The mining industry has seen a significant drive in technological advancements over the last couple of years, from the automation of mobile equipment, remote operation control centers, to smart cameras that improve surveillance and security. But one aspect of mining that, in my opinion, still has a significant opportunity to be “mined” for more benefit is the area of scheduling and planning. Schedule and planning optimization can significantly benefit from simulation in the strategic longer-term decision making and in shorter-term tactical decision-making. As I will explain, open-pit mine simulation can contribute to such improvements in planning.
I believe that the recent advancements in the various spheres of artificial intelligence (AI) will be one of the aspects to keep our eye on. Still, in the meantime, we have a set of tools that can already add a significant amount of value right away: open-pit mine simulation. Not only is simulation already being used as a decision support tool at various mines around the world, but it will likely be a necessary accompaniment to AI. When AI algorithms and policies eventually become mature enough to assist with the decision-making, simulations will be there to provide a safe, risk-free environment to test these tools before they go and do their work in the real world.
Before we dive deeper into the reality of open-pit mine simulation, let’s see why mining is such a complex environment and why making these tactical and strategic decisions can be so tricky.
Although this article focuses on simulation it by no means suggests that simulation is the “only” tool in the toolkit of mine planning and optimization.
Major challenges in mine planning and operation make step changes in improvement difficult
Managers at all levels of an organization are responsible for making reliable commitments and constantly trying to improve with ever-higher targets and incentives to improve performance. In any complex system, this is really difficult due to VUCCA. The term initially only incorporated one C, VUCA, but I prefer to use the former framework as defined by Dr. Alan Barnard at dralanabarnad.com:
As in most interconnected systems, mining experiences significant volatility. For example, there is considerable volatility in:
- demand and prices of their products.
- performance of the machines and humans.
- the variability of the ore quality.
- the efficiency at which ore can be extracted.
These factors can never truly be considered using an average value and only within a range given a specific confidence interval. This also means that providing a single number as a figure of commitment for any output level within the organization can be extremely unreliable. It would be best to specify any output as a range, based on some consideration for the best, worst and likely output due to defined input variables.
It is often difficult to determine the cause-and-effect relationships within large and complex systems. We might make a decision today that impacts performance weeks from now. How will we know that we should have made a different decision? And if we did, how sure are we that the outcome would be better than what we have now? Can we do a test and wait another few weeks and see if it improved? What if the outcome is much worse?
There are many cause-and-effect relationships associated with a high level of uncertainty in the mining industry and mining operations.
In any organization or system with many moving parts, there is a level of complexity, not only due to the many cause-and-effect relationships that exist between points but also because of the sheer number of points.
These complexities are challenging to map out in a single mind map, Excel spreadsheet, ERP system, and many other traditional decision support tools. Simulation is able to handle feedback loops which most linear tools cannot handle.
There are many constraints within which the organization must operate to survive in any organization. Among others, these can be:
- Money – Budget constraints, cash flow, loan repayment, supplier and buyer terms etc.
- Time – Limited due to environmental factors e.g., weather, or simply the competition and demand of other competitors e.g., first come first serve.
- Labor – Limited availability of skilled or other workers to complete the desired tasks in an effective and efficient manner.
- Compliance – These can be any regulatory, environmental or legal requirements.
A mine must be able to navigate and consider all of these constraints to make better tactical and strategic decisions.
The ambiguity here refers to the confusion created by having more than one, often competing, objectives within an organization. For example, in a mine, you need to be safe, productive and cost-effective. The safety rules often impact the speed of the work, thus reducing productivity, and to be more productive, you need more machinery, which might be very costly.
[Update: It is well understood that increased safety actually increases performance over the long term. These safety checks everyday waste much less time in the long run than not having them and causing serious downtime, failures or worse fatalities]
Managers need to consider the trade-off between the different objectives and ensure they can simultaneously adhere to all the requirements.
These VUCCA elements make step changes in improvement difficult within mine planning and operations since it is so hard to predict the impact of changes on the entire system. People will often make changes that improve their local environment or subsystem without knowing the overall effect on the entire system, thus potentially sacrificing a global optima rule for local optimization.
