A multi-step approach to long-term open-pit production planning

The objective of this paper is to develop, verify, and present a multi–step methodology for three interrelated key components of open–pit mine planning: controlled optimal phase–design, characterisation of selective mining–units, and long–term production scheduling optimisation. A hybrid solution methodology for open–pit phase–design using integer programming and a local search heuristic is presented. Next, a hierarchical clustering approach with size and shape control, which aggregates blocks into minable polygons constrained within the pushback boundaries, is presented; and finally, a mixed integer linear programming mathematical model, which uses the generated pushbacks and aggregates as the planning units to provide near–optimal practical life–of–mine schedules, is introduced. In addition, the model inherently solves the cut–off grade optimisation problem. Two case–studies of real–size deposits are presented to illustrate practicality of the developed methodologies, and also to compare the results against industrial conventional practices to assess validity, performance, strengths, and limitations of the developed methodologies.