EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Optimizing Concrete Compressive Strength Prediction Using ANFIS Models Enhanced by Butterfly, Beluga Whale, and Golden Eagle Metaheuristics

Rufaizal Che Mamat, Azuin Ramli, Muhammad Lukman Kirunjisman

Current Problems in Research · 03.03.2026 · 2 (1) · 1-15

DOI: 10.70028/cpir.v2i1.66 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Current Problems in Research, online erschienen am 03.03.2026. Verfasst von Rufaizal Che Mamat und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The precise prediction of concrete compressive strength is fundamental to ensuring structural integrity and optimizing material use. This study presents a robust comparative analysis of hybrid computational intelligence models, integrating the Adaptive Neuro-Fuzzy Inference System (ANFIS) with three metaheuristic algorithms: the Butterfly Optimization Algorithm (BOA), Beluga Whale Optimizer (BWO), and Golden Eagle Optimizer (GEO). Utilizing a dataset of 400 concrete mix samples, the models were trained and validated to predict mean and characteristic compressive strength. Performance was rigorously evaluated using RMSE, MAE, and R2 metrics. While all hybrid models significantly outperformed the standalone ANFIS, a critical distinction emerged between training and testing performance. The ANFIS-BWO model excelled during training, but the ANFIS-GEO model consistently demonstrated superior stability and generalization, achieving an RMSE of 2.35 and an R2 of 0.93 for mean strength prediction on unseen test data. This superior robustness highlights GEO's efficacy in navigating the ANFIS parameter space to avoid overfitting. The findings underscore that generalization capability is as critical as aggregate accuracy, identifying the ANFIS-GEO hybrid as a highly reliable tool for practical concrete engineering applications.

Zitieren

APA

Mamat, R. C., Ramli, A., & Kirunjisman, M. L. (2026). Optimizing Concrete Compressive Strength Prediction Using ANFIS Models Enhanced by Butterfly, Beluga Whale, and Golden Eagle Metaheuristics. Current Problems in Research, 2(1), 1-15. https://doi.org/10.70028/cpir.v2i1.66

Vancouver

Mamat RC, Ramli A, Kirunjisman ML. Optimizing Concrete Compressive Strength Prediction Using ANFIS Models Enhanced by Butterfly, Beluga Whale, and Golden Eagle Metaheuristics. Current Problems in Research. 2026;2(1):1-15. doi:10.70028/cpir.v2i1.66

BibTeX
@article{Mamat2026,
  author = {Mamat, Rufaizal Che and Ramli, Azuin and Kirunjisman, Muhammad Lukman},
  title = {Optimizing Concrete Compressive Strength Prediction Using ANFIS Models Enhanced by Butterfly, Beluga Whale, and Golden Eagle Metaheuristics},
  journal = {Current Problems in Research},
  year = {2026},
  volume = {2},
  number = {1},
  pages = {1--15},
  doi = {10.70028/cpir.v2i1.66}
}
Kostenlos starten →