Join Ali at Europe’s most forward thinking mining tech event – Mines and Technology London
Ali Soofastaei is a Data Scientist at Vale, who uses new models based on Artificial Intelligence (AI) methods to increase productivity, energy efficiency and reduce the total costs of mining operations.
Ahead of his presentation at Mines and Technology London, we caught up with him to discover his latest thoughts on AI and it’s usage in the mining industry.
What are the reasons for using AI and what are the most commonly adopted AI technologies implemented at mining operations today?
Artificial intelligence (AI) plays a primary, but often unrecognised, role in the modern world already. YouTube suggests what you should watch. You can ask Google Home how to repair your computer. AI is making our life more comfortable, and it is also providing new insights and opportunities for almost all industries.
It is becoming clear that AI can help mining companies improve the quality of their decisions, to increase safety and productivity and decrease their operating costs. Artificial intelligence solves problems that were impossible to address as recently as five years ago, enabled by the wealth of data now available and the rise of low-cost sensors, smart devices, cloud technology and a new generation of communication devices. Today, newly developed advanced data analysis tools, using AI algorithms enable mine managers to get new insights into how their mines are operating. Massive scale, low cost, secure data storage and increased computing power is making intelligence-driven decision making more affordable than ever. AI technologies are being used for different purposes in mining. Broadly these are to improve operations, maintenance, safety and energy efficiency. These aspects are, of course, highly coupled throughout the mining value chain and making a change in any one of them will influence all the others, very often in ways that are too complex for conventional analysis techniques to be useful. AI techniques operate successfully in these circumstances, showing patterns and behaviour previously unseen.
The application of AI is rapidly evolving today and slowly beginning to reshape how the mining sector operates. Fully automated, increasingly sophisticated mining machinery is becoming more widely used as ore bodies become more technically challenging to mine. The industry has to apply these advanced AI technologies to increase operational efficiency significantly.
What specific benefits does AI bring to mine operators?
By deploying the right AI technology, the mining industry will gain the ability to:
- gain new insights into an operational performance that can increase productivity and operational efficiencies;
- avoid mistakes and ‘human error’, by using smart systems
- achieve cost savings, by optimising business processes, workforce activities or products;
- save money and time by automating routine operations and tasks;
- make faster business decisions based on risk, cost and value-based analysis from cognitive technologies;
- predict customer preferences and offer them better, personalised experience and product;
- increase revenue by identifying and maximising sales opportunities
Artificial intelligence (AI) is steadily passing into everyday mining business use. From workflow management to trend predictions, AI has many different applications in current mining operations as well as an emerging use to support exploration and the development of new business opportunities.
Artificial intelligence is also the technology behind equipment and process automation. Automation has been and will continue to be, an innovation driver in mining. By using industrial vehicles or customised robots, the mining industry can reduce costs, improve safety and enhance productivity.
What are the business risks for mining companies when adopting AI technologies?
Artificial intelligence (AI) involves using machines to analyse data and make decisions in a more human-like manner. With this in mind, there are specific barriers and challenges to keep in mind.
One of the main barriers to the uptake of AI is cost. Creation of smart technologies can be expensive. Building a team with “double-deep” expertise – with the ability to develop AI solutions that recognise and accommodate the complex nature of mining operations – is time-consuming and difficult. Software programs need regular upgrading to adapt to the changing business environment and, in case of software failures, this presents a risk of losing critical code capability or even losing essential data. Restoring this system and technology-related failures is often time-consuming and costly.
Application of AI has limitations related to implementation times, which are often lengthy; integration challenges and lack of understanding of the state-of-the-art systems and usability and interoperability with other systems and platforms also present challenges.
With the fast development of AI, ethical issues have cropped up. These include the possibility of automation technology that creates job losses, a particular problem in countries where mining operations are seen as a means of providing much-needed work for local populations. There is also the on-going need to reorganise or retrain personnel to upgrade their skills. An open question is how to equitably distribute wealth created by machines – a general issue, not restricted to mining. The effects of machine interaction on human behaviour and attention is an issue and the need to eliminate any bias in AI generated by humans. There is also the difficult problem of mitigating against unintended consequences, a particular concern for AI, as smart algorithms (and machines) are taught to learn and develop independently.
Don’t forget, you can hear more from Ali at Mines and Money London (27-29 Nov). If you’ve not already reserved your place you can get your pass here.