The 4th industrial revolution is pushing a strong digitalization of the process industry. In fact, it focuses the attention of the manufacturing multinationals on intelligent production systems based on connected factories through integrate cyber-physical systems, cloud technologies and the internet of things. As a consequence, enormous volumes of data of different types are available every few seconds, the so called: “big data”. This poses several challenges for the process industry in data analytics and data translation. Therefore, integrating datasets, fusing them, extracting useful information in real time and suggesting actions is of paramount importance in the process. However, not only the case of “big data”, but also the case of “small data” is a challenge for the process industry.
This is the case in which a novel product and a new process are developed, where optimal process parameters should be found to obtain a product of desired quality.
In this presentation successful industrial applications of data analytics and statistical design of experiments will be shown for different tasks: data mining and exploration, data historians retrieval, process understanding and troubleshooting, online process monitoring, real-time quality monitoring and prediction, predictive maintenance, artificial vision systems, data fusion, product and process development, product/process/technology transfer, process optimization and scale-up. The considered case studies will cover a wide series of industrially relevant fields: pharmaceutical industry, food industry, semiconductor fabrication, specialty chemical manufacturing, petrochemical industry.