Mathematical Mining in 2019

Deloitte’s annual Tracking the Trends report articulates the issues that are transforming the future of mining each year. Its 2019 edition notes that the mining industry is poised for growth, but that market realities have altered significantly in recent years due to geopolitical tensions and looming asset shortages.

“Disruption and volatility has become the new normal and the pace of change is outpacing our ability to adapt,” says Philip Hopwood, Global Leader – Mining & Metals for Deloitte Touche Tohmatsu Limited. “This makes it imperative for mining companies to clarify how they plan to drive value into the future and how they intend to respond when prices inevitably drop again.”

Glenn Ives, Americas Mining & Metals Leader for Deloitte Canada, echoes this prediction, pointing out that “the fourth industrial revolution represents a new era of business that can only be harnessed by leaders who have the courage of their conviction”.

It’s clear that 2019 is not a time for shrinking violets when it comes to mining leadership.

The frontier of analytics and artificial intelligence

Unsurprisingly, technology figures strongly in the mining trends that Deloitte identifies for 2019. In this article, we take a look at Trend 2: The frontier of analytics and artificial intelligence (AI).

By now, mining companies know that the increasing connectivity between physical and digital worlds holds the promise of better planned, safer and more efficient mining operations.

The focus now has shifted to what Deloitte refers to as the “maturity curve” of analytics – how well are mining organisations using AI; what value is being delivered; and where should they be focusing their investment?

AI and analytics maturity

The report notes that, in other industries across the world, most organisations are working at “Horizon 1” of AI: assisted intelligence, in which human assistance and interpretation is needed.

Leading organisations, however, are rapidly moving towards “Horizon 2”: augmented intelligence, where machine learning (“training” algorithms with large volumes of data which evolve the algorithm without it needing to be explicitly programmed) is augmenting human decisions.

In order to progress to Horizon 2 (or Horizon 3, where AI decides and executes autonomously), organisations need to be able to answer increasingly complex questions, from “what happened?” to “why did it happen?” and “what will happen?”

This involves moving from descriptive analytics (which uses statistical methods to understand data patterns and trends) to predictive analytics (which uses machine learning algorithms to learn from these data patterns and forecast future trends) and finally to prescriptive analytics (which embeds models into an organisation’s analytics layer to enable decision support).

“Increasing analytics maturity requires greater integration of data from multiple sources, and delivery of end-to-end planning and decision-making solutions that span multiple processes and operations,” the report points out. “Although it is necessary to have a big picture to work towards, it is better to build the foundational data and infrastructure platforms in an evolutionary way while delivering specific use cases, rather than trying to build it all up-front.”

Biarri EMI: prescriptive analytics solutions

Two of the mining-specific use cases identified in the report involve maintenance planning and haul scheduling. These are addressed by two predictive mathematics products developed by Biarri EMI: DIMO and kavern.

Algorithmic web app kavern provides prescriptive analytics to guide load and haul equipment scheduling on mine sites. Working at the advanced end of the analytics maturity curve, kavern enables easy data capture, reporting and spatial mine visualisation, coupled with powerful automated scheduling capabilities.

As a result, kavern:

  • optimises throughput and increases production
  • increases visibility
  • reduces planning time
  • increases safety compliance.

Biarri EMI’s Distributed Infrastructure Maintenance Optimisation solution (DIMO) is used to schedule preventative and corrective maintenance of energy and resources assets that are distributed over large distances.

Like kavern, it introduces a spatial component to planning, and analyses data to automatically generate schedules. DIMO:

  • reduces time spent manually creating plans
  • optimises the utilisation of resources
  • drives productivity increases in maintenance and scheduling teams
  • reduces travel time and its related safety risks.

Crucially, both kavern and DIMO integrate with third-party systems and data acquisition sources such as Pitram, as well as data upload from spreadsheets. This means that, while they deliver targeted point solutions in specific areas, they are capable of interacting with other systems and extending their business value across multiple processes and operations. They provide a route for mining organisations to “think big, start small and scale fast”, as Deloitte’s research recommends. “The key is to get on with solving real problems and delivering value as quickly as possible, while keeping the big picture in mind”.

Get in touch with us to find out more about how Biarri EMI’s products are responding to the most pressing concerns in the resources sector today.

1 reply

Trackbacks & Pingbacks

  1. […] recent article, Mathematical Mining in 2019, examined the possibilities presented by analytics and artificial intelligence for today’s mining […]

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *