Optimization and Benefit Assessment of Production Supply Chain Networks Using Graph Neural Network Models

Abstract

With the flourishing development of global economy, effective management of production supply chains is crucial for the competitiveness of enterprises. Optimizing supply chain networks can not only improve the efficiency of resource allocation but also enhance market responsiveness and systemic risk resistance. Traditional supply chain network optimization methods, focusing mostly on linear models and static analysis, fall short in addressing the growing complexity and dynamism. The emergence of Graph Neural Network (GNN) models in recent years has offered new opportunities to tackle non-linearity and structural dynamism in supply chain networks. However, existing research still faces methodological limitations in supply chain node relationship mining and benefit assessment. This study introduces an optimization and benefit assessment method for production supply chain networks based on GNNs. Firstly, by developing a node role type-aware graph neural network model, this paper achieves in-depth mining and optimization of node relationships within production supply chain networks. Secondly, a hierarchical factor analysis method is used to comprehensively assess the benefits of the production supply chain. This method can dynamically capture changes in node roles and relationships within the supply chain network, optimize the network structure, and provides a multidimensional, multilevel framework for benefit assessment. This study not only expands the application of GNN in the field of supply chain management but also provides a new analytical tool for the comprehensive assessment of supply chain benefits.