The automotive industry encounters increasing supply chain challenges due to fluctuating demand, inventory imbalances, and long lead times. Traditional approaches face economic uncertainties, supplier disruptions, and trade policy changes, leading to inefficiencies and cost volatility. To address these issues, industry is adopting Industry 4.0 technologies, particularly Digital Twins, which create real-time virtual representations of physical systems, integrating sensor data and automation for enhanced decision-making, and predictive analytics. This chapter presents a Multi-Agent-Based Digital Twin (MABDT) model targeted toward automotive manufacturing supply chains. The proposed three-layer architecture, utilizing Physical, Digital Twin, and Application Layers, leverages Agent-Based Modeling simulation techniques to create real-time monitoring agents of supply chain elements using statecharts and structured agent behaviors. A case study on the Tiger Motors simulated assembly line demonstrates how this architecture is used to model cars, workstations, material supermarkets, and manufacturing cells as agents that interact dynamically with each other to develop applications that can improve takt time management, production visibility, and operational efficiency. Integrating real-time data with Digital Twin agents has the potential to enable predictive maintenance, reduce downtime, and ensure better supply-production synchronization. By aligning Digital Twins with multi-agent system architectures, this work establishes a scalable framework for intelligent manufacturing systems. Through this work we aim to contribute to adaptive supply chain management, bridging theory, and practice while serving as an educational testbed for digital manufacturing in automotive manufacturing.