Contact: dkelesis@iit.demokritos.gr
Supply chain network design involves optimizing the structure and operations of a supply chain network to minimize costs and improve efficiency. Traditional methods include mixed-integer linear programming (MILP) and heuristic algorithms, which can be computationally intensive and may not scale well for large networks. Recent advancements include the use of graph-based models and machine learning techniques to improve scalability and efficiency.
This thesis will investigate the application of graph convolutional networks (GCNs) to supply chain network design. Key research questions include: How can GCNs be leveraged to model complex supply chain networks? What are the advantages of using GCNs over traditional methods in terms of scalability and solution quality? Can GCNs provide real-time optimization capabilities for dynamic supply chain environments?
Relevant literature:
[1] https://arxiv.org/pdf/2401.15299,
[3] https://www.tandfonline.com/doi/full/10.1080/00207543.2024.2344661