Mining conglomerate BHP describes AI as a way to turn operational data into better everyday decisions. The company’s blog post highlights that it analyzes data from sensors and monitoring systems to identify patterns in plant machinery, flagging problems, and providing decision makers with options that can improve efficiency and safety, as well as reduce environmental impact.
The useful question for BHP’s business leaders was not “Where can we use AI?” But “What decisions do we make over and over again, and what information makes them better?”
Portfolio not displayed
BHP describes the end-to-end impact of AI on its operations, “from extraction of minerals to delivery to customers.” Leaders have decided to treat AI as an operational capability and move beyond pilot deployments. It started with a small series of problems that affected the company’s performance. Where you can measure changes in results.
The company realized that it could avoid unplanned machine downtime and further enhanced its energy and water usage. Each use case that addressed a small but impactful problem was given an owner and accompanying KPIs. Results were reviewed with the same regularity used to monitor other operational performance within the company.
Where BHP uses AI every day
In addition to a particular focus on areas such as predictive maintenance and energy optimization, BHP also considered using AI in more adventurous but important operations such as self-driving cars and real-time staff health monitoring. Such categories also translate well to other asset-heavy environments across logistics, manufacturing, and heavy industry.
predictive maintenance
Predictive maintenance is the process of planning repairs within planned downtime to reduce unexpected failures and costly unplanned outages. Here, AI models can analyze equipment data from onboard sensors and predict maintenance needs. This reduces the number of failures and equipment-related safety incidents. BHP runs predictive analytics across most of its handling vehicles and material handling systems. A central maintenance center provides real-time and long-range visibility of machine health and potential failures and deterioration.
Forecasting has become an essential part of machine-intensive operations. Previously, such information was presented as “just another” report that could get buried in company bureaucracy. Model and define thresholds that directly trigger actions for teams planning maintenance.
Energy and water optimization
The company reports saving more than 3 gigalitres of water and more than 118 gigawatt hours of energy in two years by deploying predictive maintenance in this way at its facility in Escondida, Chile, all thanks to AI. This technology provides operators with real-time options and analytics to identify anomalies and automate corrective actions across multiple facilities, such as concentrators and desalination plants.
The lesson learned is to put AI where decisions are made. Improvements are further enhanced when operators and control teams can act on recommendations in real time. Conversely, regular reporting means that decisions are only made if both staff members have seen the results of the data and believe it is necessary. The real-time nature of data analysis and the use of triggers for action make the difference immediately apparent.
Autonomy and remote control
BHP also utilizes more advanced technologies, such as self-driving cars and automated machines that support AI. These are high-risk areas and this technology has been shown to reduce worker exposure to risk and reduce the source of human error in incidents. The company has complex operational data flowing from remote facilities through regional centers. Therefore, without the use of AI and analytics, staff will not be able to optimize every decision the way software can.
The use of AI-integrated wearables is increasing in many industries such as engineering, utilities, manufacturing, and mining. BHP takes the lead in protecting its staff, who often work in extremely difficult conditions. Wearables can monitor an individual’s condition, read heart rate and fatigue indicators, and provide real-time alerts to supervisors. One example is the “smart” helmet sensor technology used by BHP in Escondida. This technology measures truck driver fatigue by analyzing the driver’s brain waves.
A plan that leaders can implement
Regardless of industry, decision makers can learn from BHP’s experience deploying AI at the coal site (literally). The following plan will help leaders leverage AI in business problem areas in their own strategies.
- Select one reliability issue and resource efficiency issue that your operations team is already tracking and attach a KPI to it.
- Map your workflow. Who can see the output and what actions can they take?
- Put basic governance in place for data quality and model monitoring, and review performance alongside operational KPIs.
- Start with decision support in high-risk processes and automate only after your team has validated controls.
(Image source: “Coal Mining Shovel View” by rbglasson is licensed under CC BY-NC-SA 2.0.)
Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo in Amsterdam, California, and London. This comprehensive event is part of TechEx and co-located with other major technology events. Click here for more information.
AI News is brought to you by TechForge Media. Learn about other upcoming enterprise technology events and webinars.
