Enhancing Efficiency in Manufacturing through Automated AI Agent-Based Bearing Defect Diagnostic System

Industrial bearings are critical components in aerospace, industrial, and automotive manufacturing, where their failures can result in costly downtime. Traditional fault diagnosis typically depends on time-consuming on-site inspections conducted by specialized field engineers. This study introduces an automated Artificial Intelligence virtual agent system that functions as a maintenance technician, empowering on-site personnel to perform preliminary diagnoses. By reducing the dependence on specialized engineers, this technology aims to minimize downtime. The Agentic Artificial Intelligence system leverages agents with the backbone of intelligence from Computer Vision and Large Language Models to guide the inspection process, answer queries from a comprehensive knowledge base, analyze defect images, and generate detailed reports with actionable recommendations. Multiple deep learning algorithms are provisioned as backend API tools to support the agentic workflow. This study details the architectural design of the agentic system and provides a real-time simulation of its workflow. In this study, inspection reports previously conducted by live technicians are used as a surrogate for simulating the diagnostic process carried out by agents. Validation of the system is studied by industry standard metrics like RAGAS comparing reports generated by field technicians versus AI agent generated reports. This feasibility study gets a score of 0.72, and it shows good promise for automating the time-consuming defect identification process. The concepts discussed can be extended to other similar problems, demonstrating their potential to enhance operational efficiency across sectors. This AI agentic workflow automation is constantly evolving, and further studies are needed to improve current performance levels and to mitigate the risk factors for productionizing this solution.