The use of mixed integer linear programming for long-term scheduling in block caving mines

Caving is a mass mining method that is normally applied to large, low grade orebodies because of its low production cost and high capacity. However, equipment breakdown and poor draw management can cause mining costs to exceed that of all other mining methods. One way to minimize the effect of disruptions in the mining cycle is to use Mixed Integer Linear Programming (MILP) for long-term scheduling. This paper describes a MILP goal program which has the dual objectives of minimizing deviation from the ideal draw profile while achieving a production target. Schedule optimization is performed using a life of mine approach in which all production periods are optimized simultaneously. The underlying assumption being that material mixing in the short-term (between monthly optimization cycles) has a minimal effect on the long-term state of the panel. Schedule optimization is carried out using six general constraint types to model and control production. The constraint types define: deviation from ideal practice, panel state, material flow conservation, production quality, material flow capacity and production control. These constraints define production targets and drive production toward the ideal profile while maintaining correct mining practice. Application of the model using production scheduling data for a De Beers kimberlite mine. It is shown how different production control constraints regulate production from individual drawpoints as well as recovery of the ideal panel profile by implementing an optimised draw schedule.