IIoT System for Intelligent Detection of Bottleneck in Manufacturing Lines

Abstract

Production lines face numerous challenges to meet market demands, including constant changes in products that require continuous adjustments. Efficient and rapid reconfiguration and adaptation of production processes are crucial. In cases of inadequate adaptation, bottlenecks can arise due to human errors or incorrect configurations, often introducing complexity in pinpointing the root cause and resulting in financial losses. Furthermore, improper machine maintenance contributes to this situation as well. This article seeks to establish a framework grounded in the contemporary smart factory, the IIoT, the Industry 4.0 paradigm, and Big Data. The proposed system places emphasis on leveraging real-time data analysis for predicting risks, while concurrently conducting a thorough analysis of historical data to monitor trends and enhance bottleneck identification. The defined architecture operates across multiple levels, acquiring real-time information and generating historical data for training and continuous optimization. Predictive results contribute to decision-making and assist in mitigating bottlenecks in manufacturing lines.