Leveraging Machine Learning for Optimal Advance Purchasing in Oracle Systems

The integration of Oracle-based systems using machine learning (ML) techniques for enhancing advance purchasing has attracted significant research interest over the past decade. The evolution of ML methods like time-series forecasting, reinforcement learning, and deep learning has enabled Oracle systems to make data-driven, smarter purchasing decisions, thus reducing occurrences of stockout, overstock, and procurement costs. However, there is a large research gap in effectively integrating external market situations, real-time data, and risk management methods into Oracle's procurement systems to support dynamic and robust purchasing decisions. Earlier studies focused on the implementation of certain ML algorithms like decision trees, support vector machines, and ensemble methods with minimal concern for their performance in dynamic, real-world environments where data streams and supply chain disruptions are common. Furthermore, integrating developing technologies like blockchain with machine learning to offer more transparency and security in the procurement process has also been quite underresearched. While machine learning has shown some promise to automate demand forecast and supplier management, there is little research on the challenges of integrating multiple streams of data (e.g., weather patterns, competitor behavior, and social movements) to support purchasing decisions in Oracle systems.