How Artificial Intelligence Is Radically Changing the Mining Industry
Business Solutions Jason Fletcher, vice president of customer solutions at Komatsu, shares how artificial intelligence is being utilized in the new age of the mining industry.
How is artificial intelligence changing and improving the mining industry?
There are many opportunities for artificial intelligence in the mining industry. The benefits for safety are huge because AI can detect and even predict when failures and unsafe conditions occur. Getting people and equipment out of harm’s way — AI will be a huge player in that.
AI and machine learning will create opportunities to better manage the entire value stream by linking planning with operations, processing and logistics to drive business outcomes. Mining companies could theoretically use AI to scale production levels across multiple assets and commodity types as market conditions vary to leverage their portfolio for optimum ROI and profits.
These tools can also be used to hard code lessons learned over years into codified algorithms. In the future, this will happen at a pace that is difficult for humans to achieve because of the amount of data being processed. Especially in the areas of planning systems and predictive analytics, AI is allowing us to digitally learn from the past so we can code into our systems the things observed. Then the system can alert everyone with hard-coding predictability and knowledge transfer on an enterprise-wide scale.
What role does real-time monitoring and data analysis play when working on large projects?
At the machine level, digital twins of each machine will allow us to predict failures, track real-time performance, replay events to determine root cause and simulate future performance.
At the operational level, mines can then run scenario analysis to support operational planning, increase predictability and provide quicker fixes. This approach shifts resources and costs to the planning phase so that you end up with much lower costs overall because you’re decreasing unproductive downtime and events which are reactive and costly.
What is a main challenge to the widespread adoption of AI, and what advice would you give to miners and businesses looking to utilize the technology?
You need data. Getting the right data from the right sources and doing something useful with it is harder than it sounds. You need a good picture of your intended business outcomes first. Some mining companies simply don’t have their organization set up yet to leverage these newer tools. We see the full spectrum — from operations with huge departments of people dedicated to leveraging technology to smaller operations that don’t yet have the ability to dedicate those resources. Also, mining is often done in remote locations where you typically have connectivity issues and limited bandwidth. That barrier is starting to be overcome as technology gets cheaper; the ability to get data off the site is getting better.
There’s an essential shift that needs to happen, and is starting to happen, where the industry needs to be more open to partnerships, better leveraging out-of-the-box solutions and advancements from other companies and industries. In the past, mining businesses were used to building everything themselves because what they needed wasn’t available. But now we need to be open to participating in a larger ecosystem to make AI and machine learning effective.
Where do you see the industry in the next 10-15 years?
In the next 10-15 years, there will certainly be higher levels of automation and use of machine learning algorithms. That’s going to advance quickly, better enabling remote, autonomous operations, computer-led decision making and predictive capabilities. During the most recent industry downturn, taking cost out of the business was the primary goal. But automation and data analytics have been identified as ways to improve safety and lower costs, and now that we’re out of the downturn, business are trying to implement systems now to backstop for the next downcycle.