Publication

Abstract : Industry 4.0 (I4.0) aims on making a manufacturing plant to mitigate risk & retain its profitability by predicting its future and adapt for the future challenges through the intelligence given by the Data generated through Information Technology and Operational Technology (IT & OT) convergence. For this IT & OT Convergence to happen, a pre digital plant where machines are retrofitted with automation or a plant with having isolated machines with high level of automation will get connected. By becoming a connected plant, often mentioned as Industrial Internet Of Things (IIOT) huge amount of data gets generated and to achieve the standards of I4.0, data analytics have to be applied in the entire value chain of a manufacturing plant for better decision support system for the leaders / managers as well as better algorithms in the machines for optimum results like quality and productivity. This work reviews the different state of the art approaches of applying data analytics methodologies in the context of I 4.0. The work focuses on decision support algorithms for making business decisions like optimized Supply Chain Management & Asset condition monitoring. The data set used for the study is PHM08 challenge data set from prognostic data repository. Deployment of the Remaining Useful Life (RUL) prediction algorithm along with suitable data pre-processing in an EDGE device will be objective of the work. Model building & validation will be done using R Script and the final deployment will be done in python or Java Script based on the compatibility of the EDGE Device as suggested by Industrial Internet Consortium Analytical Framework. This work can be further scaled up for economical deployment in portable smart devices & smart sensors with limited computational capabilities to achieve level 5 as per Digital Maturity Assessment (DMA) guidelines of International Society of Automation (ISA).