Creating robust solutions by means of evolutionary algorithms

  • Jürgen Branke
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1498)


For real world problems it is often not sufficient to find solutions of high quality, but the solutions should also be robust. By robust we mean that the quality of the solution does not falter completely when a slight change of the environment occurs, or that certain deviations from the solution should be tolerated without a total loss of quality.

In this paper, a number of modifications to the standard evolutionary algorithm (EA) are suggested that are supposed to lead the EA to produce more robust solutions. Some preliminary experiments are reported where the proposed approaches are compared to a standard model. As it turns out, the EA's ability to create robust solutions can be greatly enhanced even without additional function evaluations.


evolutionary algorithm robust solution 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    A. N. Aizawa and B. W. Wah. Scheduling of genetic algorithms in a noisy environment. Evolutionary Computation, pages 97–122, 1994.Google Scholar
  2. 2.
    J. Michael Fitzpatrick and John J. Greffenstette. Genetic algorithms in noisy environments. Machine Learning, 3:101–120, 1988.Google Scholar
  3. 3.
    D. E. Goldberg. Genetic Algorithms. Addison-Wesley, 1989.Google Scholar
  4. 4.
    U. Hammel and T. Bäck. Evolution strategies on noisy functions, how to improve convergence properties. In Y. Davidor, H. P. Schwefel, and R. Männer, editors, Parallel Problem Solving from Nature, number 866 in LNCS. Springer, 1994.Google Scholar
  5. 5.
    U. Kohlmorgen, H. Schmeck, and K. Haase. Experiences with fine-grained parallel genetic algorithms. Annals of Operations Research, to appear.Google Scholar
  6. 6.
    M. McIlhagga, P. Husbands, and R. Ives. A comparison of search techniques on a wing-box optimisation problem. In H.-M. Voigt, editor, Parallel Problem Solving from Nature 4, number 1141 in LNCS, pages 614–623. Springer Verlag, 1996.Google Scholar
  7. 7.
    Z. Michalewicz. Genetic Algorithms + Data Structures = Evolution Programs. Springer Verlag, 3rd edition, 1996.Google Scholar
  8. 8.
    I. C. Parmee. Cluster-oriented genetic algorithms for the identification of highperformance regions of design spaces. In EvCA96, 1996.Google Scholar
  9. 9.
    C. R. Reeves. A genetic algorithm approach to stochastic flowshop sequencing. In IEE Colloquium on Genetic Algorithms for Control and Systems Engineering, number 1992/106 in Digest, pages 13/1–13/4. IEE, London, 1992.Google Scholar
  10. 10.
    R. Roy, I. C. Parmee, and G. Purchase. Integrating the genetic algorithm with the preliminary design of gas turbine blade cooling systems. In ACEDC'96, 1996.Google Scholar
  11. 11.
    A.V. Sebald and D.B. Fogel. Design of fault tolerant neural networks for pattern classification. In D.B. Fogel and W. Atmar, editors, 1st Annual Conf. on Evolutionary Programming, pages 90–99, 1992.Google Scholar
  12. 12.
    S. Tsutsui and A. Ghosh. Genetic algorithms with a robust solution searching scheme. IEEE Transactions on Evolutionary Computation, 1(3):201–208, 1997.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Jürgen Branke
    • 1
  1. 1.Institute AIFBUniversity of KarlsruheKarlsruheGermany

Personalised recommendations