JayaL: A Novel Jaya Algorithm Based on Elite Local Search for Optimization Problems

Abstract

Many metaheuristic methods have been proposed to solve engineering problems in literature studies. One of these is the Jaya algorithm, a new population-based optimization algorithm that has been suggested in recent years to solve complex and continuous optimization problems. Jaya basically adopts the best solution by avoiding the worst ones. Although this process accelerates the convergence for the solution, it causes concessions in the population and results in inadequate local search capacity. To increase the search capability and exploitation performance of the Jaya algorithm, a new local search procedure—Elite Local Search—has been added to the Jaya algorithm in this study without making any changes in its basic search capability. Thus, an efficient and robust strategy that can overcome continuous optimization problems is presented. This new algorithm created with the elite local search procedure is called JayaL. To demonstrate the performance and accuracy of JayaL, 20 different well-known benchmark functions in the literature were used. In addition to JayaL algorithm, these functions were solved with differential evolution (DE), particle swarm optimization (PSO), artificial bee colony (ABC), dragonfly algorithm (DA), grasshopper optimization algorithm (GOA), atom search optimization (ASO) and Jaya algorithms. The performances of JayaL, DE, PSO, ABC DA, GOA, ASO and Jaya algorithms were compared with each other, and experimental results were supported by convergence graphs. At the same time, JayaL has been applied to constrained real-world engineering problems. According to the analyses, it has been concluded that JayaL algorithm is a robust and efficient method for continuous optimization problems.