A new MIP model for mining equipment scheduling under variable scenarios

Abstract

Mining is a capital-intensive industry that requires hundreds of million-dollar investment in large equipment. Mine trucks are the most important mining equipment for the material haulage. However, their maintenance cost accounts for large proportion of the overall mining operational cost. An overview of the current studies reveals that although the truck scheduling issue has been researched for decades, there is no method that can take into account all major constraints with the objective of obtaining the minimised maintenance cost.This research develops a mixed integer programming (MIP) model to optimise the heterogeneous truck fleet schedule that can minimise the maintenance cost in conjunction with satisfying the material movement schedule and considering a new truck purchase option. A hypothetical data set is implemented to validate the proposed MIP model. The validation results have proved that all formulated constraints are working correctly.The proposed MIP model is applied to a real data set from a gold mine located in Western Australia. The result is compared with that of the traditional spreadsheet-based method and original MIP model using the same data set, which indicates that the proposed MIP model can provide 21.94% and 14.77% maintenance cost savings over the other two methods throughout a 10-year period of time respectively. Since the conditions in the real mining operation have uncertainty, the input variables of the proposed MIP model for truck scheduling are dynamic. Hence, the applications of the proposed MIP model with “what if” scenario-based variables, such as age bin size, material movement requirement and truck purchase cost, are conducted in this research. Their performance and solutions are demonstrated in this thesis.An interface of the proposed MIP model is developed, which simplifies the process of dealing with a mass of data and constructing the MIP model. With the assistance of the interface, the truck scheduling time is decreased considerably. In addition, a manual of the interface is stated in the thesis.