Genetik algoritmalar ve tesis düzenlemesi probleminde kullanımı: Teori ve metodoloji (Kısım 1)

Genetik algoritmalar (GA), biyolojik sistemlerde Darwinian doğal seçim ve mutasyonlarına doğrudan analojiyi baz alan bir optimizasyon metodolojisidir. Yine GA, bireylerin belirlenen uyum değerleri ile orantılı olarak oluşturdukları bir nüfus içinde rahat etmelerine dayalı bir seçim mekanizmasını kullanan iterasyonel bir metottur. GA; çizelgeleme, seyyar satıcı problemi ve kuadratik atama problemi (QAP) gibi kombinatorial sayılı problemlerin sezgisel olarak çözümüne uygulanmaktadır. Bu çalışmada önce genetik algoritmalar; temel yapı, avantajları, kodlama ve kullanılan parametreler sistematik olarak incelenmiştir. Daha sonra QAP'nin çözümleme zamanını iyileştirmek için "genetik algoritmalar ile yerel arama (YA) teknikleri" beraber kullanılarak oluşturulan "kısmi aramalı ve tam yerel aramalı melez genetik algoritmalar" tanıtılmıştır. Bu çalışmanın amacı; genetik algoritma kavramını tanıtmak ve bu konuda yapılacak çalışmalara yardımcı olabilmek için hem metodolojik bir çerçeve hem de bir bibliyografya sunmaktır. Bu çalışmanın ikinci kısmında ise, bu alanda oldukça yeni bir konu olan genetik algoritmaların tesis yerleşim probleminde (TYP) nasıl kullanılabileceği konusunda metodolojik bir çerçeve sunulmaya çalışılmıştır.

The genetic algorithms and theirs using in the facilities layout problem: Theory and methodology (Part I)

Genetic Algorithms (GAs) are optimization methodologies which take Darwinian natural selection and mutations of biological systems analogy as a base directly. Again GA are iterational methods using a selection mechanism that depends on the aim of providing comfort for the individuals making up a population that is rational to defined adaptation values. GA are applied heuristically to solutions of combinatorial numbered problems such as graphing, salesman and Quadratic Assignment Problems (QAPs). In this study, first GAs, their basic structure, coding, advantages and used parameters are examined systematically. Then, in order to improve the solution time of QAP "partial local search and full local search hybrid GAs", which are formed by using "GAs and local search techniques" together, are introduced. The aim of this study is to present both a methologic frame and a bibliography in order to introduce GA concept and to provide help for further studies on this subject. In the second part of this study, a méthodologie frame on the subject of how to use GAs in the facility layout problem, which is a further new issue, is tried to be presented.

___

1.Ahuja, R.K., Orlin, F.B., "Developing fitter genetic algorithms", INFORMS Journal on Computing, 9 (3), 251-253, (1997).
2.Back, T., "Optimal mutation rates in genetic search", Proceedings of the 5th I. Conf. on Genetic Algorithms, 2-8, University of Dortmund, (1995).
3.Back, T., Evolutionary Algorithms in Theory and Practice, Oxford U. Press, N.Y., (1996).
4.Behe, M., Darwin's Black Box: The Biochemical challenge to evaluation, Free Press,N.Y.,(1996).
5.Bertoni, A., Campadelli, P., Carpentieri, M. ve Grossi, G, "Analysis of a genetic model", Proceedings of the 7th I. Conf. on Genetic Algorithms, 121-126, University of Milano, Italy, (1997).
6.Çela, E., "Combinatorial optimization: The Quadratic Assignment Problem, Theory and Algorithms", Kluwer Academic Publishers, Netherlands, (1998V
7.Davis, L., Handbook of genetic algorithms, Van Nostrand Reinhold: N. Y., (1991).
8.Downsland, K.A., Genetic algorithms a tool for OR?, The J, of Oper. Res. Society, 47 550-561,(1996).
9.Goldberg, D.E., Genetic algorithms in search, optimization and machine learning,Addison-Wesley Publishing Company, Reading, Massachusetts, (1989).
10.Grefenstette, J., Optimization of control parameters for genetic algorithms. IEEE Transactions on Systems, Man. and Cybernetics, SMC. 16, (1), 122-128, (1987).
11.Holland, J. H, Adaptation in natural and artificicial systems, Ann Arbor, MI: The University of Michigan Press, (1975).
12.Islier, A.A., "The design of manufacturing cell using genetic algorithm", 1st System Engineering and Defence Applications Symposium, The Turkish Military Academy, (1995).
13.Kershenbaum, A., "When genetic algorithms work best", INFORMS Journal on Computing, 9 (3), 254-255, (1997).
14.Levine, D., "Genetic algorithms: a practitioner's view", INFORMS Journal on Computing, 9 (3), 256-259, (1997).
15.Liepins, G.E., Wang, L.A., "Classifier system learning of Boolean concepts", Proceedings of the 4th I. Conf. on Genetic Algorithms, 318-323, San Diego, CA, (1991).
16.Liepins, G.E., Hillard, M.R., Richardson, J., Palmer, M., "Genetic algorithms applications to Set Covering and Travelling Salesman Problems" (Operations Research and Artificial Intelligence: The integration of problem-solving strategies), Kluwer Academic Publishers, Boston, 29-57, (1990).
17.Lim, M.H., Yuan, Y., Omatu, S., "Efficient genetic algorithms using simple genes exchange local search policy for the Quadratic Assignment Problem",Computational Optimization and Applications, 15, 249-268, (2000).
18.Merz, P., Freisleben, B., "A genetic local search approach to the Quadratic Assignment Problem", Proceedings of the 7th I. Conf. on Genetic Algorithms, 465-472, (1997).
19.Reeves, C.R., "Genetic algorithms: No panacea but a valuable tool for operations researcher", INFORMS Journal on Computing, 9 (3), 263-265, (1997).
20.Ross, P., "What are genetic algorithms good at?", INFORMS Journal on Computing, 9 (3), 260-262, (1997).
21.Shaffer, J.D., "Practical guide to genetic algorithms", VRL:http://chemdiv-www.nrl.navy.mil/6110/sensors/chemometrics/practga.html, (2001).
22.Suresh, G., Vinod, V., Sahu, S., "A genetic algorithm for facility layout", I. J, of Prod.Res., 33 (12), 3411-3423, (1995).
23.Tarn, K.Y., "Genetic algorithms, function optimization and facility layout design",European J. of Oper. Res., 63,322-346, (1992a).
24.Vignaux, G.A.,"Genetic algorithms in lecture notes", URL: http: //www. mcs. vuw. ac. nz/courses/OPRE457/Lecture-notes/genetic/Iectures/genetic. html, (2000).