Discrete Event Simulation of Mine Equipment Systems Combined with Maintenance Analysis Using Genetic Algorithms
CIM Montreal 2003
Greg Yuriy, Nick Vagenas, Tony Nuziale,
This paper discusses research carried out by the Laurentian University Mining Automation Laboratory (LUMAL) in the field of discrete-event simulation of mine equipment systems and reliability assessment of machinery using genetic algorithms. The AutoMod simulation package, from Brooks-PRI Automation Inc., is used in this research. It is a 3-D discrete-event simulation/animation software tool for creating realistic depictions of manufacturing and materials handling systems.
Discrete-event simulation involves modeling of real-life systems based on variables and events which occur or change at distinct points in time. In our case, a two level, multi-drift sublevel stoping model was created in AutoMod for the simulation of the equipment fleet employed in the development cycle (drilling, explosives loading, blasting, hauling, ground support) of a typical underground mine in the nickel mining region of Sudbury in Ontario, Canada. The equipment primarily considered in the simulation is Load-Haul-Dump (LHD) vehicles, drills, explosives-loaders and roof bolters.
The purpose of the simulation study is to assess the impact of equipment failures to mine’s production throughput. In more detail, using Mean Time Between Failures (MTBF) data of scooptrams derived from a reliability assessment model based on genetic algorithms, LHD failures are incorporated into the simulation model. These failures cause production interruptions to the development cycle and consequently loss of hauling capacity.
This paper will indicate the significance of equipment failures to mine’s production and suggest a methodology based on the combination of discrete-event simulation and a predictive reliability assessment model using genetic algorithms for the study and analysis of mine equipment systems.
Simulation, Mining equipment, Reliability