A Hybrid PSO-Jaya Algorithm for Optimization Problems

In this paper introduces and investigates a hybrid PSO-Jaya optimization algorithm based on two heuristic algorithms PSO and Jaya. Two problems: function optimization and training ANN for the classification problems Iris and breast cancer Wisconsin, are employed to evaluate the efficiencies of this new hybrid algorithm. In test calculations, the PSO, Jaya, PSO-Jaya algorithms are compared based on the average, median, standard deviation, and “best” of the best particle position after 50 independent runs for benchmark functions and 30 for network training. The results are compared with the PSO and Jaya algorithms. For all test cases, PSO-Jaya shows the best performance in terms of convergence rate and avoidance of local minima.