The job-shop scheduling problem was considered with sequence-dependent setup times and preventive maintenance constraints. Processing times were revealed when products arrived at a machine. Unknown processing times would give a more real-world representation where exact processing times weren’t available.
A hybrid model combining discrete-event simulation and an optimization algorithm in Python was applied to simulate the production process and solve the job-shop problem. The hybrid model used optimization by creating new production schedules when a job was processed and when the product arrived at the next machine. The meta-heuristics of the genetic algorithm, ant colony optimization, and simulated annealing algorithm were used.
The output of the hybrid optimization model made by connecting AnyLogic simulation and a Python optimization algorithm through the Pipeline library connector showed significantly better results than random sequencing of jobs.
The results of the meta-heuristic comparison, using r = 5