Accelerated shuffled frog-leaping algorithm

Abstract

Shuffled frog-leaping algorithm (SFLA) is a recent addition to the family of stochastic search methods that mimic the social and natural behavior of species. SFLA combines the advantages of local search process of particle swarm optimization (PSO) and mixing of information of the shuffled complex evolution. The basic idea behind modeling of such algorithms is to achieve near to global solutions to the large-scale optimization problems and complex problems which cannot be solved using deterministic or traditional numerical techniques. In this study, the searching process is accelerated using golden section-based scaling factor and the constraints are handled by the penalty functions. Penalty functions are used to find the optimal solution for restrained optimization problems in the feasible region of the total search space. The resulting algorithm is named as Accelerated-SFLA. The proposal is implemented to solve the problem of optimal selection of processes. The results illustrate the efficacy of the proposal.