AI for ore sorting and dispatch decides where every rock and every truckload goes. Sensors measure grade and mineralogy on the rock itself. Classification models then sort particles, and dispatch software routes trucks based on live data rather than a fixed plan. This category covers the sensing hardware, the sorting technologies built on it and the dispatch logic that turns graded material into a coherent plant feed.
What AI Does in Ore Sorting and Dispatch
Material moves fast between the pit and the plant, so decisions need to happen in seconds. AI helps in three ways. It senses grade and mineralogy with XRT, SWIR and belt-mounted analyzers. It classifies each particle or batch as ore or waste. Finally, it routes material by tuning dispatch rules as trucks, queues and roads change.
Where to Start
Each page below answers one “how do I do this with AI?” question.
- Sort the rock. AI ore sorting pre-concentrates ore by rejecting waste before the mill, using TOMRA and Steinert sorters.
- Pick the sensors. Choose ore sorting sensors by what physically separates your ore from waste, then fuse them with AI.
- Read the mineralogy. SWIR mineral analysis identifies clay and alteration minerals for geomet models and sorting logic.
- Check the belt. A conveyor belt analyzer such as MineSense BeltSense catches dilution before it reaches the mill.
- Route the trucks. Mine dispatch optimization adds an AI tuning layer to fleet systems like DISPATCH and Wenco FMS.
What’s Mature and What’s Emerging
Sensor-based ore sorting is mature, and TOMRA and Steinert hold over 40% of that market together. Truck dispatch systems have run for decades, and AI tuning add-ons are now commercially available. SWIR scanning and belt analyzers are established products as well. By contrast, reinforcement learning for dispatch is still mostly research. Multi-sensor fusion is also still evolving, so test on your own ore before trusting vendor claims.