Supply Chain Demand Forecasting based on Gated Graph Neural Networks and Federated Learning

Key Points

• The central aim is to enhance demand forecasting accuracy in supply chains while maintaining data privacy.
• Utilized gated graph neural networks for modeling supply chain dynamics.
• Employed federated learning to preserve data privacy across multiple sources.
• Analyzed historical demand data to develop predictive models.
• Forecasting accuracy improved compared to traditional methods.
• Data privacy was effectively maintained using federated learning techniques.
• The approach showed robust performance across various supply chain scenarios.