IMPACT OF SUPPLY CHAIN AGILITY AND SUSTAINABILITY THROUGH MACHINE LEARNING

In consonance with this background, this study aims to analyze how supply chain agility and sustainability have been impacted through the use of machine learning. Today, the nature of supply chains has become far more complicated and rapidly changing, and being able to address changes in the market while being sustainable is what matters most. Challenges such as demand forecasting, optimal pricing and order quantity, life cycle assessment, product recommender systems, supplier selection, risk assessment, and churn prediction in supply chain management are handled by the study using several machine learning algorithms such as Lasso Regression, Elastic Net Regression, Gradient Boosted Trees, Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks. These algorithms were used to manage demand and supply chain, especially in the areas of demand forecasting, inventory control, and logistics. The research shows that machine learning brings a positive degree of improvement to the responsiveness of supply chains and helps organizations reduce expenses associated with inventories and warehouse space as well as identify optimal routes for the transportation of goods. It also takes care of the sustainability of the business by cutting on wastage and carbon footprint while enhancing the operations of the business. Moreover, the demonstrated research addresses the directions for the further application of machine learning in supply chain management and provides practical guidelines for multiple industries. The application of big data and predictive modeling helps to create an efficient and flexible supply chain and, therefore, gain a competitive advantage in the existing conditions on the market. The paper adds to existing research on the use of machine learning for SC design and operation and underlines that this area is crucial for attaining both tactical and strategic organizational aims related to sustainability. There is scope for future studies looking at how these types of algorithms can be interfaced with real-time feeds and how they may be applied across a range of industries to improve the prospects for integrating machine learning into supply chain planning.