Intelligent mining systems address changing energy demands

Advanced analytics solutions provide the tools for mining enterprises to effectively scale their operations to meet increasing global demands.

By Mariana Sandin

The global demand for cleaner energy sources to meet net-zero goals, which requires a doubling of base metals output over the next few years, has spurred a significant transformation in the mining industry. To achieve and track such ambitious demands, the mining sector must outfit both greenfield investments and brownfield asset improvements with reliable data analytics and innovative software as a service technology.

As more companies embrace data-driven decision-making processes, the ability to explore and analyze data from operations, maintenance, asset performance and more across different levels of the organization will drive success. Yet, due to the complexity of gathering, cleaning, stitching and organizing the data to uncover insights, highly valuable data analyses have traditionally been performed infrequently in the mining sector.

Fortunately, the industry is shifting toward intelligent mining systems by adopting modern software technologies, such as advanced analytics solutions, to seamlessly access, cleanse and analyze data from disparate sources in a common framework. Equipped with self-service analytics, subject matter experts (SMEs) and operational talent in mining environments can leverage their domain expertise to proactively improve recovery yields, while increasing operational efficiency.

Workforce and operations
Most mining companies face significant challenges surrounding operational data and analytics. To start, most mines operate in lean environments around the clock, relying on highly experienced engineers to maintain uptime. Most engineers in mining environments are accustomed to firefighting, which limits the time available for innovation and process optimization. When time does become available to focus on these efforts, many of the spreadsheets and data sources used for analysis paralyze the task because they are disjointed and cumbersome to work with.

Spreadsheets present significant limitations, including live data connectivity difficulties, a lack of computational capability, poor online collaboration, and unsophisticated visualization and reporting functionalities. Without live connections to both historical and live data sources, engineers in mining environments must manually query each individual database, extract the necessary data, then aggregate and align mismatched timestamps in a spreadsheet. When a new time period of interest is identified, the process must be repeated.

Additionally, finding experienced engineers in mining operations has become especially difficult in recent years. As personnel retire or leave roles they held for many years, companies are struggling to find easy methods to scale process knowledge and maintain productivity. Attracting new talent is also challenging because mining companies are competing with innovative, tech-centric businesses recruiting graduates with niche process and data analytics skills.

The mining sector has met various technical and environmental challenges in the past, but the pace of change and the number of moving targets in today’s landscape is unmatched. With threats to production and incessant consumer demand, these mining companies need the best human engineering talent, along with the right software solutions that empower personnel to meet ambitious production targets.

Solutions in advanced analytics
By leveraging cloud-based, self-service advanced analytics solutions, mining companies can begin navigating these complex industry challenges. From straightforward – such as calculating a net recovery rate based on the current ore grade mix – to highly complex analyses, like forecasting water recovery rates based on ball-mill performance, these solutions can perform a wide range of computations to transform raw data into insights.

For advanced analytics solutions to make a worthwhile impact, they must provide comprehensive functionalities, such as connecting raw operational data from diverse sources, cleansing it for insight creation, integrating it into vital business and maintenance systems, and promoting team collaboration. 

Upon implementation, advanced analytics solutions connect disparate data sources in a single platform, bridging near-real time data, and enabling SMEs – including process engineers, operations managers, analysts, data scientists and more – to quickly identify insights across all sources with full contextualization. Operational data, including high-frequency time-series data, exists and is available within control networks or in data historians, and it is commonly used for operational awareness at mines and in remote operating centers.

By removing data access barriers for SMEs, advanced analytics solutions empower them to leverage point-and-click interfaces for descriptive, diagnostic, predictive and prescriptive analytics. These visual tools help users identify the impact of analyses, identify missteps and successes in near-real-time, and model and innovate operational strategies.

These solutions also enable organizations to maximize the effectiveness of SMEs – whether working at the same site or from different sites across countries or continents – by facilitating streamlined collaboration, along with knowledge capture and transfer. This helps teams not only analyze their data, but tell the story of findings at different levels throughout an enterprise. 

