4 présentations

• 
  15h30 - 15h55

  A Tabu Search Procedure for Open Pit Mine Scheduling

  • Renaud Senécal, prés., Université McGill
  • Roussos Dimitrakopoulos, COSMO Stochastic Mine Planning Laboratory, Université McGill

  This study presents a metaheuristic solution for the open pit mine scheduling problem with multiple destinations and geological uncertainty. The feasible domain is explored by a Tabu search procedure, which finds the best local solution and uses a diversification strategy to generate new initial solutions. A case study is presented to show the efficiency of the approach.

• 
  15h55 - 16h20

  Stochastic Short-term Mine Production Scheduling Accounting for Fleet Allocation and Operational Considerations

  • Martha E. Villalba Matamoros, prés., McGill University
  • Roussos Dimitrakopoulos, COSMO Stochastic Mine Planning Laboratory, Université McGill

  A stochastic short-term mine production scheduling problem is presented, where the ore body metal content, the fleet availability, and the hauling process are uncertain. We propose a stochastic integer formulation that is solved using the commercial solver CPLEX. Operational considerations, such as mining width and mining directions, are addressed and deliver practically mineable production scheduling patterns.

• 
  16h20 - 16h45

  Long-Term Planning at Raglan Mine (Part 1)

  • Jean Collard, prés., Polytechnique Montréal
  • Rafael Martinelli, GERAD - Polytechnique Montréal
  • Michel Gamache, GERAD - Polytechnique Montréal

  We present a mixed-integer model for long-term planning in a large underground nickel mining complex. The Raglan Mine is located in Nunavik region, the extreme limit of Northern Quebec, Canada. The model incorporates operating characteristics like resource constraints, precedence between activities, nickel cutoff grade, among other. Results will be presented.

• 
  16h45 - 17h10

  Long-Term Planning at Raglan Mine (Part 2)

  • Rafael Martinelli, prés., GERAD - Polytechnique Montréal
  • Jean Collard, Polytechnique Montréal
  • Michel Gamache, GERAD - Polytechnique Montréal

  In this talk, we show heuristic approaches to improve a MIP model resolution for a long-term planning in a large underground nickel mining complex. The heuristics work together with a commercial MIP solver. This technique can lead to high quality solutions fast. We compare our results with some existing ones.