Digital Twin-driven carbon emissions management in manufacturing

Abstract

With global efforts intensifying to reduce carbon emissions, the management of carbon emissions in manufacturing, particularly for energy-intensive processes e.g., aluminum casting, has become crucial. The European Union’s Carbon Border Adjustment Mechanism (CBAM) requires accurate reporting of embedded emissions, adding pressure on manufacturers to adopt dynamic, real-time monitoring and reduction systems. Traditional carbon emission management (CEM) systems, based on static data and analytical models, fail to capture the complexities of fluctuating production conditions. Digital Twin (DT) technology, a key pillar of Industry 4.0, offers a promising solution by providing real-time, dynamic representations of manufacturing processes. Despite its potential, research gaps remain, particularly limited artificial intelligence (AI) integration in CEM, and the scarcity of empirically validated DT models at the factory level. To address these gaps, this paper introduces a DT-based framework for life cycle assessment (LCA) in CEM, enabling real-time monitoring, simulation, and AI-enabled prediction and optimization of carbon emissions across the production process. The framework integrates AnyLogic for simulation and Prosys OPC UA for real-time data exchange via Python, ensuring bi-directional data flow between the physical and virtual environments. A case study in the aluminum casting production unit − smelting demonstrates the ability to track energy consumption and carbon emissions through real-time monitoring and simulation, offering insights into sustainable practices aligned with CBAM and broader environmental goals. However, AI-enabled prediction and optimization will be explored in future work. Future research will also extend the framework across the entire aluminum casting production line and validate it in real-world industrial applications.

Topics & Concepts