Developing a Multi-task Ensemble Geometric Deep Network for Supply Chain Sustainability and Risk Management

• URL: http://arxiv.org/abs/2510.26203v1
• Date: Thu, 30 Oct 2025 07:26:18 GMT
• Title: Developing a Multi-task Ensemble Geometric Deep Network for Supply Chain Sustainability and Risk Management
• Authors: Mehdi Khaleghi, Nastaran Khaleghi, Sobhan Sheykhivand, Sebelan Danishvar,
• Abstract summary: The proposed Chebyshev ensemble geometric network (Ch-EGN) is a hybrid convolutional and geometric deep learning.<n>The product classification and edge classification are performed using the SupplyGraph database to enhance the sustainability of the supply chain.<n>The results confirm an average improvement and efficiency of the proposed method compared to the state-of-the-art approaches.
• Score: 0.22020053359163297
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The sustainability of supply chain plays a key role in achieving optimal performance in controlling the supply chain. The management of risks that occur in a supply chain is a fundamental problem for the purpose of developing the sustainability of the network and elevating the performance efficiency of the supply chain. The correct classification of products is another essential element in a sustainable supply chain. Acknowledging recent breakthroughs in the context of deep networks, several architectural options have been deployed to analyze supply chain datasets. A novel geometric deep network is used to propose an ensemble deep network. The proposed Chebyshev ensemble geometric network (Ch-EGN) is a hybrid convolutional and geometric deep learning. This network is proposed to leverage the information dependencies in supply chain to derive invisible states of samples in the database. The functionality of the proposed deep network is assessed on the two different databases. The SupplyGraph Dataset and DataCo are considered in this research. The prediction of delivery status of DataCo supply chain is done for risk administration. The product classification and edge classification are performed using the SupplyGraph database to enhance the sustainability of the supply network. An average accuracy of 98.95% is obtained for the ensemble network for risk management. The average accuracy of 100% and 98.07% are obtained for sustainable supply chain in terms of 5 product group classification and 4 product relation classification, respectively. The average accuracy of 92.37% is attained for 25 company relation classification. The results confirm an average improvement and efficiency of the proposed method compared to the state-of-the-art approaches.

Related papers

• An Analytics-Driven Approach to Enhancing Supply Chain Visibility with Graph Neural Networks and Federated Learning [52.79646338275159]
  We propose a novel approach that integrates Federated Learning (FL) and Graph Convolutional Neural Networks (GCNs) to enhance supply chain visibility.<n>FL enables collaborative model training across countries by facilitating information sharing without requiring raw data exchange.<n>GCNs empower the framework to capture intricate relational patterns within knowledge graphs, enabling accurate link prediction to uncover hidden connections.
  arXiv  Detail & Related papers  (2025-03-10T12:15:45Z)
• Enhancing Supply Chain Visibility with Generative AI: An Exploratory Case Study on Relationship Prediction in Knowledge Graphs [52.79646338275159]
  Relationship prediction aims to increase the visibility of supply chains using data-driven techniques.<n>Existing methods have been successful for predicting relationships but struggle to extract the context in which these relationships are embedded.<n>Lack of context prevents practitioners from distinguishing transactional relations from established supply chain relations.
  arXiv  Detail & Related papers  (2024-12-04T15:19:01Z)
• Enhancing Supply Chain Visibility with Knowledge Graphs and Large Language Models [49.898152180805454]
  This paper presents a novel framework leveraging Knowledge Graphs (KGs) and Large Language Models (LLMs) to enhance supply chain visibility. Our zero-shot, LLM-driven approach automates the extraction of supply chain information from diverse public sources. With high accuracy in NER and RE tasks, it provides an effective tool for understanding complex, multi-tiered supply networks.
  arXiv  Detail & Related papers  (2024-08-05T17:11:29Z)
• Generative Probabilistic Planning for Optimizing Supply Chain Networks [0.0]
  This paper introduces a novel Generative AI technique, which we call Generative Probabilistic Planning (IPP) IPP generates dynamic supply action plans that are globally optimized across all network nodes over the time horizon. Our experiments using historical data from a global consumer goods company demonstrate that IPP accomplishes objective-adaptable, probabilistically resilient, and dynamic planning for supply chain networks.
  arXiv  Detail & Related papers  (2024-04-11T07:06:58Z)
• HKTGNN: Hierarchical Knowledge Transferable Graph Neural Network-based Supply Chain Risk Assessment [3.439495194421287]
  We propose a hierarchical knowledge transferable graph neural network-based (HKTGNN) supply chain risk assessment model. We embed the supply chain network corresponding to individual goods in the supply chain using the graph embedding module. Our model outperforms in experiments on a real-world supply chain dataset.
  arXiv  Detail & Related papers  (2023-11-07T00:54:04Z)
• Identifying contributors to supply chain outcomes in a multi-echelon setting: a decentralised approach [47.00450933765504]
  We propose the use of explainable artificial intelligence for decentralised computing of estimated contributions to a metric of interest. This approach mitigates the need to convince supply chain actors to share data, as all computations occur in a decentralised manner. Results demonstrate the effectiveness of our approach in detecting the source of quality variations compared to a centralised approach.
  arXiv  Detail & Related papers  (2023-07-22T20:03:16Z)
• A Knowledge Graph Perspective on Supply Chain Resilience [15.028130016717773]
  Global crises and regulatory developments require increased supply chain transparency and resilience. Information about supply chains, especially at the deeper levels, is often intransparent and incomplete. By connecting different data sources, we model the supply network as a knowledge graph and achieve transparency up to tier-3 suppliers.
  arXiv  Detail & Related papers  (2023-05-15T10:14:30Z)
• Will bots take over the supply chain? Revisiting Agent-based supply chain automation [71.77396882936951]
  Agent-based supply chains have been proposed since early 2000; industrial uptake has been lagging. We find that agent-based technology has matured, and other supporting technologies that are penetrating supply chains are filling in gaps. For example, the ubiquity of IoT technology helps agents "sense" the state of affairs in a supply chain and opens up new possibilities for automation.
  arXiv  Detail & Related papers  (2021-09-03T18:44:26Z)
• Data Considerations in Graph Representation Learning for Supply Chain Networks [64.72135325074963]
  We present a graph representation learning approach to uncover hidden dependency links. We demonstrate that our representation facilitates state-of-the-art performance on link prediction of a global automotive supply chain network.
  arXiv  Detail & Related papers  (2021-07-22T12:28:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.