AI-Driven Automated Quality Inspection for Beverage Bottles: Leveraging Object Detection Models for Enhanced Supply Chain Efficiency

In the beverage industry, maintaining product quality during packaging and throughout the supply chain is critical to ensuring customer satisfaction and brand integrity. This research addresses the challenge of automating quality inspection for beverage bottles by leveraging cutting-edge AI-based object detection models. The study focuses on identifying and classifying six key quality defects particularly Cracked_Bottle, Misaligned_Label, Missing_Cap, Normal_ Bottle, Overfilled_Bottle, and Underfilled_Bottle. These defects, if undetected, can lead to customer dissatisfaction, increased return rates, and potential brand damage. To tackle this problem, we implemented and evaluated three advanced object detection architectures— YOLOv8, YOLOv9, and YOLOv11—on a custom dataset comprising thousands of images of beverage bottles captured under diverse conditions, including varying lighting, angles, and backgrounds. Among the models, YOLOv8 emerged as the most effective, achieving an impressive 78% accuracy across all defect classes. The model demonstrated exceptional performance in detecting subtle defects such as misaligned labels and minor cracks, which are often overlooked in manual inspections. The integration of AI-driven quality control systems into the beverage supply chain not only minimizes human error but also significantly enhances operational efficiency. By automating the detection of defects, this approach ensures that only products meeting stringent quality standards reach consumers. Furthermore, the system provides real-time feedback, enabling swift corrective actions and reducing waste. This research underscores the transformative potential of AI in revolutionizing quality assurance processes within the beverage industry, ultimately driving customer trust, reducing costs, and improving overall supply chain performance.