ÇOK AMAÇLI GENETİK ALGORİTMA YÖNTEMLERİNİN BAŞARIMININ BELİRLENMESİ İÇİN İKİ YENİ ÖLÇÜT ÖNERİSİ

Öz

Genetik algoritmalar, çok amaçlı optimizasyon problemlerinin çözümünde kullanılan etkili yöntemlerdir. Çok amaçlı optimizasyon problemlerinin doğası gereği, bu problemleri çözebilecek birçok çok amaçlı genetik algoritma (ÇAGA) yöntemi önerilmiştir. Bu yöntemlerin, optimizasyon problemlerini ne kadar iyi çözdüğünün belirlenmesi için literatürde birçok başarım ölçütü önerilmiştir. Bu çalışmada, ÇAGA yöntemlerinin sıralama (puan atama) yeteneklerinin ölçülmesi için Ceza ve Ödül başarım ölçütleri önerilmektedir. Bu iki ölçüt ile bir ÇAGA yöntemi tarafından seçme mekanizmasına ne kadar nitelikli bilgi aktarıldığı sezgisel ve istatistiksel olarak tespit edilebilmektedir. Literatürde çok kullanılan SPEA yöntemi ile yeni önerilmiş DOPGA yöntemi, 4 farklı test fonksiyonu üzerinde çalıştırılmış ve sonuçlar Ceza ve Ödül ölçütleri kullanılarak değerlendirilmiştir.

Anahtar Kelimeler

• Genetik Algoritma, Optimizasyon, Başarım Ölçütü, Ceza, Ödül

PROPOSAL OF TWO NEW METRICS TO DETERMINE THE PERFORMANCE OF MULTI-OBJECTIVE EVOLUTIONARY ALGORITHMS

Abstract

Genetic algorithms are effective methods to solve the multi-objective optimization problems. Due to the nature of multi-objective optimization problems, a lot of multi-objective evolutionary algorithms (MOEAs) are proposed to solve these problems. In literature, a lot of performance metrics are proposed for determining the performance of MOEAs. In this paper, Punishment and Reward metrics are proposed to measure fitness assignment capabilities of MOEAs. With the help of two proposed metrics, how much useful information can be generated and passed into the selection mechanism by MOEA methods can now be determined heuristically and statistically. The state of the art SPEA and newly proposed DOPGA methods are tested on four test functions and the results are evaluated by using Punishment and Reward metrics

Keywords

• Genetic Algorithms,Optimization,Performance Metric,Punishment,Reward

References

1. Coello, C. A. C., Lamont, G. B., Veldhuizen, D. A., (2007). Evolutionary Algorithms For Solving Multi-Objective Problems, Second Edition. Springer, New York, U.S.A.
2. Deb, K., (2001). Multiobjective Optimization Using Evolutionary Algorithms, Wiley, Chichester, U.K.
3. Coello Coello, C. A., (1999). A Comprehensive Survey of Evolutionary-Based Multiobjective Optimization Techniques, Knowledge and Information Systems, 1(3), pp.269-308.
4. Ghosh A., Dehuri S., (2004). Evolutionary Algorithms for Multi-Criterion Optimization: A Survey, International Journal of Computing & Information Sciences, 2(1).
5. Schott, J. R., (1995). Fault Tolerant Design Using Single and Multi-Criteria Genetic Algorithms, Master Thesis, Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Boston, USA.
6. Van Veldhuizen, D.A., (1999). Multiobjective Evolutionary Algorithms: Classifications, Analyses, and New Innovations, PhD. Thesis, Air Force Institute of Technology, Dayton.
7. Zitzler, E., (1999). Evolutionary Algorithms for Multiobjective Optimization: Methods and Applications, Ph.D Thesis, Swiss Federal Institute of Technology, Switzerland.
8. Zitzler, E., Thiele, L., (1999). Multiobjective Evolutionary Algorithms: A Comparative Case Study and the Strength Pareto Approach, IEEE Transactions on Evolutionary Computation, 3(4), 257-271.

9. Zitzler, E., Thiele, L., Laumans, M., Fonseca, C. M., Fonseca, V. G., (2003). Performance Assesments of Multiobjective Optimizers: An Analysis and Review, IEEE Transactions on Evolutionary Computation, 7(2), 117-132.
10. Deb, K., Pratap, A., Agarwal., S., Meyarivan, T., (2002). A Fast and Elitist Multi- objective Genetic Algorithm: NSGA-II, IEEE Transactions on Evolutionary Computation, 6,182-197.
11. Van Veldhuizen, D.A., and Lamont G.B., (2000). Multiobjective Evolutionary Algorithms: Analyzing the State-of-the Art, Evolutionary Computation 7(3), 1-26.
12. Knowles J., Thiele L., and Zitzler E., (2006). A Tutorial on the Performance Assesment of Stochastic Multiobjective Optimizers, 214, Computer Engineering and Networks Laboratory (TIK), ETH Zurich, Switzerland, Revised Version.
13. Knowles, J.D., Corne, D.W., (2002). On Metrics for Comparing Nondominated Sets, Proceedings of the 2002 Congress on Evolutionary Computation, CEC '02, 1, 711-716.
14. Ergul, E.U., Eminoglu, I., (2014). DOPGA: A New Fitness Assignment Scheme for Multi-objective Evolutionary Algorithms, International Journal of Systems Science, 45(3), 407-426.
15. Fonseca, C.M., Fleming, P.J., (1993). Genetic Algorithms for Multi-objective Optimization: Formulation, Discussion and Generalization, In Proceedings of the Fifth International Conference on Genetic Algorithms, 416-423.
16. Chiu, S., (1994). Fuzzy Model Identification Based on Cluster Estimation, Journal of Intelligent & Fuzzy Systems, 2(3). Huband, S., Hingston, P., Barone, L., While, L., (2006). A Review of Multiobjective Test Problems and A Scalable Test Problem Toolkit, IEEE Transactions on Evolutionary Computation, 10(5), 1-30.