Today, in Industry 4.0, Big data analytics – which is used to sort through massive amounts of data and identifies important patterns – have become useful in the advancement of industrial cognitive systems in process industries & are a major theme in current industrial technology development. The challenge is to achieve the advantage of using a data-driven cognitive system by integrating the heterogeneous data from multiple sources that can easily be used in a machine learning model and adjust the algorithms. The objective of DIGICOGS is to provide a digital twin that combines sensor information, AI and machine learning and big data analytics that underpin the new wave of the cognitive system. In DIGICOGS, cutting-edge solutions will be achieved through data-driven analytics, real-time monitoring and intelligent adaptive prediction based on combination of information i.e, sensor data, domain and context. DIGICOGS comprises MDH (a research group), Seco Tools (supplier to process industry and to GKN) & GKN (manufacturing industry). The project tasks will be performed as several work packages (WPs) including different methodologies, such as determining the state of the art, studying the data, defining the use-cases, developing the tools and evaluating the results. The predictive analytic tools developed in the project will assist operational operators, engineers and maintenance staff in fast & efficient process monitoring, predictive maintenance, improving productivity and energy efficiency. Thus, it is believed that the DIGICOGS technologies will strengthen Swedish industrial competence and competitiveness. The results from the project can be re-used and repeated with other Seco Tools customers and companies in process industry.
|First Name||Last Name||Title|
|Sharmin Sultana||Sheuly||Doctoral student|
Machine Learning Based Digital Twin in Manufacturing: A Bibliometric Analysis and Evolutionary Overview (Jul 2022) Sharmin Sultana Sheuly, Mobyen Uddin Ahmed, Shahina Begum Applied Sciences- Computing and Artificial Intelligence(Special issue:Artificial Intelligence and Optimization in Industry 4.0) (applsci)
Explainable Machine Learning to Improve Assembly Line Automation (Dec 2021) Sharmin Sultana Sheuly, Mobyen Uddin Ahmed, Shahina Begum, Michael Osbakk 4th International Conference on Artificial Intelligence for Industries (ai4i 2021)
Data Analytics using Statistical Methods and Machine Learning: A Case Study of Power Transfer Units (Mar 2021) Sharmin Sultana Sheuly, Shaibal Barua, Shahina Begum, Mobyen Uddin Ahmed, Ekrem Güclü , Michael Osbakk International Journal of Advanced Manufacturing Technology (IJAMT)