Enhanced Local Search in Shuffled Frog Leaping Algorithm

Abstract

Shuffled frog leaping algorithm (SFLA), a recent memetic metaheuristic proposed to solve combinatorial optimization problems. In SFLA, the population consists of group of frogs which are equally distributed into memeplexes. The population of frogs performs cooperative search for food, i.e., local search in each memeplex and global search among memeplexes. In this work, the performance of SFLA is enhanced by accelerating the convergence speed. This performance has been achieved by implementing self adaptive scheme in local search process in the structure of basic SFLA. The enhanced scheme is named as enhanced SFLA (ESFLA) which is validated on a real-time human resource problem. The optimal results of ESFLA present the efficiency of the proposal.