Parallel Simulated Annealing Algorithm for Cyclic Flexible Job Shop Scheduling Problem
This paper deals with scheduling of tasks in cyclic flexible job shop scheduling problem (CFJSSP). We have proposed a new method of computing cyclic time for CFJSSP. This method is based on the known properties of the job shop problem as well as new properties of cyclic scheduling. We have developed two versions of proposed method: sequential and parallel. The parallel version is dedicated to the computing devices supporting vector processing. Finally, we have developed double paralyzed simulated annealing algorithms: fine grained - vector processing, multiple walk - multi core processing. Computation results, provided on market multicore processors, are presented for a set of benchmark instances from the literature.
Unable to display preview. Download preview PDF.
- 1.Barnes, J.W., Chambers, J.B.: Flexible Job Shop Scheduling by tabu search, Graduate program in operations research and industrial engineering, Technical Report ORP 9609, University of Texas, Austin (1996)Google Scholar
- 3.Bożejko, W., Uchroński, M., Wodecki, M.: The new golf neighborhood for the flexible job shop problem. In: Proceedings of the ICCS 2010. Procedia Computer Science, vol. 1, pp. 289–296. Elsevier (2010)Google Scholar
- 12.Kampmeyer, T.: Cyclic Scheduling Problems, Ph. D. Thesis, University Osnabrück (2006)Google Scholar