Boliden Edge Lab will pave the way for new ways of working.

Digitalization, automation, and artificial intelligence are transforming workplaces across industries and sectors – from construction to energy and industrials. Major investments are being made around the world. A subset of artificial intelligence, machine learning, is emerging as an invaluable asset. It enables machines to acquire existing data and improve it to make decisions or predictions. Large amounts of data from the production environments can be interpreted and inserted into models.

At Boliden a world-unique work is being done to develop machine learning and automation in our mines. The development takes place in the form of a cross-functional R&D program in collaboration with partners. Artificial intelligence (AI) shows great potential for a variety of applications and is, together with other technologies, making automation possible. 

In the spring of 2023, the project Boliden Edge Lab was initiated. At the time, there were lots of good ideas within the organization, but we had a hard time implementing them. The project goal was to continuously enable digitalization initiatives, to increase productivity, improve safety and utilize new technology advancements. There was a need to create context where we can move forward at a fast pace despite not knowing everything about the future.  

Boliden Edge Lab is a modern AI development and accelerator area. It can be explained as an inhouse hybrid cloud solution that combines the ways of working from the cloud through a cloud-native architecture, along with powerful hardware inside the firewall. This allows us to explore and realize applied AI functions with a much shorter timeframe from idea to deployed AI solution. 

Boliden Edge Lab in practice  

Thanks to Boliden Edge Lab’s cloud-native infrastructure, we are able to conduct Research and development (R&D) activities in a faster and more flexible way. For example, we can design state-of-the art model-based controllers, or AI-driven models, and execute them as containerized applications in Boliden Edge Lab. 

One of our research projects, PID-MPC Hybrid Control, investigates the advantages of a model-based control structure for one of our open pit mines, Aitik’s flotation process. To keep the levels of slurry just right in each tank so the process works smoothly, we have historically relied on a traditional regulator structure. However, when the operating conditions change, we need to handle it differently. This project is trying out new ways to control the levels. The aim is to try the control structure in live production without affecting the operations negatively, by testing it all in computer simulations instead of changing things at the actual location. One interesting aspect this project addresses is how to send a control signal securely and reliably from Boliden Edge Lab back to the main control system where all the other controls are managed. This will pave the way for new ways of working in terms of how to design smart automation solutions by utilizing modern cloud-native technologies and software components. 

Boliden Edge Lab was nominated within the category Data and AI-driven Transformation Project of the year at the 2023 Nordic DAIR Awards.