1. How to blend the ore best?
2. And does the chosen blending recipe or logic allow for meeting production targets?
Optika Solutions has been engaged previously by a large mining company to answer those questions while the process was still in its design phase.
Following extensive workshops with the company’s engineering team; it was decided to model the system by Discrete Event Simulation. Since then, the Discrete Event Simulation model has yielded in several business benefits including:
• Understanding the impact of grade variability on the feed, as derived from the mine-plan. This can be assessed and allow the decision maker to get a clearer indication of the risk involved in the planned operations.
• A better understanding of the proposed operational changes, (given the current plant design and mine plan), to allow the decision maker to make changes to increase the chances of on-plan delivery, prior to the start of the operations.
• Allowing the decision maker to automate the feasibility assessment process and costing, freeing uptime from this otherwise tedious and labor-intensive task.
In order to be able to answer the two questions posed earlier, it was necessary to run countless “What-if” scenarios. A “What-if” scenario is usually branched off the baseline model for the purpose of answering questions like “What if we improve the availability of our crushers?” or “What if we alter our blending recipe that way?”.
Optika Solution’s analytics platform Akumen has been proven to be the right tool for this problem through its inbuilt scenario management and execution features. Furthermore, Akumen’s Asset Library was used as a single source of truth for all asset related data helping to uncover and settling disagreements in process configurations.
Scope of the system that was to be modelled is the material flow between mining pits (cut-backs in the deposit) and the processing plant. Mining pits and processing plant were treated as “black boxes”, meaning their behavior was only modelled on an abstract level. Significantly more detailed were the models for trucks, conveyors, crushers and stockpiles.
Each one of those “mining objects” obtained its own logic such that their internal processes could be represented realistically.
For instance, a logic was implemented that allows trucks to be re-routed to a stockpile with a grade and hardness range fitting the loaded ore, if the tip pocket is utilized fully.
In addition, two methods for the grade blending were implemented. Blending on the coarse ore stockpile happens according to a user-specified recipe. In other words, the user specifies a percentage for each pile that is converted into a rate at which the pile is drawn down. This allows balancing out periods of low or high grade by blending in stocked ore. Blending on the ROM on the other hand is done through trucks that transport reclaimed material to the tip pocket. A continuous reclaim rate based on a recipe is not applicable here.
For that reason, a complex logic was implemented that picks specific stockpiles to reclaim from based on a range of criteria to achieve an optimal material feed to the plant.
Based on the developed model, it was shown that a coarse ore stockpile between crushing and the plant is beneficial in multiple aspects. First, it allows keeping the grade of the plant feed in over 95% of the time within the target range. Second, it enables meeting operational targets on throughput and utilization, since it decouples crushing and ore processing.
Overall, it turned out to be financially advantageous to set up the process with a coarse ore stockpile.
In conclusion, Optika Solutions has helped the mining company in determining an optimal blending strategy as well as in gaining confidence in their system configuration through Discrete Event Simulation.