摘要:Three alternatives within the method of assignment by classes are presented for thecalculation of individuals lifetime in genetic algorithms with varying population size.(GAVaPS).In the proposed strategy (assignment by classes) individuals are grouped according to theirfitness. The purpose is to use the allowed range of lifetime values in a way which is moresuitable to search the optimum than proportional, linear and bilinear strategies.A comparative study of three possibilities of assignment by classes as related to the traditionalmethods is carried out, and results are shown over five functions. Finally, some conclusionsare presented, along with possible future lines of work