The optimal basics for GAs
H.B. Kamepalli · IEEE Potentials · 2001
Genetic algorithms were introduced by John Holland in early 1970s as a special technique for function optimization. They are quite different from other more conventional optimization methods that are mainly stochastic in nature. A typical GA will have three phases; i.e., initialization, evaluation and genetic operation. In each phase, various parameters of GA need to be selected based on the nature of the optimization problem. A genetic algorithm is also classified based on the various combinations of parameters and strategies employed. However, the designer is free to develop a hybrid genetic algorithm. The main goal is to deliver the most enhanced performance possible to the optimization problem.