The original Handbook of Evolutionary Computation (Back et a1 1997) was designed to fulfil1 the need for a broad-based reference book reflecting the important role that evolutionary computation plays in a variety of disciplinesranging from the natural sciences and engineering to evolutionary biology and computer sciences. The basic idea of evolutionary computation, which came onto the scene in the 195Os, has been to make use of the powerful process of natural evolution as a problem-solving paradigm, either by simulating it (‘by hand’ or automatically) in a laboratory, or by simulating it on a computer. As the history of evolutionary computation is the topic of one of the introductory sections of the Handbook, we will not go into the details here but simply mention that genetic algorithms, evolution strategies, and evolutionary programming are the three independently developed mainstream representatives of evolutionary computation techniques, and genetic programming and classifier systems are the most prominent derivative methods.

In the 1960s, visionary researchers developed these mainstream methods of evolutionary computation, namely J H Holland (1 962) at Ann Arbor, Michigan, H J Bremermann (1962) at Berkeley, California, and A S Fraser (1957) at Canberra, Australia, for genetic algorithms, L J Fogel (1962) at San Diego, California, for evolutionary programming, and I Rechenberg ( 1965) and H P Schwefel (1965) at Berlin, Germany, for evolution strategies. The first generation of books on the topic of evolutionary compuation, written by several of the pioneers themselves, still gives an impressive demonstration of the capabilities of evolutionary algorithms, especially if one takes account of the limited hardware capacity available at that time (see Fogel et a1 (1966), Rechenberg ( I 973), Holland ( 1975), and Schwefel ( 1977)).