Leveraging Machine Learning Tools to Identify and Manage Suppliers Corresponded to Defective Parts During Warranty

Managing defective parts during the warranty period presents significant challenges in the automotive industry, including accurately identifying responsible suppliers, reducing human errors, and handling large volumes of data. Moreover, ensuring product quality and providing excellent after-sales services are crucial for customer satisfaction and increasing customer retention rates. This paper addresses these challenges by introducing a machine learning (ML) approach to automate supplier verification and identify misclassified faulty parts. Specifically, we employ Gradient Boosting Machines (GBM) due to their superior performance in handling complex datasets and imbalanced classifications. We trained and evaluated our model using 401,168 data collected from multiple analysis centers nationwide. The model achieved a recall of 92.3%, a precision of 90.7%, an F1-score of 91.5%, and an AUC of 0.96, ensuring accurate identification of faulty parts' suppliers across our diverse environments. By integrating AI-driven methodologies, our system not only reduces manual workload and human error but also enhances supplier accountability and operational efficiency, which shows the promising perspective of using AI in large industries like the automotive industry. While traditional manual processes are time-consuming, error-prone, and inefficient, leading to delays and increased operational costs, this ML framework is a useful tool for addressing those issues and can enhance warranty management, product quality, and customer satisfaction in the automotive industry, which can lead to securing market share and competitive advantage.