Novel Data Science Methods for Rapid Mineralogical and Grade Prediction in Magmatic and Hydrothermal Ni-Co-Cu-PGE Deposits

This project aims at generating numerical tools and application of Machine Learning Algorithms in understanding the geochemical characteristics of Ni-Co-Cu-PGE deposits.

This approach will predict the problematic zones within the orebody that is linked to decarbonisation and a green future.

This will be achieved using cheap available geochemical data and has proven to be cost-effective because of the turnaround time.

This approach can save time and influence the timelines in the decision-making of mining operations in connection to the grades and geometallurgical properties of the ore to be extracted.

This research can also help greatly in the classification of mineral resources and mineral reserves which is a standard international practice in declaring mineral resource assets.

Predictive modelling of geochemical properties of ore, wastes in mines also helps in the proper design of processing procedures and mine waste management.

The remedial process of dealing with tailings can be improved by predetermining the properties of the waste before they are deposited in their secondary environment which is better as "prevention is better than cure".

Most environmental remediation methods take long to take effect so with predictive modelling its valuable to capture the properties and the continuous process of environmental contamination.