Hybrid ELM models-based strength prediction model for self-compacting-concrete

The study presents a review on hybrid Extreme Learning Machine (ELM)-based soft-computing methodology for computing the compressive strength of self-compacting concretes (SCC). Due to its advantages of better quality and aesthetic, as well as suitability for addition of supplementary environment-friendly cement substitutes, SCCs have gathered enormous attention in construction engineering. While the strength prediction of SCCs remains problematic due to constraints like complex constitution, ML-based methodologies have received enormous attention in the field. The application of hybrid ELM models is novel in the field of SCCs, though it has been proved to be a robust alternative to traditional methods in many other fields of engineering. The study develops three hybrid ELM models by integrating three efficient optimization algorithms to the ELM algorithm, namely Particle Swarm Optimization (PSO), Improved firefly algorithm (IFF) and Equilibrium Optimizer (EO). The results report that ELM-EO (R² = 0.916, RMSE = 0.065) is the best performing model in comparative analysis and outperforms the traditional ELM model. The results of the study are compared from the previous studies in literature and the ELM-EO model is concluded as best among them. The proposed methodology provides a robust and efficient alternative for SCC strength prediction, offering potential for practical implementation in the construction industry.