This thesis deals with the development of a flow scheduling/optimization approach applied to the field of high-variety, low-volume production systems called HVLV (HighVariety, Low-Volume) systems. In this context, the flow of parts is represented by a discreet flow model. The discontinuous behavior of HVLV systems can be characterized by the knowledge of the starting and ending times of its activities. (Max, +) algebra is used to represent these kinds of systems where relationships between the starting times of the activities require both the maximum and addition operators. In order to use (max, +) algebra for HVLV systems scheduling, it is necessary to solve into this algebra an optimization problem subject to conflicts and constraints. In this research, we have first of all developed a scheduling (max, +) model for HVLV systems where decision variables are introduced to solve the conflict problem between operations carried out on the machines. Then, we have improved the proposed model to deal with preventive maintenance. Two kinds of maintenance are considered: Repetitive Periodic Maintenance (RPM) and Flexible Periodic Maintenance (FPM). In both cases, a non-linear optimization problem with constraints is solved to minimize some performance criteria. Lastly, simulation results on some complex HVLV job-shop systems are presented to illustrate the feasibility of the proposed methodology..
