An ANP-Hopfield Neural Network Based Approach for Supply Chain Stress Testing

Supply chain resilience is increasingly critical in today's globalized and volatile business environment. This study proposes a novel approach to supply chain stress testing by combining Analytic Network Process (ANP) with Hopfield Neural Networks. The method constructs a stress testing index system based on product review, elasticity, agility, and cultural motivation. ANP is used to weight each index, while a discrete Hopfield neural network is employed to design equilibrium points corresponding to different stress levels. The model is applied to an automobile manufacturing case study, demonstrating its effectiveness in classifying supply chain stress levels. Results show that the proposed method can effectively identify key factors affecting supply chain stress and provide a comprehensive evaluation of supply chain resilience. This approach offers a new tool for supply chain managers to assess and enhance their networks' ability to withstand external pressures.