Reproduction

 Reproduction is the first step of the genetic algorithm. During this process individual

strings are copied according to their objective function values, or fitness values. Copying

strings according to this measure means that strings with higher fitness values have a

greater probability of propagating through the next generation. This is an artificial

version of "natural selection." In terms of an algorithm, reproduction may be

implemented in several ways. Goldberg creates a biased roulette wheel where each

current string in the population has a slot proportional to its fitness [6]. An illustration of

this idea is shown in figure 4.3.


String # String Fitness % of Total
 1 11011 385 45.9
 2 10001 141 16.8
 3 00110 62 7.4
 4 10101 251 29.9
 Total 839 100


 Figure 4.3: Sample string types and biased roulette wheel

A simple spin of the weighted roulette wheel yields the reproduction candidate, which is

solely based on fitness values and the overall probabilities. Using this method, more

highly fit strings have a greater number of offspring in the succeeding generation. Once a

string has been selected for reproduction an exact replica of the string is created and it is

then entered into the mating pool, resulting in a tentative new population.

Crossover

 After reproduction, simple crossover proceeds in two steps. Members of the new

population are mated at random. Next, each pair of strings undergoes crossover: a

position i along the string is selected at random between 1 and then string length (L)