UNDERSTANDING AND USING GENETIC ALGORITHMS .1. CONCEPTS, PROPERTIES AND CONTEXT

Genetic algorithms are search algorithms founded upon the principles o f natural evolution laid down by Darwin. They tum out to be competitiv e for a certain class of problems - complex large-scale problems, as a rule. Among the favorable properties of genetic algorithms are: effic iency, robustness and versatility. A less favorable property of geneti c algorithms is the imprecision as a result of the noise used by the m ethod. This tutorial consists of two parts which treat a variety of ke y issues concerning genetic algorithms. The first part emphasizes the principles underlying genetic algorithms, their search characteristics and the broader perspective in which they fit. This serves as a gener al, comprehensive introduction. Starting from the first part, the seco nd part of this tutorial elaborates on practical issues such as repres entation, configuration and hybridization with other techniques. There by, some hands-on information is provided, so that common pitfalls can be avoided in using a methodology that exhibits its full power only w hen handled according to the principles it is based upon.