I suggest a query in the subject literature and Internet resources. For example: https://www.tutorialspoint.com/genetic_algorithms/genetic_algorithms_parent_selection.htm

There are also materials on the web with source codes in programming languages. Familiarizing yourself with GA will certainly be more fruitful than asking about individual issues. Regards.

thank you Maciej Karczyński ,

I am sorry to answer with delay. actually, I have already read a lot about metaheuristic optimization algorithms, especially genetic algorithm. anyway, I took a look at the link you'd prepared for me in your answer and moreover, I researched more and more on genetic algorithm selection methods. But, as mentioned in the aforementioned link, in roullete selection method, probability of choosing a specific individual is dependent on its cost function. the better the cost function is, the more probable is to be chosen. So, having a constant number (like 0.3) for selection probability is still meaningless to me.

@Mohammad Reza Moravej

Please look at the roulette wheel. The areas of the slices contained in it are proportional to the fitness function of a given individual (must be > 0) in relation to the SUM of the fitness functions of ALL individuals. So with the "turn" of the wheel, BETTER-fit individuals have MORE CHANCE of choice. A selection algorithm according to the "roulette wheel" is in the link .

Dear Mohammad Reza Moravej. I made a HUGE mistake. I beg your pardon. I just looked at the picture you posted. Of course, CONSTANT probability is incompatible with the "roulette wheel selection" scheme. I think that the solution should be sought in the publication in which the drawing was presented.

It's certainly not about "roulette wheel" selection. Here is a quote from the article "Determination of the Optimal Design of ThreeRing Concentric Circular Antenna Array Using Evolutionary Optimization Techniques":

>

I DON'T KNOW what "natural selection" is in this context.

Dear Maciej Karczyński ,

here is the articles referring to GA using natural selection theory:

(1) (PDF) Genetic Algorithm Based on Natural Selection Theory for Optimization Problems (researchgate.net)

(2) (PDF) Genetic Algorithm Optimization by Natural Selection (researchgate.net)

although I have studied already this article completely, I could not still understand where selection probability term is used in this approach.

Thank you