AI-Powered Smart Supply Chains and Logistics Optimization
Authors: Manisha Sunil Bhende, Babasaheb Ramdas Jadhav, Suriya Murugan, Thella Preethi Priyanka
Journal: International Conference on Electronics and Sustainable Communication Systems (ICESC)
Publication year: 2025
DOI: 10.1109/icesc65114.2025.11212574
Abstract
Global supply chains are growing more sophisticated, revealing inherent inefficiencies in forecasting, routing, and inventory control in traditional logistics systems. While past research has exhibited the promise of artificial intelligence (AI) through machine learning, deep learning, reinforcement learning, and natural language processing, most applications are still siloed and do not deliver an integrated platform for end-to-end optimization. This study fills this vacuum through the creation and testing of an AI-based logistics system integrating fundamental AI methods with facilitating technologies like IoT and blockchain. A mixed-methodology was used, involving literature review, comparative performance testing against conventional systems, and high-demand variability scenario testing. Seven key performance indicators, forecast accuracy, delivery time, stockout rate, inventory turnover, procurement cycle, warehouse space utilization, and carbon footprint, were rigorously evaluated. Findings showed impressive gains: forecast accuracy rose by $26 \%$, stockouts dropped by $58 \%$, carbon emissions fell by $22 \%$, procurement cycles were reduced by $35.7 \%$, and warehouse utilization rose by $30.8 \%$. In spite of some data quality, legacy integration issues, and ethical issues, the results show that AI plays a revolutionary part in making supply chain networks efficient, sustainable, and robust. The research offers a complete conceptual framework and empirical analysis to inform broader AI adoption in logistics.