Challenges in open-pit mining addressable with simulation
Dynamic simulation can be applied to solve a wide range of particular challenges in open-pit mine planning and operation. In the figure below, you will find an illustration of the primary operations and processes in open-pit coal mining.
This example is simplified, but nevertheless, it highlights some of the major challenges that cannot be solved with strictly analytical approaches.
The process starts with drilling holes in the overburden, then blasting the overburden, dozing the overburden, and removal by trucks and shovels.
Once the coal layer has been uncovered, there is another round of drilling and blasting before the coal can be hauled and taken to stockpiles, where it can be processed further. As a simulation expert when interacting with mining managers in operations of this kind, I often encounter the following questions related to production planning and operations management:
As you can see, these are non-trivial questions to answer. For example, the same trucks are used for overburden removal and coal haulage so distributing them across these two types of work is a dynamic problem that requires a tool to evaluate the impact with a high degree of accuracy. The result of these decisions can be that the mine either makes money or goes bankrupt…. so how will the mine manager solve this conundrum?
Open-pit mine simulation improves planning and decision making
Although many alternative methodologies exist to help answer the questions stated above, one specific methodology stands out amongst all others: Simulation. The reasons for this are manifold:
- Consider all the critical system interdependencies, constraints, complexities and variability.
- Provide a range of likely outcomes for single scenarios, do sensitivity analysis and direct scenario comparisons.
- Provide a low-risk, low-cost way to test the impact of any changes on both operational and financial performance.
- Get fast feedback by testing various scenarios in a relatively short time.
When built correctly, simulation models furthermore allow for scalability and flexibility, meaning that the same model can be used to simulate the mine now and in the future. This allows for a much higher return on investment.
Solving mine planning and decision making problems with MineTwin, an exemplary tool for open-pit mine simulation
Today I will take a quick look at a very specific simulation tool that is specifically used for simulation of both open-pit and underground mines and test just how powerful it is in taking into consideration all the VUCCA elements and providing the benefits stated above.
Demo model using MineTwin to model and investigate operations in a big open-pit mine in Africa
Please watch the video below for the whole demo.
In the video above, we use a real mining environment, one of the big open-pit mines in Southern Africa. The specific question we try to answer is what is the number of trucks and dozers required to meet the mine plan, and then subsequently, what is the optimal number of excavators required. The mine plan is based on the processing plant’s capacity, which is the bottleneck, so we do not want to starve the processing plant, and we do not want to build up massive stockpiles in front of it. Below is a quick overview of the steps shown in the video.
Step 1: Setting up the open-pit mine simulation model
The first step to set up the simulation is to import the mine layout and specify the equipment, the mine plan, and many other input parameters and constraints inside the model. Once the scenario has been imported and checked for setup errors by the platform, we can verify the setup with the mine personnel through an interactive animation.
Step 2: Setting up the simulation experiment
We create a new sensitivity analysis experiment, set the number of replications, to ensure we get a statistically valid sample, and hit the play button.
The output indicates that the mine plan can be achieved with six trucks and seven dozers. Adding in more trucks actually leads to a deterioration of performance, even if it is just by 0.1%, as the more trucks we have, the more congestion we get inside the pit.
NB! This is a unique insight that no statistical or deterministic model will ever be able to show you!
In the previous example, the number of excavators was set extremely high to ensure they were never a bottleneck. Now that we know the number of dozers and trucks required, we can set up a similar sensitivity analysis for the excavators.
It appears we have answered all the questions, in record time and with a high level of confidence that we have considered a significant number of elements.
Summary and conclusion
In conclusion, many constraints and interdependencies make decision-making in mines difficult, and many static and deterministic tools are inefficient in supporting critical decisions. The use of simulation overcomes many shortcomings and provides real insight with a range of outputs instead of just a single number.
Simulations provide a cost-effective, risk-free environment that can deliver results promptly again and again.
I am passionate about solving seemingly complex problems with simple solutions and teaching others how to do the same. This passion has led me to focus on using technology, especially simulation to, find, test and validate potential solutions to common business problems.
I have over a decade of experience in using simulation and other tools to provide valuable insights to clients and partners.
Visit my website, The AnyLogic Modeler, where I and other simulation experts regularly write about Simulations, with a special focus on AnyLogic.