(75e) Keynote Talk - Digital Transformation, Data Science, and Industrial Intelligence

Data analysis has played an important role in process monitoring and operations for over three decades, with early methods focusing on multivariate statistics and artificial neural networks. Over the last ten years, along with the rise of big data, industrial internet of things (IIoT), statistical machine learning, and deep neural networks, digitalization and data analytics have taken place in nearly all industry sectors with an unprecedented pace. The McKinsey report in 2011 envisioned that analyzing large data sets would become a key basis of competition, underpinning new waves of productivity growth and innovations. Recent surveys by McKinsey revealed that the adoption of machine learning varied significantly from sector to sector.

In this keynote we offer a few perspectives on the development of data analytics and machine learning in process systems engineering and how we can maneuver their healthy development to harness the full potential of the digital revolution. Digitalization and the IIoT adoption will provide the process industries with massive amount of data from process operations to process maintenance. The power of data science and analytics from statistical learning methods to deep neural networks will need to be scrutinized and tailored for process applications. In driving toward industrial intelligence, we attempt to address the following big-picture questions in this talk.

i) What is the role of data science and analytics in transforming data into information and knowledge?

ii) Will human be out of the loop and replaced by artificial intelligence or autonomous systems?

iii) Will industry need deep learning or explainable machine learning?

iv) Will first principles be replaced with data-driven deep learning?