Optimization and Parametrization of Reinforced Concrete Cantilever Retaining Walls - Parametric optimization of reinforced concrete cantilever retaining walls using genetic algorithms

Abstract

Reinforced concrete cantilever retaining walls (RCCRW) are among the most com mon type of infrastructure projects and require detailed, time-consuming calcula tions and designs. Time is a key parameter in the construction industry and opti mizing the process is therefore advantageous, especially during the bidding phase, where projects are secured. This thesis aims to investigate the potential to stream line the design process using an optimization algorithm. The study consisted of a literature review, development of a calculation model, and implementation of a genetic algorithm (GA) through a Python script to minimize investment cost and environmental impact. A convergence and case study were conducted to evaluate and compare the results. The RCCRW was designed in accordance with current Eurocodes and the Swedish Transport Agency’s national specifications for their implementation, Transportstyrelsens Författningssamling (TSFS). The results show that optimization using GA enables one to quickly develop build able solutions for RCCRW using optimization algorithms, which can be highly advantageous. The study has also indicated that a certain level of detail in the calculations is required for further optimization, highlighting a trade-off between optimization and generalization.