Big Data Analytics for Predictive Manufacturing Control - A Case Study from Process Industry

Julian Krumeich; Sven Jacobi; Dirk Werth; Peter Loos(2014)
In: 3rd IEEE International Congress on BigData (BigData 2014), Anchorage, Alaska, June 2014, IEEE Xplore® Digital Library, S. 530-537, ISBN: 978-1-4799-5057-7, DOI: 10.1109/BigData.Congress.2014.83

Link zur Publikation:
http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6906825&newsearch=true&searchWithin=%22First%20Name%22:dirk&searchWithin=%22Last%20Name%22:werth

Abstract:

To keep up with increasing market demands in global competition, companies are forced to dynamically adapt each of their business process executions to individual business situations. Companies that are able to analyze the current states of their business processes, forecast their most optimal progress and proactively control them based on the derived knowledge, are an essential step ahead competitors. The paper at hand exploits the potentials through the usage of predictive analytics on big data aiming at event-based forecasts and proactive control of business processes. In doing so, the paper outlines—based on a case study of a large steel producing company—which production-related data can be collected by applied sensor technology at present; hence, forming a potential foundation for accurate forecasts. However, without dedicated methods of big data analytics, the company cannot utilize the potential of already available data for a proactive process control. Hence, the article forms a working and discussion basis for further research in big data analytics.