AIChE: Big Data Acuity a Must

Tuesday, March 28, 2017 @ 04:03 PM gHale


By Gregory Hale
Throughout the industry there has been an explosion of data and its capabilities over the past 10 years.

“What is interesting is the capability existed 10 to 20 years before that,” said Lloyd Colegrove, data services director and fundamental problem solving director at the Dow Chemical Co. during his Monday presentation entitled “Big Data Analytics Skills – A Revolution Lacking in Revolutionaries” at the AIChE Spring Meeting and 13th Global Congress on Process Safety in San Antonio, TX. “The greatest challenge to us today is improving the data acuity of humans. Data acuity is to numbers as verbal literacy is to words.”

RELATED STORIES
AIChE: Big Data for Safety
AIChE: Chemical Engineers at Heart of Safety
AIChE: Safety Taken Personally
Chemical Leak Hurts 1, Forces Evac

There is no doubt there has been an avalanche of data cascading through manufacturing systems over the past decade or so, but with the onslaught of the Industrial Internet of Things (IIoT), it looks as though the number will increase at an even larger scale.

That can play into the hands of Big Data, but as Colegrove said, when you get those numbers everything has to be put into proper context.

“The goal is to make better decisions; make timely decisions,” he said.

With all that data coming in, Colegrove said recipients need to know if it is any good. “How do you know? How should one respond to the data they see?” Colegrove asked.

Data contextualization requires the user to understand a process variation. To point that out, he showed a chart that looked good in terms of product production at the outset, but after doing a “Big Data deep dive,” the results were quite the opposite. The manufacturer had a client, but the client would claim every year one product shipment was not up to specs. That would mean the manufacturer would lose all profits with the delivery of one bad batch of product that was out of spec. After conducting a Big Data analysis, they ended up proving they were making four different products instead of the one.

“We are touching data today, but we have to improve our data acuity across the spectrum.”

Since 2011, there has been a rise in predictive analytics masters and other programs, Colegrove said. “We need to improve data acuity for each student.” With all the potential for Big Data, “we don’t see the opportunity in front of us.”

How can the industry move in the right direction?

The greatest challenge right now is culture, he said. People are locked into doing things their own way, but the issue is Big Data is a very dynamic environment. Thinking differently can help move everyone in the right direction.



Leave a Reply

You must be logged in to post a comment.