Automobile collaborative supply chain accessory demand intelligent prediction method based on space-time diagram convolution

CN119962756AActive Publication Date: 2025-05-09CHONGQING UNIV
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202510146022.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-10
Publication Date
2025-05-09
Estimated Expiration
2045-02-10

AI Technical Summary

Technical Problem

The prior art is difficult to accurately predict the spatial and temporal relationship between nodes in the automotive parts supply chain, resulting in limited demand prediction accuracy and adaptability.

Method used

The intelligent prediction method of automotive collaborative supply chain accessories demand based on spatiotemporal graph convolution is adopted to generate multi-dimensional adjacency matrix through graph construction technology, combining spatiotemporal attention mechanism, LSTM network and GCN network to comprehensively model the dynamic correlation of spatiotemporal in the supply chain.

Benefits of technology

It significantly improves the accuracy and reliability of automotive collaborative supply chain accessories demand forecasts, and can better capture the dynamic temporal relationships and complex relationships in the supply chain.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119962756A_ABST
    Figure CN119962756A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of internet big data and new-generation information, in particular to an automobile collaborative supply chain accessory demand intelligent prediction method based on space-time diagram convolution, which comprises the following steps: S1, acquiring historical accessory demand time sequence data; s2, analyzing a supply chain structure based on historical accessory demand time sequence data; s3, mapping of a plurality of spatial dependencies is carried out based on the supply chain structure, and a plurality of adjacency matrixes are obtained through calculation; s4, inputting historical accessory demand time series data and various adjacent matrixes into the trained demand prediction model, and outputting accessory demand prediction data; multi-dimensional decoupling is performed on a supply chain network through a graph construction technology, an independent adjacency matrix is generated for each type of spatial dependency relationship, supply chain interaction characteristics of different semantic levels are reserved, and meanwhile, space-time dynamic association among nodes in an automobile part supply chain is effectively captured by introducing a space-time graph convolutional neural network, so that the spatial-temporal dynamic association of the nodes in the automobile part supply chain is effectively captured. And comprehensive modeling of the complexity of key nodes and multi-source data is realized.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Chain retail store sales prediction method based on space-time diagram convolutional network

    CN113837794A

  • Traffic flow prediction method based on combination of multiple attention and adaptive graph convolution

    CN116911460A

  • Artificial intelligence method for supply chain purchase

    CN118052506A

  • Traffic state prediction method based on adaptive dynamic space-time diagram convolutional network

    CN118116194A

  • Big data-based purchase and sale data analysis method and system, and storage medium

    CN118134533A