Title: AI, Innovation and Global Transformation: Interdisciplinary Perspectives on Technology, Business and Society
Editors: Dr. J. Preetha, and Dr. Siddhartha Mehrotra
ISBN: 978-81-69857-64-2
Chapter: 24
DOI: https://doi.org/10.59646/809/24
Author: Dr. G. Pandi Selvi
Abstract
Modern supply chain networks operate under volatile market dynamics, non-linear demand shifts, and systemic geopolitical disruptions. Traditional linear supply chain models rely on retrospective data, causing information distortions such as the bullwhip effect. This chapter presents an end-to-end framework for Intelligent Supply Chain Management (ISCM) and predictive operations by combining cyber-physical visibility, edge Internet of Things (IoT) telemetry, and advanced machine learning algorithms. Utilizing Temporal Fusion Transformers (TFT) for probabilistic multi-horizon demand forecasting, Graph Neural Networks (GNN) for structural supply topology risk modeling, and Reinforcement Learning (RL) for autonomous inventory rebalancing, the architecture bridges tactical planning and physical execution. The framework was evaluated across a multi-echelon global consumer goods supply chain comprising 18 manufacturing facilities, 42 regional distribution centers, and over 120,000 retail endpoints. The empirical results demonstrate a 34.2% reduction in mean absolute scaled forecast error, a 28.6% drop in safety stock carrying costs, and a 42.1% acceleration in disruptive recovery velocity.
Keywords: Intelligent Supply Chain, Predictive Operations, Demand Forecasting, Temporal Fusion Transformers, Reinforcement Learning, Multi-Echelon Inventory Optimization.