IIoT based framework for data communication and prediction using augmented reality for legacy machine artifacts

Abstract

Machine data automation, process feedback, and visualization are critical elements in monitoring and improving system performance. Legacy machines constitute a large part of the factory floor assets, especially in small and medium enterprises (SMEs). Digitalization enables them to seamlessly connect within the smart factory framework to exchange product and process information. This paper extends our work on extracting digitized data from legacy machines using computer vision in the Industrial Internet of things (IIoT) environment. The proposed framework includes data communication through the cloud for process automation, performing data analytics, interpretation, and visualization using Augmented Reality (AR) based smart devices through human-in-the-loop feedback. Real-time process recommendations are provided on an AR app by implementing a Bayesian Network to enable the user to input desired quality metrics. A case study of capturing injection molding data represented through different process parameters is presented to evaluate the performance. Further, its extension to using other machine learning (ML) models for process feedback, the latency of real-time data presented, communication with different IoT protocols, and authentication are discussed. This AR-based Digital Thread framework has promising applications for visualizing machine process parameters and performance, real-time feedback, troubleshooting, and maintenance activities for legacy and smart machines.