NLP)techniques. The development of Industry 4.0 technologies is leadingto a growing interest in using digital twins in many sectors.Digital twin-based services are revolutionizing design,manufacturing, product use, and maintenance (diagnosis, prognosis,and decision-making). This paper proposes a human knowledgecentered intelligent maintenance decision support. The proposedservice can find solutions to new maintenance problems usingknowledge in past maintenance records in a digital twinenvironment. The architecture of the proposed service and itsconnections with Physical Space (PS), Virtual Space (VS) andDigital Twin Data (DTD) are presented in this paper. Theperformance of the service is validated using a case study on anopen-source dataset of real MWOs from mining excavators. Resultsindicate that state-of-the-art NLP techniques can be used toprocess human knowledge in MWOs and generates interesting patterns.This study is also a step forward towards application of TechnicalLanguage Processing (TLP) in a smart manufacturing setup