Live Memory Forensics Integrated Reinforcement Learning Model for Optimized Demand Forecasting in Autonomous Supply Chain Management System

Today's highly interconnected global world economy operates in a highly agile manner. Smooth and timely coordination amongst the suppliers and manufacturers ensures a seamless supply chain which serves as the backbone for modern day businesses. Unforeseen demands and slightest fluctuations in demand/supply may lead to a highly unpredictable environment thereby leading to optimized demand forecasting causing a considerable loss for stakeholders throughout the supply chain. Reinforcement learning can help in dynamically adjusting the demand and supply across global value chains leading to optimized delivery efficiency. Moreover, cyberattacks in industrial framework may also lead to considerable losses. In this work, we propose a reinforcement learning based model for optimizing demand forecasting in autonomous supply chain management systems. The proposed model integrates live memory forensics module to enhance the security of autonomous supply chain management systems and to provide an ability to track the alterations made by various actors across the autonomous supply chain management system.