Enhanced Ruppel’s Fox Optimizer (eRFO): An Improved Metaheuristic With Adaptive Control and Chaotic Diversity

Abstract

The proposed study introduces an enhanced Ruppel’s Fox Optimizer (eRFO) algorithm as an improved version of the RFO algorithm. This algorithm is designed to address the issues of early convergence, diversity loss, and limited benefit. The proposed algorithm combines four complementary mechanisms. Sinusoidal adaptive control factor phase regulates exploration intensity and step size. Logistic map–based chaotic factors preserve population diversity. Opposition-based learning strategy expands the search region, while the Nelder–Mead local search module aims to improve the best candidate solution. The efficacy of the proposed algorithm has been substantiated using the Congress on Evolutionary Computation 2019 Benchmark Suite(CEC-2019) test set. A comparison has been made between the obtained results and both standard and state-of-the-art search algorithms. These comparisons show that eRFO achieved the best performance in eight out of ten functions. It ranked second in the remaining two. Moreover, eRFO exhibited a 62.8% reduction in average error compared to RFO, while demonstrating comparable computational efficiency. Lower best, mean, and SD values were also offered.

Cite this article as: G. Yüksek, “Enhanced Ruppel’s Fox Optimizer (eRFO): an improved metaheuristic with adaptive control and chaotic diversity,” Electrica, 26, 0298, doi: 10.5152/electrica.2026.25298.