Decision Making applied to Shift Change in Stochastic Open-pit Mining Truck Dispatching

Material transportation is one of the most important aspects of open-pit mine operations. The problem usually involves a truck dispatching system in which decisions on truck assignments and destinations are taken in real-time. The dispatch is usually done in a mine with many trucks working together to achieve a common goal, which can be the maximization of transported material in the end of the shift. This task is held in a stochastic environment with a lot of uncertainties, such as, quantity of transported material, size of truck queues formed in shovels and crushers, and involved times in the system. The shift change, which involves time, is a critical issue to be tackled in truck dispatching. It is very hard to preview with certainty when a truck must stop to change the driver, because, in the most of cases, there is time to remain working, but queues and/or distances may imply in delays and overtime pay. Thus, it is very important to define when the shift change must occur, even when it implies in idleness. This work presents a decision making methodology based on Time-dependent Markov Decision Process (TiMDP) to minimize the shift change problem in a stochastic example mining with 15 trucks, 3 shovels, and 1 crusher. The system is modeled and simulated in SimEventsr comparing its performance to the MTCT (Minimization of Truck Cycle Times) heuristic results in the shift change problem.