Change Control Design in Product Supply Chain System Based on Radial Basis Function Neural Network

Product supply chain systems are structurally complex infophysical systems that contain numerous unmodeled dynamics and uncertainties. Drastic fluctuations in user demand and sudden unexpected events—such as epidemics, trade wars, or cyber-attacks—can lead to changes in system structure or parameters or even destabilize the system. Designing changes within the product supply chain is an important strategy to meet user demand and maintain stable system operation. In this paper, we explore the use of artificial intelligence (AI) to enhance the analysis and control of complex product supply chain systems. We design a radial basis function neural network (RBFNN) to address Denial of Service (DoS) attacks. This RBFNN is designed to predict trends in inventory changes following a system attack and to develop optimal control strategies accordingly. First, we construct a mathematical model of the product supply chain system. Second, we leverage the predictive capability of the RBFNN to handle the effects of system changes and cyber-attacks through feed-forward compensatory control. A Linear Quadratic Regulator (LQR) is then designed under the nominal system. Finally, we verify the effectiveness of the proposed method through simulation experiments.