<p>The automotive parts supply chain is a multi-commodity, complex network influenced by macroeconomic fluctuations, social factors, costs, and uncertain risks. This study aims to optimize the automotive parts supply chain by maintaining network stability, enhancing efficiency, and improving supply chain effectiveness, providing competitive advantages for manufacturers. A key innovation of this study is integrating demand forecasting techniques with data-driven and scenario-based mathematical modeling. By utilizing neural networks optimized with meta-heuristic algorithms, demand forecast accuracy is significantly improved, reducing prediction time and aligning production and inventory with actual demand. The developed data-driven mathematical model balances economic, environmental, and social goals, employing advanced optimization techniques to handle uncertainties and disruptions, enhancing resilience and adaptability. It includes indices such as suppliers, proposed locations for parts manufacturers, distributors, dealerships, collection centers, and recycling centers, and technical parameters like demand, costs, capacities, disruption percentages, carbon dioxide emissions, and job creation. The model has three main objectives: maximizing economic profit by optimizing costs, minimizing environmental impact by reducing carbon dioxide emissions, and maximizing job opportunities and regional development. This research also examines data from car manufacturers and dealerships in Iran, reflecting unique market conditions and operational challenges.</p>