Research on Material Demand Forecasting Algorithm Based on Multi-Dimensional Feature Fusion

Shi-Yao She, Fang-Fang Yuan, Jun-Ke Li and Hong-Wei Dai
Additional contact information
Shi-Yao She: Purification Equipment Research Institute, China Shipbuilding Industry Corporation, China
Fang-Fang Yuan: Purification Equipment Research Institute, China Shipbuilding Industry Corporation, China
Jun-Ke Li: Purification Equipment Research Institute, China Shipbuilding Industry Corporation, China
Hong-Wei Dai: Dongfang Electronics Co., Ltd., Harbin, China

International Journal of Information System Modeling and Design (IJISMD), 2023, vol. 14, issue 1, 1-13

Abstract: Material demand forecasting has a profound impact on the supply chain and is an important prerequisite for manufacturing enterprises to produce. In order to accurately predict the material demand of enterprises, this paper proposes a material demand forecasting algorithm based on multi-dimensional feature fusion (DFMF). Secondly, in order to obtain the spatial features, the vector representation of the relevant materials of a material is obtained through the attention mechanism. Then, the authors aggregate the relevant material representation and material vector representation of materials to obtain the final material vector representation through aggregation function. Then the final material vector representation under different time scales is used as input, and the prediction value of material demand is obtained by using BP neural network. Finally, experiments show that the model can effectively obtain multi-dimensional features of materials for prediction, and the prediction results have high accuracy.

Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/IJISMD.330137 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:igg:jismd0:v:14:y:2023:i:1:p:1-13

Access Statistics for this article

International Journal of Information System Modeling and Design (IJISMD) is currently edited by Thierry O. C. Edoh

More articles in International Journal of Information System Modeling and Design (IJISMD) from IGI Global
Bibliographic data for series maintained by Journal Editor ().