Process Data Analytics in the Era of Big Data

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

328 Scopus Citations
View graph of relations



Original languageEnglish
Pages (from-to)3092-3100
Number of pages9
Journal / PublicationAICHE Journal
Issue number9
Online published2 Jul 2014
Publication statusPublished - Sept 2014
Externally publishedYes


For engineering systems where processes, units, and equipment are designed with clear objectives and are usually operated under well‐controlled circumstances as designed, mechanistic models and first principles are dependable. However, for emerging circumstances that are not factored into the design, data become indispensable assets for decision making in safe and efficient operations. In this Perspective article, we offer a brief introduction to the essence of big data, a description of how data have been effectively used in process operations and control, and new perspectives on how chemical process systems might evolve into a new paradigm of data‐enhanced operations and control. The discussed perspectives include (1) the mining of time‐series data with expanded depth in history and breadth in location for event discovery, decision making, and causality analysis; (2) the exploration of the power of new machine‐learning techniques that have enjoyed tremendous development in nearly 2 decades; and (3) the anticipation of a system architecture shift towards a data‐friendly information system to complement the current distributed control systems centric information system. In addition, high‐level systems engineering tasks such as planning and scheduling1 can also benefit from information extracted from big data since optimization and control have always relied on the interplay between models and data. We note that big data is not the answer to everything, but historical and real‐time data are valuable for safe and efficient operations, especially for abnormal process behaviors or circumstances that are not considered in the design phase.

Research Area(s)

  • big data, machine learning, optimization, process data analytics, process control, process improvement

Citation Format(s)

Process Data Analytics in the Era of Big Data. / Qin, S. Joe.
In: AICHE Journal, Vol. 60, No. 9, 09.2014, p. 3092-3100.

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review