Mechanical Workshop Scheduling using Genetic Algorithms

The goal of this paper is to develop an algorithm which allows obtaining an optimal operation schedule for some parts and machines with a low computational time.

Present algorithms for scheduling like Lomnicki or Johnson only are capable to schedule three machines and n parts at maximum, being so quite limited in its use.

Since as at the moment no effective deterministic algorithm has been developed for parts and m machines, only is possible to obtain schedules by means of heuristic techniques. These techniques do not guarantee to obtain the global optimum, although they provide reasonably good local optimums.

From this techniques, genetic algorithms have been chosen, which provide local optimums in a reasonably low time and are also capable to obtain global optimum if enough time is given.