Deep Reinforcement Learning for Dynamic Compliance Management in Blockchain Supply Chains: A CNN Perspective

This paper focuses on the implementation of deep reinforcement learning (DRL) coupled with convolutional neural network (CNN) in regulating dynamic compliance in blockchain-enabled supply-chain systems. These elements include the use of blockchain technology for data integrity of the data collected, DRL agent for decision-making, and a CNN for feature extraction of images or data captured. The work shows that the proposed system increases compliance accuracy, flexibility, and performance when compared to the existing approaches, and thus proves the system's effectiveness as a rule-based tool for dynamic compliance enhancement. In light of the above findings, this research brings value in raising the reliability and performance of supply chains while other regulations alter.