The Exploratory Programme project PROTEAS caters to growing demand in the digital twin market by demonstrating a functional proof of concept (POC) that empowers the training and deployment of machine learning (ML) models, thereby leveraging competencies in the global digital twin market. PROTEAS achieved its primary objective of establishing an edge artificial intelligence (AI) framework that enables the efficient utilisation of ML models on digital twins in a distributed fashion.
Successful completion of the project has resulted in the integration of cutting-edge cloud-native tools, enabling Edge ML operations (MLOps) and utilisation of graphics processing units (GPUs) on the Edge for hardware-accelerated training. The implementation of PROTEAS has yielded significant benefits for TWI and its Members engaged in digital twins and ML endeavours, including a reduction in the ML lifecycle time by over 20% and an increase in the accuracy of data-driven digital twins by more than 10%. PROTEAS has also enabled continuous monitoringof digital twin operations, verificationof accuracy and retraining if needed.
This Exploratory Programme project has enabled the project team to harness the enhanced capabilities of digital twin and ML to deliver a collaborative project for TWI Hellas, based in Athens, Greece, opening up a new domain in the space and data sector, which holds immense potential. This accomplishment illustrates the tangible benefits of the project's advancements in digital twin and ML, enabling TWI Hellas to venture into new and lucrative areas of opportunity.