Elite-initial population for efficient topology optimization using multi-objective genetic algorithms

  • Shin, Hyunjin ;
  • Todoroki, Akira ;
  • Hirano, Yoshiyasu
  • Received : 2013.09.22
  • Accepted : 2013.10.28
  • Published : 2013.12.30


The purpose of this paper is to improve the efficiency of multi-objective topology optimization using a genetic algorithm (GA) with bar-system representation. We proposed a new GA using an elite initial population obtained from a Solid Isotropic Material with Penalization (SIMP) using a weighted sum method. SIMP with a weighted sum method is one of the most established methods using sensitivity analysis. Although the implementation of the SIMP method is straightforward and computationally effective, it may be difficult to find a complete Pareto-optimal set in a multi-objective optimization problem. In this study, to build a more convergent and diverse global Pareto-optimal set and reduce the GA computational cost, some individuals, with similar topology to the local optimum solution obtained from the SIMP using the weighted sum method, were introduced for the initial population of the GA. The proposed method was applied to a structural topology optimization example and the results of the proposed method were compared with those of the traditional method using standard random initialization for the initial population of the GA.


Topology optimization;Multi-objective optimization problem;Genetic algorithm


  1. Bendsoe, M. P. and Kikuchi, N., "Generating optimal topologies in structural design using a homogenization method", Computer Methods in Applied Mechanics and Engineering, Vol. 71, No. 2, 1998, pp. 197-224.
  2. Bendsoe, M. P. and Sigmund, O., Topology Optimization, Theory, Methods and Applications, Springer, Berlin, 2003.
  3. Huang, X., and Xie, Y.M., Evolutionary topology optimization of continuum structures: methods and applications, John Wiley & Sons, Chichester, 2010.
  4. Tada, Y., Seguchi, Y., and Soh, T., "Shape Determination Problem of Structures Considering Multiple Loading Conditions : Externsion of Inverse Variational Shape Determination Method" [in Japanese], Transactions of the Japan Society of Mechanical Engineers. A, Vol. 52, No.476, 1986, pp. 233-239.
  5. Deb, K., Multi-objective Optimization using Evolutionary Algorithms. John Wiley and Son Ltd, New York, 2001.
  6. Yamada, T., Nishiwaki, S., Izui, K., Yoshimura, M., and Takezawa, A., "A Structural Optimization Method Incorporating Level Set Boundary Expressions Based on the Concept of the Phase Field Method.", Journal of Environment and Engineering, Vol. 6, No. 3, 2011, pp. 567-578.
  7. Otomori, M., Yamada, T., Izui, K., and Nishiwaki S., "Topology Optimization Based on the Level Set Method Using Mathematical Programming" [in Japanese], Transactions of Japan Society of Mechanical Engineers Series C, Vol. 77, No. 783, 2011, pp. 4001-0414.
  8. Wang, S. Y. and Tai, K., "Bar-system representation method for topology optimization using the genetic algorithms", Engineering Computations, Vol. 22, No. 2, 2005, pp.206-231.
  9. Deb, K., Agrawal, S., Pratap, and A., Meyarivan, T., "A fast and elitist multi-objective genetic algorithm for multiobjective optimization: NSGA-II", in: Proceedings of the Parallel Problem Solving from Nature VI Conference, Paris, 2000, pp.849-858.
  10. Goldberg, D., Genetic Algorithms in Search, Optimization, and Machine Learning, Addison Wesley Longman, 1989
  11. Deb, K., Anand, A., and Joshi, D., "A computationally efficient evolutionary algorithm for real-parameter optimization", Evolutionary Computation, Vol. 10, No. 4, 2002, pp.371-395
  12. Deb, K. and Goyal, M., "A combined genetic adaptive search (GeneAS) for engineering design", Computer Science and Informatics, Vol. 26, No. 4, 1996, pp.30-45.

Cited by

  1. A meta-heuristic for topology optimization using probabilistic learning vol.48, pp.11, 2018,
  2. Provisional-Ideal-Point-Based Multi-objective Optimization Method for Drone Delivery Problem vol.19, pp.1, 2018,