The quality of optimisation by genetic algorithms Original Research Article

The recently introduced quality criteria for optimisation, describing the coverage of the search and solution spaces as well as the reproducibility of both, are applied in combination with experimental design to fine-tune parameter settings and fitness function of a genetic algorithm for the structure optimisation of a heptapeptide. A series of influences of the investigated parameters are revealed by these criteria, while none of them seem significant from the fitness values of the last population alone. It is therefore suggested to apply these criteria, which are not based on the fitness value of the final population, when developing genetic algorithms. It is shown that they are easily adaptable to specific problems.