Lower bounds and multiobjective evolutionary optimization for combined maintenance and production scheduling in job shop

We study in this paper the combined maintenance and production scheduling. Such problem is traditionally treated independently specially for the multi product environment (like the job shop). We look for optimizing jointly two criteria, one for production which is the makespan, and one for maintenance, which is the total cost. These two criteria are antagonistic, that is why we develop a multiobjective optimization and genetic algorithm based method that determines simultaneously maintenance and production schedules for a job shop. The genetic algorithm uses a Pareto optimal selection keeping only the more adapted solutions. The Pareto optimal solutions given by the genetic algorithm are validated using some lower bounds. For the purpose of reducing the frequency of breakdowns, we apply the systematic preventive maintenance to the various resources of the workshop. We compare in this paper two manners of choosing periods of maintenance: the first consists in maintaining machines each a fix period of time. The second applies maintenance according to a fix work load of each machine.