These findings accelerate organization-wide sharing of best practices for predicting and preventing common failure modes among critical assets, which can then be shared and scaled across multiple sites and used to train new personnel throughout the enterprise.

The successful deployment of digital solutions largely depends on the maturity of operational data utilization. Typically, this progression spans from descriptive analytics – which detail current conditions – to prescriptive analytics, which suggest ways to achieve desired outcomes.

Advanced process control for ore grinding
Ore grinding is very energy intensive, requiring nearly half of all electric power used throughout the entire mining process. This procedure is the bottleneck in most operations, so any improvements to this step are felt in full force at the output stage. For example, increasing ore grinding production by one percent would result in a one percent overall increase in the entire operation.

As corporate initiatives increasingly focus on sustainability, mining organizations are looking for new opportunities to maximize energy use with long, sustained periods of high production.

The Swedish metals and mining giant Boliden leverages Seeq to determine ways to increase these sustained periods of high production. The company began by running a simulation model of its grinding process to benchmark current operations against a new control system to identify scenarios that would increase production.

Next, the team used the advanced analytics solution to look at historical data for both the absolute power utilization and the relative power distribution utilization to calculate the available power. Using the solution to complete the calculation not only saved Boliden significant time, but it also empowered the company to visualize the calculations, motivating team members to buy into the idea of implementing a new control (see Figure 1).


Figure 1. Swedish metals and mining giant Boliden leveraged Seeq to identify ways of increasing sustained periods of high production by examining historical data for various power utilization properties in their operations.

Figure 2. Boliden used Seeq to establish multiple conditions for an accurate analysis, which revealed a 15% increase in sustained periods of high production.

Using the benchmark and baseline calculations, the team decided the next step was implementing a new advanced process control for the grinding circuit. However, before evaluating the new control system, the team leveraged the advanced analytics solution’s capabilities to improve, refine and add nuance to the analysis, including setting up conditions to identify and correct load limitations for a more accurate evaluation (see Figure 2).

Monitoring and evaluating the new control within the advanced analytics solution, the data revealed a 15% increase in long, sustained periods of high production. It also helped identify the new operating regime’s impact on grinding energy consumption, revealing encouraging early results of a 2.8% efficiency increase.

Power export commitment predictor
When a new local government regulation required facilities to report predicted daily power exports, carbon black additive manufacturer Birla Carbon sought a method for predicting its power export commitment. The changing dynamics of daily production and consumption made this a difficult task, with multiple lines producing different product grades at any given time, using different feedstocks of varying quality, and not all equipment continually running at scale.

To avoid revenue loss from either under- or over-committing to an export, the company leveraged an advanced analytics solution to build a machine learning model for predicting the facility’s daily power export. By connecting all data sources in the solution, Birla Carbon could visualize the effects of the entire facility’s power generation and consumption, and accurately predict – within a 2% margin of error – the next day’s power export (see Figure 3).

As a result, Birla Carbon eliminated cost penalties due to short supply and increased revenue with accurate predictions of their power export.

Figure 3. Birla Carbon used Seeq to connect multiple data sources and create a machine learning model for predicting next-day power export from a facility within a 2% margin of error.

Advanced analytics solutions empower adaptation
Global sustainability initiatives are rapidly transforming energy demand, and mining companies must adapt to meet these changing needs. Left to the tools of yesteryear, there is too little time and too few knowledgeable resources to make the necessary adjustments.

But advanced analytics provide the tools for these companies to efficiently analyze their process data and optimize operations to increase production, accurately forecast output and improve energy consumption efficiency. These outcomes make organizations more resilient and competitive, empowering them to face the fast-paced changes today’s industries demand.

Mariana Sandin has more than 15 years of experience in enterprise industrial software and analytics, and she leads the mining, metals, and materials practice at Seeq. Mariana has an MBA from the University of St. Thomas in Houston, TX, and a BS in Chemical Engineering from Nuevo Leon, Mexico. She also has several certifications from TAPPI and IIMCh. Mariana lives in Houston, TX, and enjoys family life and painting, when she is not traveling to attend industry-related conferences and symposiums.