PyEye: A Real-Time Edge-Based Data Capture System for Error Reduction in Industrial Logistics

Abstract

In modern industrial logistics environments, manual data capture and delayed system updates frequently lead to operational errors, reduced traceability, and production risks. This paper presents PyEye, a real-time edge-based data capture system designed to improve accuracy, reliability, and responsiveness in manufacturing logistics operations. PyEye integrates barcode scanning, automated image capture, and local-first data processing to ensure immediate validation and traceability at the point of operation. The system employs hybrid architecture combining edge computation with cloud-based synchronization, enabling continuous operation under network constraints while maintaining data integrity. The proposed solution has been deployed in a live production environment at Jaguar Land Rover’s Battery Assembly Centre (BAC), where it supports high-frequency handling unit tracking and visual verification. Experimental results demonstrate a 98% reduction in data capture errors, elimination of production line stoppages due to incorrect material delivery, and an estimated annual cost saving of £40,000. Compared to traditional manual and semi-automated approaches, PyEye significantly reduces latency from hours to seconds and enhances operational visibility. These results highlight the effectiveness of edge-based automation systems in improving reliability and efficiency in industrial logistics.