Incluant une Journée industrielle de l'optimisation

HEC Montréal, 7 - 9 mai 2012


HEC Montréal, 7 — 9 mai 2012

Horaire Auteurs Mon horaire

TB4 Applications liées à la foresterie 1 / Applications in Forestry 1

8 mai 2012 10h30 – 12h10

Salle: Quebecor

Présidée par Bernard Gendron

3 présentations

  • 10h30 - 10h55

    A Column Generation Approach for Demand-Driven Harvest Scheduling

    • Géraldine Gemieux, prés., Université de Montréal
    • Bernard Gendron, Université de Montréal, CIRRELT
    • Jacques Ferland, Université de Montreal

    We consider the problem of assigning to each harvest team an annual schedule to satisfy the demands at the mills, while minimizing the costs associated to each activity along the value chain. A MIP model and a heuristic based on column generation have been used, where columns represent harvest schedules. We present computational results in the context of eastern Canadian forests.

  • 10h55 - 11h20

    Solving a Synchronized Log-Truck Scheduling Problem with Column Generation

    • Greg Rix, prés., Polytechnique Montréal
    • Louis-Martin Rousseau, Polytechnique Montréal
    • Gilles Pesant, Polytechnique Montréal

    We present a synchronized routing and scheduling problem that arises in the forestry industry. We allocate harvested volumes to mills, determine inter-period storage, and construct log-truck routes to deliver the harvest while synchronizing the trucks with log-loaders. A column generation methodology is proposed, and results given on several case studies.

  • 11h20 - 11h45

    Modeling and Solving a Complex Dynamic Facility Location Problem

    • Sanjay Dominik Jena, prés., Université du Québec à Montréal
    • Bernard Gendron, Université de Montréal, CIRRELT
    • Jean-François Cordeau, HEC Montréal, GERAD, CIRRELT

    We study a complex extension of a multi-period multi-commodity facility location problem appearing in the Forestry Industry. A set of valid inequalities and heuristic starting solutions significantly improve the solutions found by CPLEX. Two simplified versions of the problem are efficiently solved by Lagrangean Relaxation, often outperforming CPLEX.