Abstract Genetic algorithms are a random search method based on the biological model of evolution by mating and mutation. In classical genetic algorithm, solutions to problems are encoded into strings of bits that are tested for suitability, and then the best strings of bits are combined to form new solutions using methods that mimic the Darwinian process of "survival." of the strongest” and the exchange of DNA that occurs during mating in biological systems. Programming genetic algorithms involves little more than manipulating bits and evaluating the quality of solutions. Genetic algorithms have been applied to problems as diverse as graph partitioning and the automatic creation of programs corresponding to mathematical functions. Introduction Genetic algorithms are a random search method that "generates" efficient solutions to problems through simulation of Darwinian evolution. A large number of potential solutions are created randomly. The most promising solutions are then grouped together to produce new solutions that receive most of their “genetic stock” from the best solutions of the previous generation. This is similar to the “survival of the fittest” observed in biological systems, where individuals best adapted to their environment produce more offspring, allowing the best-adapted genetic material to be passed on to future generations. The history of genetic algorithms is most commonly attributed to Holland's text "Adaptation in Natural and Artificial Systems", published in 1975. Earlier work by Holland and others shows that the concept of genetic algorithms began to form at the end of the 1960s [1]. Around this time, Bagley first coined the term "genetic algorithm" in his dissertation. Holland...... middle of paper ......Come of Age, David E. Goldberg, Communications of the ACM, March 1994, pp 113-119.[3] Genetic Algorithms: A Survey, M. Srinivas, Latit M. Patnaik, IEEE Computer, June 1994, pp 17-26.[4] Genetic Algorithm and Graph Partitioning, Thang Nguyen Bui, Byung Ro Moon, IEEE Transactions on Computers, July 1996, pp 841-855.[5] http://www.mines.edu/students/d/drferrin/Cool_Beans/GeneMachiene.html[6] Genetic Programming: A Paradigm for Genetically Reproducing Populations of Problem-Solving Computer Programs, John R. Koza, Report Stanford Technical STAN-CS-90-1314, June 1990. Available at ftp://elib.stanford.edu/pub/reports/cs/tr/90/1314/. This material is also covered in Koza's text "Genetic Programming: On Programming Computers Using Natural Selection".[7] http://www.ifh.ee.ethz.ch/~gerber/approx/default.html[8] http://www.novagenetica.com