Smart equipment allocation decisions and updating short-term stochastic production planning in mining complexes through reinforcement learning APCOM 2021

This paper presents an actor-critic reinforcement learning (RL) method to allocate trucks and shovels in open-pit mines, while defining short-term production scheduling of an industrial mining complex, given new information collected during operations. This new data obtained from sensors update the uncertainty of the orebody models. The RL agents interact with the mining complex environment; exploring the solution space, learning actions to improve metal production and equipment utilisation. A case study is presented at a copper mining complex, highlighting the method’s ability to adapt and make informed decisions while collecting new information.