Multi-level modeling and optimization of forest transport.

Summary This thesis is a contribution to the modeling, planning and optimization of transport for the supply of forest wood to primary processing industries. In this field, the climatic hazards (grounding of wood by storms), sanitary hazards (bacteriological and fungal attacks on wood) and commercial hazards (variability and increasing demand of the markets) push the various actors of the sector (forestry contractors and operators, transporters) to review the organization of the logistic supply chain, in order to improve the quality of service (adequacy of supply-demand) and to decrease the costs.The main objective of this thesis was to propose a steering model to improve the performance of forest transport, while respecting the constraints and practices of the sector.The results establish a hierarchical planning approach for transport activities at two decision levels, tactical and operational. At the tactical level, a multi-period optimization allows to respond to orders by minimizing the global transport activity, under the constraint of the aggregate capacity of the accessible means of transport. This level allows the implementation of load smoothing policies and the organization of subcontracting or partnerships between transport actors. At the operational level, the tactical plans allocated to each carrier are disaggregated, to allow an optimization of the fleets' routes, under the constraint of the physical capacities of these fleets. The optimization models of each level are formalized in mixed linear programming with binary variables. The applicability of the models was tested using an industrial dataset in the Aquitaine region and showed significant improvements in the exploitation of transport capacities compared to current practices.The decision models were designed to adapt to any organizational context, whether partnership or not: the production of the tactical plan has a generic character without presumption of the organization, this one being taken into account, in a second time, at the level of the operational optimization of the transport plan of each actor.