Augmenting the Crayfish Optimization with Gaussian Distribution Parameter for Improved Optimization Efficiency

In the realm of optimization algorithms, the Crayfish Optimization Algorithm (COA) has emerged as a promising metaheuristic approach influenced by the swarming behavior of crayfish. Many adjustments have been made to successfully improve the search efficiency of meta-heuristic algorithms. Since the beginning, these adjustments have mainly been made to elevate the caliber of the solutions by using various clustering techniques and learning theories, frequently ignoring the population and fitness diversity.When designing an optimization algorithm, diversity is more important. Although COA has strong optimization capabilities, it frequently experiences premature convergence in complex optimization environments. Additionally, two exploitation stages make its exploration capabilities less robust. To overcome these limitations, this paper suggests a control approach that assesses COA’s population and fitness diversity by incorporating the Gaussian Distribution (GD) parameter. The new algorithm is called GD-COA, and its performance is verified by optimizing 27 uni-model and multi-model benchmark functions. The comparison results for the relatively few iterations show that GD-COA is more effective and efficient than traditional COA.