Data Science and Traditional Engineering and Technology Programs - How to Improve Operational Excellence

Abstract

This paper focuses on the application of Data Science to improving operational excellence (OE) for the process industries. By operational excellence, we mean the ability of a processing facility to operate safely, reliably and in an environmentally friendly manner while producing quality products, minimizing operating costs, and adapting to changing market conditions. Improving OE requires collaboration among professionals from different technical backgrounds. In addition, it involves the analysis of large amounts of data to help make intelligent business decisions. Recent advances in data science, big data analytics, and machine learning could help in that regard. For instance, such techniques can help improve process safety, predict equipment failure, provide better data visualization, manage alarms and abnormal operating conditions, develop inferential models, improve performance of automation strategies and real-time optimization, and help make decisions in planning long term operations. The paper provides an overview of various techniques that have been used in the process industries to improve OE and suggests curricular enhancements to traditional engineering and technology programs to educate engineers in data science, big data analytics, and machine learning.