A unified framework for digital twin development in manufacturing

The concept of digital twin (DT) is undergoing rapid transformation and attracting increased attention across industries. It is recognised as an innovative technology offering real-time monitoring, simulation, optimisation, accurate forecasting and bi-directional feedback between physical and digital objects. Despite extensive academic and industrial research, DT has not yet been properly understood and implemented by many industries, due to challenges identified during its development. Existing literature shows that there is a lack of a unified framework to build DT, a lack of standardisation in the development, and challenges related to coherent goals of DT in a multi-disciplinary team engaged in the design, development and implementation of DT to a larger scale system. To address these challenges, this study introduces a unified framework for DT development, emphasising reusability and scalability. The framework harmonises existing DT frameworks by unifying concepts and process development. It facilitates the integration of heterogeneous data types and ensures a continuous flow of information among data sources, simulation models and visualisation platforms. Scalability is achieved through ontology implementation, while employing an agent-based approach, it monitors physical asset performance, automatically detects faults, checks repair status and offers operators feedback on asset demand, availability and health conditions. The effectiveness of the proposed DT framework is validated through its application to a real-world case study involving five interconnected air compressors located at the Connected Facility at Devonport Royal Dockyard, UK. The DT automatically and remotely monitors the performance and health status of compressors, providing guidance to humans on fault repair. This guidance dynamically adapts based on feedback from the DT. Analyses of the results demonstrate that the proposed DT increases the facility’s operation availability and enhances decision-making by promptly and accurately detecting faults.