Improved reptile search algorithm with novel mean transition mechanism for constrained industrial engineering problems

Engineering designs are common industrial optimization problems that need an efficient method to determine the parameters of the problems. This paper proposes a novel engineering design parameters identification method based on an enhanced optimization method called IRSA. The conventional reptile search algorithm (RSA) is utilized in the proposed IRSA method with the mutation technique (MT). These two search methods are used to find the optimal parameters values for the given problems and are employed based on a novel mean transition mechanism . The proposed mean transition mechanism adjusts the searching process by changing between the search process (i.e., RSA or MT) to avoid the main weaknesses of the original RSA: the permutation convergence and unbalance between the search methods. Experiments are conducted on ten benchmark functions from CEC2019 and five industrial engineering design problems. The results are evaluated using worst, mean, and best fitness function values. The proposed method is compared with other well-established methods, and it got better and promising results. The proposed IRSA method's performance proved its ability to address the mathematical benchmark functions and engineering design problems.