The Knowledge Transfer Partnership set out to embed skill sets in digital twin development to maximise the operational efficiency of a core company product.
Key Facts
The project was a Knowledge Transfer Partnership (KTP), a grant-funded opportunity that drives business growth and innovation, between Cranfield University’s Centre for Digital and Design Engineering and advanced materials company Levidian. Innovate UK who funded the project graded the final report as ‘Outstanding’, the highest possible grade.
The aim of the KTP was to develop a digital twin of Levidian’s LOOP decarbonisation system to improve operational efficiency, scalability, and predictive maintenance. The rationale was to bridge academic expertise in digital engineering and machine learning with industrial needs for reliable hydrogen and graphene production. By creating a real-time virtual model of the LOOP reactor, the project sought to improve monitoring, forecasting, and maintenance planning, while embedding a digital engineering capability within Levidian’s team to support future innovation and commercialisation.
The KTP was successful in delivering a real-time, cloud-based digital twin of Levidian’s LOOP system, achieving accurate monitoring, predictive maintenance, and performance forecasting. This resulted in operational efficiency improvement, reduced downtime, and a scalable architecture for future systems.
The partnership provided Cranfield University with an opportunity to use an applied research platform, develop a new case-study and teaching material, and other opportunities for publications and teaching.
Impact of our research
Overall, the digital twin delivered beyond its core objectives, introducing advanced capabilities that added tangible value. Strategic enhancements beyond the original scope improved its flexibility, scalability, and long-term relevance. These highlighted a strong emphasis on innovation and alignment with broader business goals.
“The KTP directly supported our strategic goal to embed digital capability across our product range. By developing and validating a digital twin prototype, the project provided the logic and framework for data-led LOOP system design, commissioning, and support”. Quote from Levidian.
Why the research was commissioned
The primary objective was to develop a digital twin for the company to enhance the operational efficiency, reliability, and scalability of the LOOP decarbonisation system. LOOP is Levidian's proprietary, technology for converting methane-containing gas into hydrogen and graphene. The goal of the digital twin solution was to facilitate the LOOP system's transition from the pilot phase to full commercial deployment. More specifically, the project sought to develop a customised digital twin for the company's pilot-scale reactor. The insights and modelling tools produced would be crucial for the design and scaling of upcoming LOOP models, in addition to enhancing the performance and dependability in practical deployments.
Why Cranfield?
This partnership helped Cranfield University’s research in digital twin technologies, industrial AI, and data-driven sustainability. The collaboration provided access to real-world industrial datasets from Levidian’s LOOP system, enabling validation of predictive maintenance and machine learning models under authentic operating conditions. These findings will directly support new research outputs and journal submissions in green technologies. The project opened new interdisciplinary avenues between digital engineering and sustainable manufacturing, establishing a strong foundation for potential future collaboration.
Facilities used
The methodological focus was on predictive maintenance and graphene-output predictions. Both supervised and unsupervised models were analysed, including Azure AutoML. The final model was deployed as a REST API. Exploratory techniques such as causal analysis, anomaly detection, and feature importance were also employed, supported by tools like Azure Databricks.
