(59o) Scaling Manufacturing Data Analytics, from Statistical Process Control to Machine Learning
AIChE Spring Meeting and Global Congress on Process Safety
2020
2020 Virtual Spring Meeting and 16th GCPS
Industry 4.0 Topical Conference
Big Data Analytics and Statistics
Friday, August 21, 2020 - 2:10pm to 2:30pm
The chemical manufacturing environment is filled with thousands of missed signals, where providing the right type of analyzed data to the right person, at the right time, can reduce out of spec product, minimize equipment downtime, and improve raw material utilization. After having success using Statistical Process Control, Albemarle wanted to expand its capabilities with their existing analytics partner, Northwest Analytics, into Machine Learning. A pilot across six production units found hidden signals and unknown correlations among thousands of data points, but scaling machine learning globally would be necessary to enable true digital transformation and drive ROI. Currently, Albemarle is rebuilding their data infrastructure, using OSIsoft PI, and creating a platform and strategy for their global enterprise. Albemarleâs vision for Digital Transformation includes leveraging Data Analytics and Machine Learning in their chemical manufacturing facilities to empower their employees to make data driven decisions.