Iot and Big Data Analytics Platform to Analyze the Faults in the Automated Manufacturing Process Unit

As more data is gathered at every production stage, monitoring systems take part in a managerially more significant pastime. The Internet of Things (IoT) and other current sensor technologies may provide the solution for comprehensively monitoring industrial processes. IoT sensor data, in-depth data analysis, and a hybrid prediction model all come together in this study to provide a means of tracking. To get started, we built a sensor for the Internet of Things that can measure and record acceleration, rotational velocity, temperature, and humidity. The sensor data generated by the IoT in manufacturing is real-time, massive, and unstructured. The proposed tracking uses Apache Kafka for contact management. MongoDB holds sensor data acquired throughout the process, whereas Apache Storm is a real-time processing engine. Second, Density-Based Spatial Clustering of Applications with Noise (DBSCAN)-based outlier identification was used to exclude anomalous sensor data from the new model. Random Forest categorization was used to identify production issues. In a Korean auto factory, the proposed vehicle is scanned and evaluated. Results validate the usefulness of the proposed large-scale data processing system and Internet of Things-based sensors for monitoring the manufacturing process. In addition, the model’s failure prediction accuracy is improved when sensor data is input. The proposed strategy would aid management in making more informed choices and avoiding financial losses caused by manufacturing flaws.