: LEVERAGING TECHNOLOGY FOR REAL-TIME SUPPLY CHAIN OPTIMIZATION AND SUSTAINABLE INVENTORY PRACTICES

As the type of market and company competition gets more unpredictable, supply chain management takes a central place for companies who aim to maintain their competitive advantage. This paper aims to discuss the extent to which real-time supply chain operation can be enhanced by incorporating technologies and enhanced knowledge from machine learning techniques in a bid to achieve sustainable inventory management solutions. The paper looks into the viability of Lasso and Elastic Net regression for evasion over resection and categorization, where efficient data partitioning is done with K-means clustering and anomaly detection performed by DBSCAN. Further, a Convolution Neural Network(CNN) is used to decide suitable logistic paths to make them efficient and eco-friendly. By so doing, the research illustrates how supply chain dynamics can be enhanced through real-time technology; how decision-making can be enhanced, and how risks reduced while achieving sustainable goals. The contribution of the findings is to the modern literature on advanced SCM since the study presents raw information on the application of innovative technologies that enable supply chains to achieve increased efficiency and sustainability. Apart from contributing to the current knowledge of the subject, this research provides practical recommendations on how organizations can strengthen and enhance their sustainable supply chain management efforts.