Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Reference Architecture framework for enhanced social media data analytics for Predictive Maintenance models

: Grambau, J.; Hitzges, A.; Otto, B.


Institute of Electrical and Electronics Engineers -IEEE-:
IEEE International Conference on Engineering, Technology and Innovation, ICE/ITMC 2019. Proceedings : Co-creating our future: Scaling-up innovation capacities through the desgin and engineering of immersive, collaborative, empathic and cognitive systems, 17-19 June 2019, Valbonne Sophia-Antipolis, France
Piscataway, NJ: IEEE, 2019
ISBN: 978-1-7281-3401-7
ISBN: 978-1-7281-3402-4
ISBN: 978-1-7281-3400-0
International Conference on Engineering, Technology and Innovation (ICE) <25, 2019, Valbonne Sophia-Antipolis>
Fraunhofer ISST ()

Social Media data contains a lot of hidden information which is currently rarely used in the manner of service topics on product level. However, a deep analysis of existing predictive maintenance models shows, that the combined use of social media data with already existing data from products or internal service data can improve existing and new analytical models for an enhancing predictive maintenance. Therefore, this framework paper describes an approach how to gather, process and analyze Social Media data related to products of a power tool producer. The defined processes are executed with the Azure Machine Learning Studio and are visualized with Power Bi. The main result of this paper is the Reference Architecture which combines several processes combine heterogenous data sources and enable the "First time to Incident" algorithm which helps companies to to increase the precision of Predictive Maintenance.