crossover

crossover

<< Click to Display Table of Contents >>

Navigation:  »No topics above this level«

crossover

Return to chapter overview

Crossover is an operator which is applied to two parent individuals from the current generation in order to produce two new offspring.  One of the offspring passes into the next generation.  The mates for the crossover are chosen randomly from the initial population with the restriction that no individual can crossover with itself.  If the mates consist of binary chromosomes, the crossover is implemented by choosing a random position (crossing point) in the chromosome, and exchanging the segments either to the right or left of this crossing point with another chromosome similarly partitioned.  The crossover operator gives a new point within the problem parameter space for testing.  In unique gene representation, partially-matched crossover is applied by choosing two random, different crossing points and exchanging and swapping the segment contents from one individual to another.