Multi-objective optimization of reinforced concrete cantilever retaining wall: a comparative study

Abstract This paper investigates the performance of four multi-objective optimization algorithms, namely non-dominated sorting genetic algorithm II (NSGA-II), particle swarm (MOPSO), strength Pareto evolutionary (SPEA2), and multi-verse (MVO), in developing an optimal reinforced concrete cantilever (RCC) retaining wall. The wall design was based on two major requirements: geotechnical stability structural strength. Optimality criteria were defined as reducing total cost, weight, CO 2 emission, etc. In this study, sets bi-objective strategies considered: (1) minimum cost maximum factor safety, (2) weight safety. proposed method's efficiency examined using numerical examples, one with a base shear key without key. A sensitivity analysis conducted variation significant parameters, including backfill slope, soil’s friction angle, surcharge load. Three well-known coverage set measures, diversity, hypervolume selected to compare algorithms’ results, which further assessed basic statistical measures (i.e., min, max, standard deviation) Friedman test 95% level confidence. results demonstrated that NSGA-II has higher rank terms for both cost-based weight-based designs. SPEA2 MOPSO outperformed solutions diversity examples effects key, respectively. However, measure, MVO have respectively,