Mining Seeks to Bring AI to Operations

The “Mining Summit: Empowering Mining Through AI and Autonomous Solutions,” held on Sept. 24 in Santiago, Chile, addressed the adoption of artificial intelligence and autonomous solutions in mining operations. Organized by AmCham Chile and Honeywell, with sponsorship from Microsoft, the event linked the demand for minerals with the challenge of leveraging available technology and expertise to make timely decisions.

Data centers require electrical infrastructure and minerals such as copper. Mining, in turn, seeks to use AI to improve its production. This relationship framed the discussions on productivity, technology adoption, and safety.

José Simón, Vice President and General Manager for Latin America of Process Automation, Honeywell Technologies, described the difficulties of producing more copper when its concentration in the rock decreases. This forces the processing of more material to obtain the same amount of metal. Added to this demand are the lack of specialized personnel and the cost of unscheduled downtime.

To address these pressures, he proposed expanding the use of advanced process control, training simulators, remote operation centers, and systems to manage equipment performance. “Once implemented and in service, these technologies become the fundamental foundation for deploying AI and autonomous operations,” he explained. His approach is based on leveraging the infrastructure of existing plants.

Purchasing technology, however, does not guarantee that a company knows how to use it. “Mining companies are in a situation where the capacity to acquire technology far exceeds the capacity to use and leverage it,” warned Joaquín Villarino, Executive President, Mining Council. For the executive, adoption requires management and boards of directors to dedicate time, training, and attention to the transformation, in addition to reducing internal barriers to introduce changes in continuous operations.

Villarino also pointed out the risk of losing knowledge when experienced personnel retire. The generational turnover requires preparing new profiles and preserving accumulated experience, a task that extends to both mining companies and their suppliers.

Jason Urso, Technology Director of Process Automation and Industrial Automation, Honeywell Technologies, presented “digital cognition” as a combination of sensors, process models, and AI to interpret situations and recommend actions. His proposal would complement deterministic controls, which execute rules with defined responses, in decisions that still depend on the operator, such as responding to an alarm, adjusting an operation, or scheduling a repair.

According to Urso, plants already preserve part of that knowledge in maintenance logs, alarm sequences, and operator actions. Analyzing these records would allow for the retrieval of previous responses to guide workers with less experience. He acknowledged that AI can make mistakes and proposed incorporating personnel evaluation to improve its recommendations.

He suggested starting with a specific operational problem and testing the solution on a unit or piece of equipment before expanding it. “Do not start with AI as a new tool to then look for where to use it,” he noted. Unplanned downtime or process variability would be starting points to define what needs to be solved.

Armando Pazos, President of IMC, Honeywell Technologies, and Global Head of Instrumentation, Measurement, Control, and Gas Detection, steered the discussion toward worker protection. He described systems that, he explained, combine local measurements and centralized monitoring to warn of risks and track a person’s location and exposure in the event of an emergency. The goal is to “be proactive regarding gas exposure and avoid having to take action when it is already too late.” He linked this prevention to reliable sensors and the need to avoid false alarms or late detections.

José Magalhães Fernandes, President for Latin America and Vice President and General Manager of Process Technologies, Honeywell Technologies, raised the challenge of resources: producing more also requires more water, energy, and workers. He linked electrification and the growth of data centers to opportunities for critical minerals. He acknowledged the difficulty of meeting demand and maintained that “it is technology that will help minimize the existing gap between production and demand.”