Scheduling of production soundly affects its capacity especially if system does complex production jobs. In the theoretical part of the article an overview of the scheduling methods proposed in the literature was presented. In this paper it was stated a variant of job shop problem, in which jobs can overlap in some machines and omit others. Authors designed and presented here genetic algorithm to optimize solution of such a problem. The algorithm finds jobs sequence priority and in accordance with it schedules operations and calculates their completion time. An adequate problem was met in an examined plant, where 20 production jobs consisted of 11 to 20 operations assigned to at most 15 machines. Such big parameter numbers are crucial for big formal models and their solution algorithms. The designed algorithm proved to deal with parameters scale, as it found the schedule with 23,8% shorter jobs completion time in comparison with FIFO heuristic, that has been used so far by the plant.

Publication timeframe:
4 times per year
Journal Subjects:
Computer Sciences, Artificial Intelligence, Software Development