Upscaled mining talent with advanced analytics

Mining companies have set targets to decarbonize and mostly aim to reach net zero by 2040 and 2050, but some are struggling to keep up. According to the World Bank, the mining industry consumes up to 11 percent of global energy use. In comparison, 70 percent of mining projects from the six largest mining companies operate in water-stressed regions. To steer the direction of the process industries toward more sustainable practices, more than 40 governments have adopted a price on carbon. This measure includes additional taxes on the use of fossil fuels, trade programs, new standards for renewable energy, energy efficiency mandates, more stringent emissions regulation, and carbon offsets. This emphasis is also supported by investors shifting their priorities to reduce risk by focusing on long-term plans, including environmentally friendly operations. The metals and mining industry is at the core of the green energy mix and electrification of vehicles. This new trend creates an astronomical increase in demand for minerals. Under a 2-degree scenario, according to the latest analysis from the World Bank, as new energy technologies continue to grow, it is expected that all minerals will double in demand, and the distribution of these materials will depend on their application: Solar photovoltaics will account for 87 percent of aluminum demand and more than 95 percent of silver demand.Wind turbines and geothermal plants will account for 98 percent of zinc demand and 64 percent of titanium demand.Solar photovoltaics and wind turbines combined account for more than 74 percent of all copper demand, and this is without considering wiring or other infrastructure to connect the grid.Battery storage accounts for 100 percent of graphite and lithium demand (Hund, 2020).Other materials, such as nickel and steel, are also projected to more than double in demand as they are cross-technology minerals. These materials require a manufacturing process that is accountable for the emissions proper of the transformation process. It requires that these processes move from a hindsight mode focused on reporting to a forecasting mode based on run-time measurement of greenhouse gas (GHG) emissions. This is a critical first step in companies realizing an economic return for doing the right thing for the environment and their communities. Already, some companies are taking action. BHP is shifting its energy consumption to 100 percent renewables in its La Escondida and Spence copper mines and eliminating groundwater usage in Chile by 2030 by investing in desalination plants. Equipment suppliers like Komatsu, XCMG and Sandvik are investing in battery-electric mining vehicles, some of which are already commercially available. Further down the value chain, Ford Motor Co. (USA), LG Chem (South Korea) and Huayou Cobalt (China) are using blockchain technology to track the origin of cobalt from the Democratic Republic of Congo, hence aligning with the UN Sustainable Development Goals for circular economies. These initiatives are necessary for a profound transformation of the industry and all the other manufacturing industries dependent on it. Digital technologies have a part to play in accelerating their deployment and adoption but also act as a bridge. Most mining and metals companies want to be data driven and make decisions based on key performance indicators and operational performance indicators rather than assumptions. To get there, companies have different strategies to find insights into the data that would enable them to make improvements in the process to keep up with the increased demand and to enable the adoption of new technologies. These strategies require new organizational capabilities. Companies can build or develop their abilities or buy or acquire the necessary skills. Both options present different challenges. Buying versus building In the case of buying, the first challenge is budget. Recruiting data scientists comes with a high price tag due to competition; manufacturers are not only competing against each other but also competing against big technology, financial and consulting firms for this talent. Indeed, hiring a data scientist will bring the organization the skills to work with advanced mathematical models to analyze data, proficiency in different development languages, and expertise in data analytics tools. Nevertheless, the second challenge for data scientists working alone is that they do not have the context of the process and will not be able to determine whether the correlation is the cause of the scenario being analyzed. The third challenge is well known, and it has to do with the amount of time necessary to prepare a good data set for analysis — 80 percent of the time is dedicated to finding, preparing, cleaning and aligning the data for the data set. In comparison, 20 percent is used to do the actual analysis. This challenge is aggravated by the lack of process knowledge a data scientist may have. In the case of building, the first challenge is recognizing the organizational changes necessary to enable collaboration and capture knowledge, for example, creating the means to connect data scientists, if they are available in house, with subject matter experts (SMEs). The second is removing the information technology (IT) hurdles for experts in the process to access the data and perform the analysis; in many cases, this involves access to cloud technologies still not widely adopted in the industry. The third is to develop the skills for a range of baseline analytics and programming capabilities. This is not an easy task, and people need to have different entry points depending on the individuals and their experience in data analysis; most engineering curricula already have statistics requirements, but not all of them do. Impact of digital technologies and data analysis on operations To understand the areas of opportunity to maximize production capacity, one must know the historical losses, the signs leading to these losses, and the past actions for mitigation. Tracking and categorizing losses, such as water and energy consumption while in idle mode, helps identify bad actors and justify improvement projects. Performing this kind of root-cause analysis leads to identifying areas for further analytics: for example, benchmarking different ore grades’ efficiency or forecasting consumptions based on other operating conditions. Based on this analysis, summary views can be assembled in a report to share the insights necessary for collaboration. A similar approach can be taken for GHG reporting. Self-service analysis tools can provide engineers with the means to streamline to perform the calculations and reports necessary for compliance. In many cases, soft sensors can be created based on the current telemetry and stichometry of the process. As more sensors are put in place for accurate run-time readings of emissions, the data-flow processes already in place will continue to drive the reports put together. Use case in the mining industry Typical concerns for less-experienced customers in data analysis include: How to use subject-matter expertise to build applications or domain-specific models.How to reliably apply the technology in a plant. In other words, how to demonstrate that the solution is not a black box and how it fits into existing workflows.How users can collaborate and communicate insights and value created to stakeholders. SCM Minera Lumina Copper, Caserones is leveraging its data infrastructure in place. Caserones began by building a digital twin of its operations and connecting the different objects for developing a complete mass balance model by connecting all other process units, from the grinding to the tailing sections. Another critical aspect is understanding the recovery curves for the plant. These recovery curves do not isolate the stages in the process. Instead, they look at the particle size in the crushing and grinding process and its impact on the flotation and, subsequently, the water and mineral recovery from tailings. In their case, the Gaudin-Schuhmann approach is required for a hydrodynamic model to predictions in particle distribution that can be used to optimize the mine operations. (Bascur, Retamal and Espinosa, 2022). Regression models to forecast operational envelops can either be built using self-service analytics tools or developed manually using Python code. In the case of Lumina Copper, code development was kept to a minimum and used to integrate different systems to streamline workflows. With this approach, Lumina Copper has improved its copper recovery by 4 percent, and water recovery has been enhanced with 1,360 t/h (1,500 stph) put back into the grinding system, saving millions of dollars. Conclusion Mining and metals companies must be aware of the demand pressures they experience and have clear strategies to comply in sustainable ways to supply the minerals they produce. Transformational changes will likely need to happen, like adding renewable energy to the operations, electrification of shipping methods to decrease the consumption of fossil fuels, and collaborating across borders and industries to uptake circular economies. Although these measures have a significant impact, they are also the most expensive and hard to implement in a few months. As increasing digitalization creates a corresponding shift in employee expectations, manufacturers must provide appropriate solutions and environments to ensure success in today’s evolving professional world. Examples include remote work options, interactive training, agile and rapid collaboration platforms, career development, work-life balance and others. Empowering the SMEs at the mills or working remotely to make a significant impact on operations and achieve savings in the consumptions of energy, water and time necessary for compliance reporting needs removing barriers to data. SMEs can leverage self-service applications designed for descriptive, diagnostic, predictive and prescriptive analytics. These tools incorporate visualization into the data-analysis process, empowering SMEs to visualize the impact of their data analysis immediately, identify missteps and successes in real time, and iterate and innovate faster. Organizations need to enable this by building or buying modern analytics capabilities. To assess whether organizations are ready, they should ask themselves: Do I have a healthy level of automation? Does that include an operational data management infrastructure? Do I have resources that understand my operations and processes well? If the answer to the two questions is positive, then the mine is ready to adopt self-service analytics to leverage the “gold mine” of data they sit on. The gains will be translated into more efficient operations, higher water recovery, savings in energy consumption, and freeing engineering time for continuous improvement. References Bascur, O., Ramayal, F. and Espinosa, S. (2022). Treating Low Grades Ores using Digital Twins in Mining and Mineral Processing Plants. Santiago, Chile: Copper International Conference 2022. Buenemann, Allison (2022). “The data science dilemma.” Control. https://www.controlglobal.com/manage/dataanalytics/ article/11287230/the-data-science- dilemma. Deloitte Global. (2022). Retrieved from Energy and Resources: https://www2.deloitte.com/global/en/pages/ energ y- and-resources/articles/the-2030- decarbonizationchallenge. html. Gupta, A. and Yan, D. (2016). Mineral Processing Design and Operations. In A. Gupta, and D. Yan. Elsevier. Hund, K. L. (2020). Minerals for Climate Action: The Mineral Intensity of the Clean Energy Transition. Larson, Keith (2022). “Get a handle on your GHG metrics” Control. https://www.controlglobal.com/home/ article/11287192/get-a- handle-on-your-ghg-metrics. Ottewell, Seán (2022) “Asset Management: AI Stands For “Applications Increase” Chemical Processing. https://www.chemicalprocessing.com/asset- management/ digitalization-iiot/article/11414853/asset- management-aistands- for-applications-increase-chemical- processing Sandin, M. (2021). A Digital Path to Improved Sustainability. MLC Journal, https://www. manufacturingleadershipcouncil.com/a- digital-path-toimproved- sustainability24663/?stream=ml-journal Washington, DC.: World Bank Group.

