Potential corrections to grey wolf optimizer

Grey wolf optimizer (GWO), a well-known powerful algorithm that simulates the leadership hierarchy and hunting mechanisms of grey wolves in nature, has garnered significant attention from researchers recently. However, parts of GWO formulations have been shown to be unfit. Moreover, GWO generates outstanding results only for functions with optimal values of 0. Thus, in this paper, the inherent flaws of GWO are discussed and corrected variants are proposed to resolve its inherent flaws. The three corrections to the original GWO proposal made in this study include eliminating coefficient vector C, eliminating the absolute sign for factor D, and introducing a current-to-prey approach. Based on a numerical validation using CEC2005 and CEC2019 benchmark functions, one of the proposed corrected variants performs comparably with other popular optimization algorithms and handles high-dimensional functions exceptionally well. Numerical simulations have elucidated the efficacy of the suggested corrections in mitigating the inherent flaws present in the original GWO. The corrected variants of GWO proposed in this study may be useful in developing future GWO applications and other GWO variants. •This paper fixes GWO flaws via three key changes.•Removal of the coefficient factor C improves algorithm performance.•Eliminating the absolute sign for factor D rectifies potential biases.•A novel strategy curbs greed in GWO by using current-to-prey search.•The improved GWO excels at high-dimensional function optimization.