Optimal design of cantilever soldier pile retaining walls embedded in frictional soils with harmony search algorithm

Creative Commons License

Bekdas G. , Arama Z. , Kayabekir A. E. , Geem Z. W.

Applied Sciences (Switzerland), vol.10, no.9, 2020 (Journal Indexed in SCI Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 10 Issue: 9
  • Publication Date: 2020
  • Doi Number: 10.3390/app10093232
  • Title of Journal : Applied Sciences (Switzerland)
  • Keywords: cantilever soldier piles, embedment depth, optimization, frictional soils, harmony search algorithm


In this paper, the design of cantilever soldier pile retaining walls embedded in frictional soils is investigated within the insight of an optimization algorithm to acquire cost and dimension equilibrium by ensuring both geotechnical and structural requirements simultaneously. Multivariate parametric analyses with different fictionalized cases are performed to evaluate the effects of design variants and to compare the effectiveness of the preference of optimization solutions rather than detailed advanced modeling software. The harmony search algorithm is used to conduct parametrical analyses to take into consideration the effects of the change of excavation depth, shear strength angle, and unit weight of soil, external loading condition, and coefficient of soil reaction. The embedment depth and diameter of the soldier pile are searched as design dimensions, and the total cost of a cantilever soldier pile wall is calculated as an objective function. The design dimension results of the parametric optimization analysis are used to perform finite element analysis with a well-known commercial geotechnical analysis software. The results of optimization and finite element solutions are compared with the use of maximum bending moment, factor of safety, and pivot point location values. As the consequence of the study, the influence rates of design variants are procured, and the effectiveness of the usage of optimization algorithms for both cost and dimensional equilibrium is presented.