Upscaled mining talent with advanced analytics

Authored by:
Mariana Sandin

To view the past issues of the UCA magazine you must be a member of SME/UCA.

From sessions to strategy: Revealing long-term and emerging innovation trends in data analytics, intelligent systems and automation

Mining education gets federal boost as industry faces workforce crunch

Critical minerals management on the federal seabed

From sessions to strategy: Revealing long-term and emerging innovation trends in data analytics, intelligent systems and automation

Mining education gets federal boost as industry faces workforce crunch

Critical minerals management on the federal seabed

To view the past issues of the UCA magazine you must be a member of SME/UCA.

SME GPAC provides comment on recent Interagency Working Group report

Sandvik, SSAB partner for future fossil-free steel mining equipment

Related Articles

From sessions to strategy: Revealing long-term and emerging innovation trends in data analytics, intelligent systems and automation

Critical minerals: Technical and policy considerations

Historic day in Washington DC for mining education

Copper stockpiling, transformation capacity and industrial synchronization in the US

From sessions to strategy: Revealing long-term and emerging innovation trends in data analytics, intelligent systems and automation

Mining education gets federal boost as industry faces workforce crunch

Critical minerals management on the federal seabed

From sessions to strategy: Revealing long-term and emerging innovation trends in data analytics, intelligent systems and automation

Mining education gets federal boost as industry faces workforce crunch

Critical minerals management on the federal seabed

Use Single Sign On