Federated learning empowered Time-Sensitive Dynamic Data Modeling for applied computational problem of supply chain sector

• Melbin George,
• S. Baskar,

Vol. 19, No. 1, January 31, 2025
10.3837/tiis.2025.01.008, Download Paper (Free):

• optical network
• Supply Chain Management (SCM)
• Dynamic Data Modeling
• data protection

Abstract

Federated learning (FL) improves the product movement from raw material transformation to consumer goods. The optical network controls the flow of goods and services to achieve low-cost operation for data modelling with less energy usage. Data handling and protection are significant drawbacks in optical communication and applied computational problems in the supply chain sector. This article proposes a Time-Sensitive Dynamic Data Modeling (TSD2M) method for addressing the network-to-network synchronization error. The synchronization error is identified through network switching and goods delivery. The squirrel search algorithm includes A seasonal monitoring condition to avoid becoming stuck on a subset of optimum solutions. The network utilization and data models are revamped based on the best-afford demands. Global best solution update retains the best-performing solution with less time implementation through the learning process. Therefore, the proposed method maximizes delivery precision, data acquisition, and demand identification.

Statistics

Cite this article