Fertilization Operator for Multi-Modal Dynamic Optimization

  • Khalid Jebari
  • Abdelaziz Bouroumi
  • Aziz Ettouhami
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 229)


Solving Multi-modal Dynamic Optimization problems (MDO) has been a challenge for genetic algorithms (GAs). In this kind of optimization, an algorithm requires not only to find the multiple optimal solutions but also to locate a changing optimum dynamically. To enhance the performance of GAs in MDO, this paper proposes a New Genetic Operator NGO. The NGO is built on three components. First, a novel Genetic Algorithm with Dynamic Niche Sharing (GADNS) which permits to encourage the speciation. Second, an unsupervised fuzzy clustering that tracks multiple optima and enhances GADNS. Third, Spacial Separation (SS) which induces the stable sub-populations and allows local competition. In addition, NGO maintains diversity by a new genetic operators. To control the selection pressure, a new tournament selection is presented. Moving Peaks benchmark is applied to test the performance of NGO. The ability of the NGO to track multiple optima is demonstrated by a new diversity measure.


Dynamic niche sharing Dynamic optimization Evolutionary computation Fuzzy clustering Genetic algorithms Unsupervised learning 


  1. 1.
    Bezdec J (1981) Pattern recognition with fuzzy objective function algorithms. Plenum Press, New YorkCrossRefGoogle Scholar
  2. 2.
    Bouroumi A, Essaïdi A (2000) Unsupervised fuzzy learning and cluster seeking. Intell Data Anal 4(3):241–253Google Scholar
  3. 3.
    Branke J (2002) Evolutionary optimization in dynamic environments. Kluwer Academic, DordrechtMATHCrossRefGoogle Scholar
  4. 4.
    Branke J, Schmidt L, Schmeck H (2000) A multi-population approach to dynamic optimization problems. In: Parmee IC (ed) 4th international conference on adaptive computing in design and manufacture (ACDM 2000). Springer, Berlin, pp 299–308Google Scholar
  5. 5.
    Cedeno W, Vemuri VR (2007) A self-organizing random immigrants genetic algorithm for dynamic optimization problems. Genet Program Evol Mach 8(3):255–286CrossRefGoogle Scholar
  6. 6.
    Cobb HG (1990) An investigation into the use of hypermutation as an adaptive operator in genetic algorithms having continuous, time-dependent non-stationary environments. TIK-report 6760, NLR memorandum, Naval Research Laboratory, Washington, DC, USAGoogle Scholar
  7. 7.
    Deb K, Agrawal RB (1995) Simulated binary crossover for continuous search space. Complex Syst 9(2):115–148MathSciNetMATHGoogle Scholar
  8. 8.
    Deb K, Kumar A (1995) Real-coded genetic algorithms with simulated binary crossover: studies on multimodal and multiobjective problems. Complex Syst 9(6):431–454Google Scholar
  9. 9.
    Goldberg DE, Wang L (1997) Adaptive niching via coevolutionary sharing. TIK-report 97007, Illinois Genetic Algorithms Laboratory, University of Illinois at Urbana-Champaign, 117 Transportation Building, 104 S. Mathews Avenue Urbana, IL 61801Google Scholar
  10. 10.
    Grefenstette JJ (2007) A hybrid immigrants scheme for genetic algorithms in dynamic environments. Int J Autom Comput 4(3):243–254CrossRefGoogle Scholar
  11. 11.
    Jebari K, Bouroumi A, Ettouhami A (2012) New genetic operator for dynamic optimization. In: Lecture notes in engineering and computer science: proceedings of The world congress on engineering 2012, WCE 2012, London, UK, 4–6 July 2012, pp 742–747.
  12. 12.
    Jebari K, Bouroumi A et al (2011) Unsupervised fuzzy tournament selection. Appl Math Sci 28(1):2863–2881MathSciNetGoogle Scholar
  13. 13.
    Kuncheva LI, Bezdec JC (1998) Nearest prototype classification: clustering, genetic algorithms or random search. IEEE Trans Syst Man Cybernet 28(1):160–164CrossRefGoogle Scholar
  14. 14.
    Louis SJ, Xu Z (2008) An immigrants scheme based on environmental information for genetic algorithms in changing environments. In: The 2008 IEEE congress on evolutionary computation, Hong Kong. IEEE, pp 1141–1147Google Scholar
  15. 15.
    Miller BL, Shaw MJ (1995) Genetic algorithms with dynamic niche sharing for multimodal function optimization. TIK-report 95010, Illinois Genetic Algorithms Laboratory, University of Illinois at Urbana-Champaign, 117 Transportation Building, 104 S. Mathews Avenue, Urbana, IL 61801Google Scholar
  16. 16.
    Saäreni B, Krähenbühl L (1998) Fitness sharing and niching methods revisited. IEEE Trans Evol Comput 2(3):97–106CrossRefGoogle Scholar
  17. 17.
    Yang S, Yao S (2007) Population-based incremental learning with associative memory for dynamic environments. IEEE Trans Evol Comput 12(5):542–561MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2013

Authors and Affiliations

  • Khalid Jebari
    • 1
  • Abdelaziz Bouroumi
    • 2
  • Aziz Ettouhami
    • 1
  1. 1.Conception and Systems laboratory, Faculty of SciencesUniversity (UM5A)RabatMorocco
  2. 2.Faculty of sciences Ben MsikHassan II Mohammadia-Casablanca UniversityCasablancaMorocco

Personalised recommendations