a search procedure from artificial intelligence in which populations of solutions to a problem (usually encoded in strings of bits) are combined to make new possible solutions for the problem. A fitness measure is used to determine which solutions are suitable for making the new populations of solutions. Genetic operators, such as crossover (where sections of two solutions are interchanged) and mutation (where various bits in a solution are switched), are used to produce the new generations of solutions. This approach to creating problem solutions is intended to be an analogue of natural selection in actual evolutionary processes.