A numerical sensitivity study – The effectiveness of RFID-based ore tracking through a simulated coarse ore stockpile and the impacts of key process variables

The ability to understand ore characteristics in real-time during mining processes is vital for ensuring product quality control. However, it is challenging to continuously track ore flow from the mine to the mill due to the blending of ore batches, especially within stockpiles. This paper presents a numerical study of copper ore tracking through a coarse ore stockpile. A discrete element model of a 3D stockpile was created using the EDEM software to evaluate the effectiveness of using RFID tags for ore tracking. To identify the primary variables and their effect on ore transport and tracking through the stockpile, a sensitivity study was conducted to investigate a range of process variables, such as ore size distribution, ore size range, RFID tag size, wall friction, the trajectory of charging particles and stockpile charging methods. The results show that the stockpile model is not sensitive to variables such as the ore size distribution, ore size range and RFID tag size, while wall friction, stockpile feed belt speed, segregation in the ore flow region and contact model have a significant effect on ore blending within the stockpile. It was found that the overall performance of RFID-based ore tracking through the stockpile is poor. For cases with only one or a few tags per ore batch the order in which the tags are read did not provide a good representation of the ore distribution for most scenarios This sensitivity study provides insights into new tracking strategies given the poor performance of RFID tracking shown by the simulation study.