Resilience Assessment of Global Manufacturing Value Chains Under the Influence of the Carbon Border Adjustment Mechanism Using Machine Learning and Knowledge Management

The introduction of the Carbon Border Adjustment Mechanism (CBAM) that balances the requirements of sustainability, competitiveness, and compliance in the international manufacturing value chain creates a new level of complexity. The present paper will be a proposal of a new Knowledge Resilience Assessment Framework (ML-KRAF) based on Machine Learning, which can be utilized to determine and improve the resilience of manufacturing global value chains under the effects of CBAM. The model integrates the approach of knowledge management theory and machine learning intelligence to gather, examine, and utilize the organization and cross-organization knowledge in the field of carbon management, supply chain flexibility, and production efficiency. It explains the Hybrid Random Forest-Graph Neural Network (HRF-GNN) as a new hybrid analysis model that is purportedly employed to simulate complex relationships between global value networks and offer ways of potential disruption and support strategic planning of resilience. Along with that, a Knowledge-Based Reinforcement Module (KBRM) refers to a continuous decision-maker optimizer by bringing real-time information and professional insights into dynamic knowledge bases. The model emphasizes active learning, carbon intelligence exchange, and dynamism of sustainability strategies in response to changes in regulations. The paper can be used to complete a new paradigm on how resilience is understood and managed in manufacturing systems within carbon-regulated global contexts by relating machine learning analytics and knowledge management systems, which provides a new theoretical and practical guidance to the leaders of policy, industry and sustainability strategies.