Date of Award


Publication Type

Master Thesis

Degree Name



Computer Science


Evolutionary Algorithm, Multilevel Selection, Multilevel Selection in MPCA, Multi-Population Cultural Algorithm, Optimization


Kobti, Ziad




This is a study on the effects of multilevel selection (MLS) theory in optimizing numerical functions. Based on this theory, a new architecture for Multi-Population Cultural Algorithm is proposed which incorporates a new multilevel selection framework (ML-MPCA). The approach used in this paper is based on biological group selection theory that states natural selection acts collectively on all the members of a given group. The effects of cooperation are studied using n-player prisoner’s dilemma. In this game, N individuals are randomly divided into m groups and individuals independently choose to be either cooperator or defector. A two-level selection process is introduced namely within group selection and between group selection. Individuals interact with the other members of the group in an evolutionary game that determines their fitness. The principal idea behind incorporating this multilevel selection model is to avoid premature convergence and to escape from local optima and for better exploration of the search space. We test our algorithm using the CEC 2015 expensive benchmark functions to evaluate its performance. These problems are a set of 15 functions which includes varied function categories. We show that our proposed algorithm improves solution accuracy and consistency. For 10 dimensional problems, the proposed method has 8 out 15 better results and for 30-dimensional problems we have 11 out of 15 better results when compared to the existing algorithms. The proposed model can be extended to more than two levels of selection and can also include migration.