Skip to main content Skip to Footer


Analytics in mining: Q&A with Constantino Seixas

Accenture's Constantino Seixas outlines how to use analytics to increase efficiency in mining processes.
LinkedIn - Constantino Seixas Filho


Twitter - Accenture Natural Resources   Mail to Accenture Natural Resources. This opens a new window.

Are mining companies doing enough with their data?
The short answer is probably not. Many miners are building data management platforms that pull together data from programmable logic controllers, SCADA systems, specialized applications and sensors. They typically use the data to conduct high-level analytics and produce reports, but I believe they are missing a much larger opportunity to use analytics to gain deeper, more timely insight into operations.

By taking a more sophisticated approach—and taking advantage of advancing analytics tools and techniques—mining companies can extract significantly more value from the wealth of data they are collecting, and use it to drive greater efficiency and effectiveness in the business.

What kind of analytics can miners use and how can they benefit?
Analytics technologies are advancing rapidly. Mining companies can now use a range of analytics tools to measure more accurately several aspects of their businesses and be more proactive in managing operations. Examples include:

  • Video analytics—Use various types of cameras for real-time measurement of activities, such as product size distribution on conveyor belts, or froth size and bubble collapsing rate in the flotation process.

  • Asset analytics—Identify equipment problems before they happen, enabling predictive maintenance and rapid repairs. For instance, miners can install asset health sensors to measure equipment vibration and bearing temperature.

  • Process analytics—Find causes of quality problems and increase efficiency of both the process and the engineering talent managing the process. For example, process analytics can help miners better understand how the recovery in flotation correlates with floculant concentration and pH.

  • Production analytics—Determine what to do to improve production performance in mines, concentration plants and ports using a real-time process model to generate “what-if” analyses and answer questions such as “What is more influential to improve mine throughput—adding one more shovel or more trucks?”

What are some analytics best practices that mining companies can adopt?
My advice is to take a look at what other industries (banking, insurance, retail) are doing to proactively use information to support their businesses. Companies like Amazon know their customers’ behaviors and offer additional products based on these interests. For instance, the retailer probably knows how frequently I buy printer supplies, who my favorite book authors are and the style of puzzle I enjoy. Mining companies can take a similar approach and use analytics to understand the patterns of behavior in their processes.

As for leading practices, high-performance miners use a wide variety of data sources to make the business more profitable. They embed analytics in decision making, target business outcomes with analytics and establish a centralized analytics function with leadership at the C-suite level.

Rumor has it you enjoy traveling with your family. Are there any analytics capabilities that apply to family management on trips?
I make it a point to travel twice per year with my family. We have been to Disneyland, Las Vegas, Italy, Cape Town and so many other interesting places. Like many travelers, I use online tools to select hotels and read other travelers’ opinions about the destination. I even do a virtual “fly over” to assess a hotel’s swimming pool and surrounding neighborhoods before I make a booking. I think of this as using available data to make a more informed decision for my family.

You use words like “multivariate statistics” and “similarity-based modeling,” which make you sound like a big data buff. Do you have any non-data-related interests?
I’m into music, especially Brazilian Bossa Nova and Jazz. But since we’re talking about music, have you ever wondered if the Beatles or the Rolling Stones were more influential to music development? I recently read an article1 that shows how analytics and a tool called principal component analysis can provide a definitive answer. Check it out to see if you agree.

1The Economist, May 9, 2015