The mining industry is in a period of transition.
Operating conditions are becoming more challenging and the industry itself is now more competitive.
Forward-thinking mining companies, including the 'Big Four,' are looking at ways to optimise their supply chains by refining processes to get more out of their assets while reducing capital expenditure.
Their goal is to build more profitable mines in lower-cost environments and they're using modelling and simulation technology to do so.
Like any major business initiative, the development of a mine involves substantial time frames and vast amounts of capital.
Because a mine can be in operation for anywhere up to 80 years after it has been developed, the decision to dig (and the specifics of how best to do so) needs to be made based on intelligence that's as comprehensive as possible.
In order to ensure the profitability and success of the mine long-term, ongoing evaluation and monitoring of the supply chain and logistics are essential.
A growing number of miners, including several of our customers, are finding that by applying modelling and simulation techniques, processes can be streamlined to enhance the decision-making process, particularly regarding "future-proofing" areas such as risk evaluation and mitigation. Moreover, these techniques and the processes they result in can also improve the efficiency and profitability of the mine itself.
Scalable, usable, interactive
Although the mining industry in Australia has been slow to pick up on the benefits of process and data modelling, those in the know are using simulation tools for all levels of decision making. From reducing exploration and development costs to optimising processes in order to increase production rates, these tools need to be scalable to an unlimited number of applications and scenarios.
Perhaps more importantly for adoption, miners need to combine data from multiple sources into a single, integrated environment.
To date, most miners have built up arrays of systems that each have different data formats and extraction methods, making it difficult for managers to gain a holistic view of how changes to one area may affect another.
A data modelling tool such as MATLAB, developed by MathWorks, collates and analyses data from a full range of sources, enabling customers to transition from one piece of analysis to the next without having to change tools or reinvent the wheel.
Preliminary modelling to qualify risk
Within the mining industry, MATLAB is already being used to improve configurations across the supply chain, conduct geospatial and seismic analysis, forecast economic risk and profitability, and streamline the development of mining equipment.
Many customers using the software in the preliminary phases of construction are finding that a little groundwork up front can go a long way toward ensuring the viability of the mine long-term, as well as forecasting any threats that may impact its profitability.
For instance, one MathWorks customer has taken on this sort of analysis to better evaluate how the location of transport infrastructure will impact the mine financially over a period of time.
The customer's goal is to find the "golden mean" between time to market and product quality, focusing on storage elements in the supply chain.
Since larger storage amounts give a miner potentially higher product quality but slow down the time to market, it's in a miner's best interest to find the amount of storage that maximises margins while staying competitive.
Modelling software has allowed this customer to properly evaluate and compare the costs associated with different build options, particularly those that may require additional infrastructure in more remote areas.
They're using what we call "Real Options Valuation," a reliable way to assess the outcomes of different scenarios. Using this method, a separate model is created for each aspect of the project.
The models are then used to simulate distributions of outcomes for different economic scenarios. Using the resultant scenarios, analysts can more accurately assess the upside and downside economic risks and recommend possible responses, such as deferring, abandoning, expanding, staging, or contracting capital investment in the project.
Statistical and process modelling
While this technique can be used to aid high-level supply chain decisions, it's also used on a day-to-day basis to assist with more granular considerations.
Apart from modelling the correlations between storage volume and profit without expensive "trial and error," miners can also use statistical models of ore and equipment to evaluate the quality and quantity that a processing plant will generate.
Looking more broadly at the supply chain, process managers can use time-domain models to identify bottlenecks and estimate delivery schedules with far more accuracy and insight. Even mine sites themselves can benefit from modelling power and water supply requirements that each specific operation requires, as well as examining how different supply and treatment strategies will interface with the site's unique conditions.
In high-cost and high-risk industries such as mining, this sort of predictive technology can be the difference between a profitable or insolvent operation -- especially when its modelling range is expanded across the entire operation and supply chain.
For many miners, process modelling technology is unchartered territory and considered time-consuming to master. When evaluating tools, mining leaders should prioritise access to a simple, graphic interface that's both interactive and scalable.
MATLAB, for example, maps across the collection of data, simulation, processing, optimisation, and automation so that users are never limited to how they can apply that information. Our customers are finding that when it comes to supply chain management, process modelling leads to more informed decision making across all aspects of the mine.
And, it may just prevent you from throwing money down the mineshaft.
*Sam Oliver is an application engineering at The Mathworks.