Statistical Analysis and Comparison of the Performance of Meta-Heuristic Methods According to their Application as well as Defining New Criteria

Abstract

Intelligent meta-heuristic algorithms are used in various scientific branches including engineering sciences due to the importance of optimization. These algorithms are able to solve most complex engineering issues in very high dimensions and in desirable time and obtain answers very close to the optimal answers, and also, they have a high ability of adaptability and flexibility with a variety of optimization issues. In recent years several research efforts have been made in this area, but this approach has become so important that still novel methods are being added to this area, which caused an increase in the application of this method in other sciences. Undoubtedly, due to the excessive number of meta-heuristic methods, it’s time to write or even invent criteria and standards for classification and rating them. The development of such criteria is for explaining the most important question that arises for a user when facing the variety of these tools, and that is which method and in which area can be used? And what are the real challenges ahead for using these methods? The purpose of this paper is to address these issues. For this, statistical analysis is conducted on each of these methods to clarify the application of each of these methods for the use of the users. Also, new criteria are defined to compare the performance of the meta-heuristic methods to introduce suitable substrate and suitable quantitative parameters for this purpose.