The allocation of trucks in open pit mines is a field with great potential for optimizing resources and applying advanced computer modeling techniques, mainly because many companies still choose to use manual allocation, which is premised on the decisions made by the operator, being subject to common failures and not reaching the maximum potential that the equipment can provide. Therefore, this work focuses on optimizing the allocation of trucks in order to increase production, reducing queue time and keeping ore grades within proper limits. The proposed algorithm was based on the differential evolution technique, where two types of mutation operators were used: rand/1/bin and best/1/bin, thus verifying the most suitable to solve the problem. The trucks were allocated in the ore loading and unloading process, aiming to improve the production capacity in a virtual mine. The results brought a convergence to the maximum global production, in addition to which, the allocation of unnecessary transport equipment to the planned routes was avoided. The two mutation operators compared had certain advantages and disadvantages, each better adapting to certain types of situations. The proposed technique can still be extended to other areas, for example, in the transport of grain on the road network or in the implementation of an allocation in freight cars that transport grain.
Sufficient Conditions for Global Convergence of Differential Evolution Algorithm
