AI-Driven Intelligent Assessment System for Supply Chain Risk Visualization Using Image Segmentation and Graph Neural Networks
Bibliographic record
Abstract
In the context of global supply chains facing multiple risk shocks such as natural disasters and geopolitical instability, traditional risk assessment methods reliant on manual analysis and static data face challenges such as information latency and insufficient visualization capabilities.These issues hinder their ability to address the uncertainty and transmission of risks.Existing research in supply chain risk assessment has significant limitations: U-Netbased segmentation algorithms lack adaptive mechanisms for scale adjustment, resulting in insufficient accuracy in extracting multi-scale features from complex supply chain risk visualization images; attention-based methods like CLIP cannot achieve deep semantic associations between images and language; and risk matrix methods fail to dynamically adapt to changes in supply chain network topology.To address these challenges, this paper focuses on an AI-driven intelligent evaluation system for supply chain risk visualization, proposing a three-layer technical architecture: "feature extraction-fusion reasoningevaluation output."At the bottom layer, an improved central difference convolution (CDC) operator is proposed to extract multi-scale features from images; the middle layer constructs a bi-directional image-language mapping network based on graph neural networks (GNNs) for cross-modal fusion; the top layer generates three-dimensional risk assessment outputs by integrating image segmentation results.The innovations of this study are: 1) the proposed improvement mechanism enhances the completeness and accuracy of complex image feature extraction; 2) the establishment of a deep image-language fusion model driven by GNNs addresses the issue of insufficient semantic association; and 3) the creation of dynamic and intuitive risk assessment outputs.This research provides a new technological path for supply chain risk visualization and assessment, improving both the accuracy and response efficiency of risk evaluations, while enriching the theoretical applications of crossmodal learning in industrial scenarios.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".