Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Genetic algorithm with genetic engineering technology for multi-objective dynamic job shop scheduling problems

: Dimitrov, T.; Baumann, M.

Postprint urn:nbn:de:0011-n-1782857 (470 KByte PDF)
MD5 Fingerprint: 40e02834edefeaa28d19e311458c14ea
© ACM 2011 This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution.
Erstellt am: 6.9.2011

Krasnogor, N. ; Association for Computing Machinery -ACM-; Association for Computing Machinery -ACM-, Special Interest Group on Genetic and Evolutionary Computation:
GECCO 2011, 13th Annual Conference Companion on Genetic and Evolutionary Computation. Proceedings : Ireland, Dublin, July 12 - 16, 2011
New York: ACM, 2011
ISBN: 978-1-4503-0690-4
Annual Conference Companion on Genetic and Evolutionary Computation (GECCO) <13, 2011, Dublin>
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IOSB ()
genetic algorithm; genetic engineering; managed; scheduling; job shop; incremental evaluation; changeover; tardiness

Genetic algorithms were intensively investigated in various modifications and in combinations with other algorithms for solving the NP-hard scheduling problem. This extended abstract describes a genetic algorithm approach for solving large job shop problems that uses hints from the schedule evaluation in the genetic operators. The result is a hybrid genetic algorithm with smaller randomness and more managed search to find better solutions in shorter processing time. The hybridized genetic algorithm was tested with data from wafer production with thousands of jobs and hundreds of machine alternatives. The hybridized genetic algorithm not only achieved smaller tardiness in shorter computation time but was also able to reduce the sequence dependent change-over times between jobs in comparison with the classical genetic algorithm.