Digital Twin Framework for Real-Time Energy Optimization and Carbon Footprint Reduction in Manufacturing Supply Chains

Azadeh Sohrabinejd¹ Mehdi Rahimi²

1) University professor, PhD graduate in Industrial Engineering
2) University professor, DBA graduate

IOI Code : AADG-AAHDBA
Publication : The second international conference on new approaches in engineering and basic sciences(icnabs.ir/2nd)
Abstract :
    Manufacturing industries face increasing pressure to reduce carbon emissions while maintaining productivity. This research presents a comprehensive digital twin (DT) framework for real-time simulation and optimization of energy consumption and carbon emissions in manufacturing supply chains. The framework integrates Industrial Internet of Things (IIoT), machine learning (ML), and multi-objective optimization algorithms (NSGA-II). A six-month case study in automotive parts manufacturing, using data from 150 sensors (48 million energy records), demonstrated 25% reduction in energy consumption, 30% reduction in carbon emissions, and 4.8% improvement in production efficiency. The ensemble ML model achieved 3.7% MAPE with statistical significance (t-test, p<0.001). Economic analysis showed ROI within 7 months and NPV of $2.9M over 5 years. Despite implementation challenges including initial investment ($500K) and technical complexity, the framework provides a practical roadmap for sustainable manufacturing.

Keywords : energy optimization sustainable supply chain machine learning carbon footprint nsga ii digital twin