Comparison of Heuristic Algorithms with Performance Metrics on Multimodal Benchmark Functions

Abstract

The solution of difficult problems can be realized in shorter time with heuristic algorithms. There are many heuristic algorithms. In this study, artificial bee colony (ABC), biogeography based optimization (BBO), cuckoo bird search algorithm (CSO), differential evolution (DE), imperialist competitive algorithm (ICA) and particle swarm algorithm (PSO) have been chosen due to reasons such as the widespread use in the literature and the large number of open source code applications to use it widely in the literature and to have a lot of open source code applications. Each of these preferred algorithms has been run 30 times in a 10-dimensional search space with the same initial positions and conditions to find the global minimum point on the 5 benchmark function, which is also frequently used in the scientific world. The performance of the algorithms based on the results obtained from the runs has been determined by the best cost, worst cost, accuracy, stability, time and standard deviation performance metrics. The performance scores of the algorithms are evaluated based on the cumulative average ranking value generated from the results of these performance metrics.