DELVING INTO THICK DATA

Australia’s Mining Monthly – Big data is being touted as a way to save mining companies billions, however, like most things, context is all important.

 


While it brings a flood of information, finding the one piece that can unlock savings can be tough.
Dingo managing director Paul Higgins said the sort of context a human could bring to the data could make all the difference.

Take the Queensland mining company that was having problems with the planetary gearboxes on its bulldozer fleet.

The drives were suffering unexplained wear. It turns out the site had been affected by flooding and, having that knowledge, a Dingo staff member discovered the dozers were being driven through mud and that mud was finding its way into the planetary gearboxes.

This is what Higgins calls “thick” data. It is adding human understanding and domain expertise to bring fresh perspective and insight to the data.

That data does not have to come from sensors either.

One of the most effective tools in Higgins’ opinion is a magnetic plug that can be fitted to planetary gearboxes.
These plugs are simply unscrewed and inspected to see what ferrous metals have accumulated on them. These metal shavings can indicate problems within the gearbox or in other parts linked to it.

Dingo is working with Google on an app that makes use of the Google API tool to include an image library so maintenance staff can take photos of the plugs and send them to the cloud. An expert can compare those images to the image library to identify whether there are any signs of existing or impending issues.

Higgins said having an understanding of the mining sector was important because it could help with the interpretation of the data.

“We continue to see a lot of the big mining companies issuing these technology challenges,” he said.
“They are giving companies a big chunk of data and asking them what they can do with it.”
Higgins said the companies taking part in the challenges said they were often marked down for their lack of domain expertise.

“Some think the next round of mining advancement will come from this big data,” he said.
“We agree but if you don’t connect with the people using the data and understand how these insights will be employed you can miss the boat.

“You’ve got to understand what the people in the mining companies that are going to be consuming this information, how do they work and how do you prepare and put the insights in a way that make sense in their world.

“We’ve essentially got three customers on each mine site. We have the general manager who is big picture. They want the data in the simplest form.

“Then there are the maintenance managers dealing with large groups of people doing
the work. A lot of their work is people management and change management.

“And then you have the mechanics doing the work and the reliability engineers.
“If you don’t understand the needs of these different groups of people and just spit out a generic result it can backfire.

“Adding that layer of context is something we work hard at.”
To do that Dingo makes people a central part of its service delivery model.
“We take more of a laser beam approach,” Higgins said.

“Often the more valuable data is hard to collect. Are you getting the right data? What are you going to do with the data?”

It is about more than the sheer volume of data and processing power of technology.
It is about knowing how the data is going to be applied, having the right systems and processes in place, and keeping the people who use the technology central to the process.