Leveraging Machine Learning for Mineral Exploration
Mineral exploration is a complex endeavour for explorers. Mine prospecting is becoming more and more expensive as new mineral deposits become harder to find. The iterative process of collecting different datasets, followed by geological interpretation, can take an extremely long time. Vast amounts of data are collected and processed, very often without any significant mineral discovery. Therefore, explorers are seeking new approaches and innovative processes that can drive up discovery rates and speed up the exploration life cycle. In order to predict and discover new mineral deposits, therefore, we sought to develop solutions capable of handling data at scale so that it can extract insights with greater precision and in less time.
The advancement in technology of data collection and storage has increased the volume of data available to the mining industry, it has made data of the geography of an area through geological, geochemical and magnetic surveys made by geologists and a historic record of successful explorations for mines abundant. There are numerous variables which are to be considered in finding a prospective mine. The current method of finding prospective mineral deposits in an area involves rudimentary calculations from the available survey data and records which is very time consuming, error-prone considering the huge volume of data along with numerous variables to be evaluated and has very low success rate thereby increasing time and cost involved in the process of exploration. An intelligent system which is capable of providing information(type of mineral and exact location) on prospective mineral deposits in an area through analysing the vast amount of data available can help in revolutionising the way mine explorations are done. The new solution needs to reduce the cycle time of whole exploration, as well as provide new insights with precision.
Profile of the End-User
Digital Officer: Analysing data of geochemical, geological, magnetic surveys and record of successful explorations in the past and provide insights on the probability of finding mineral deposits in a given area.
Current Method : Presentations showing data collected and basic rudimentary data analysis using excel sheets.
Gaps : Complex data based on different surveys with each kind of survey having respective parameters are analysed using basic data storage and management systems such as excel sheets, to give an abstraction to each survey and parameters, based on the abstracted data decisions are taken without cohesion and relation between parameters which is leading to errors, increase in time consumption and often has very low success rate.
Functional Requirements of the End-User
- Analyse data available from surveys and past explorations.
- Provide insights on prospective mineral deposits in an area.
Functional & Operational Capabilities
- Analyse different data of geological, geochemical, magnetic surveys and exploration history
- Provide insights on prospective mineral deposits in an area from analysing the available data
- Incorporate exploratory geology approach
- Output interpretable with suitable geographic and location visualisation.
- Data Security
- Availability of Clean data for each parameter to be analysed
- Import of data from logbooks and other legacy storage systems.
Expected Tangible Benefits and Measurable Gains
- Lowering of cost involved in exploration for mines
- Reduction in the effort of data analysis with precise insights
- Reduction in cycle time of the whole exploration process
- Better insights and accurate forecasting based on the available data sets.
Performance Metrics or Outcomes
- Target probability vs Actual result (accuracy)
- Success rate on prospecting mineral deposits