Mutation Operator

It is a background operator that produces spontaneous random changes in several chromosomes. An easy way to get mutation would be to modify one or more genes. In genetic algorithm, the crucial role is serves by mutation either

(a) Replacing the genes lost from the population throughout the selection process hence they can be tried in a new context, or

(b) Giving the genes such were not present in the initial population.

The mutation rate or pm is explained as the percentage of the total number of genes in the population. Its rate controls the rate at that new genes are introducing into the population for trial. If it is also low, many genes that would have been helpful are never tried out; however if it is too high, there will be much randomly perturbations, the offspring will start losing their resemblance to the parents, and the algorithm will lose the capability to learn from the history of the search. The pseudo-code conventional genetic algorithms are presented in Programme 2.

Four major parameters that affect the performance of genetic algorithm are as: population size, crossover rate and number of generations, mutation rate. Larger population size that is hundreds of chromosomes and large number of generations or thousands raise the likelihood of getting a global optimum solution, however substantially increase processing time.        

Begin

        Generate random population of P solutions (chromosomes);

        For each individual i ∈ P ; calculate fitness (i);

                For i = 1 to number of generations;

                      //Randomly selects mutation operation or crossover operation   

                      If crossover;  // check if cross over

;                               Select two parents at random i_a  and i_b

                             Generate on offspring i_c = crossover (i_a and i_b)  

                     Else If mutation;   // else condition is use

                             Select one chromosome i at random;

                             Generate on offspring i_c = mutate (i);  

                      End if; // end of if condition

;                         determine the fitness of the offspring i_c is better than the worst chromosome then                         If i_c

                       Replace the worst chromosome by i_c

                  Next i; // increment the number of generations

             Check if termination = true

End

Programme no. 2: Pseudo-Code of Conventional Genetic Algorithm