Similar to the process of evolution in nature, in a genetic algorithm we will have to select the best evolved ones by deciding which ones are superior. A chromosome’s superiority is called its fitness. Normally in most practical use of genetic algorithm, the better the fitness, the higher the chance the chromosome will be selected. A way of determining the fitness of a chromosome is to compare how good it is to the initially proposed characteristics we want represented by the chromosomes to have. Mathematically we could define a fitness function which returns the fitness of chromosome . The larger value returns, the better the fitness the chromosome has. The idea of a fitness function may seem abstruse to many, a simple example could explain it more clearly - consider if we are evolving a car to run on a bumpy road as far as possible without falling over, one possible fitness function is simply the distance traveled by the car represented by its chromosome .

Moreover, if we plot a fitness landscape while we evaluate the average fitness we therefore can represent the solutions visually -