Genetic Algorithms (GA) is an optimization technique that is used in many fields of science and engineering. It is a natural-inspired process evolved based on the Darwinism Theory developed by Charles Darwin.

Charles Darwin was a British scientist, who laid the foundations of the theory of evolution. In 1831, he joined a five year scientific expedition on the survey ship Beagle. Darwin found that evolution occurs through random variation of heritable characteristics, coupled with natural selection (survival of the fittest). It is called Darwinism Theory. He worked on his theory for 20 years and assumed that parents' qualities mix together in the children's organism. Favorable variations are preserved, while the unfavorable are rejected. On the other hand, there are more individuals born than can survive, so there is a continuous struggle for life. The living beings (animals or plants) best suited to their environment are more likely to survive and reproduce, passing on the characteristics which helped them survive to their offspring. As a result, the species changes over time gradually.

Gregor Mendel is another scientist who conducted experiments about genetics. He was born in Czech Republic in the 19th century and known as the "father of modern genetics”. He cultivated garden peas and conducted hybridization experiments. He studied seven traits of pea plants that seemed to inherit independently of other traits (flower color is purple or white, flower position is axial or terminal, stem length is long or short, seed shape is round or wrinkled, seed color is yellow or green, pod shape is inflated or constricted and pod color is yellow or green). He demonstrated that the inheritance of certain traits in pea plants follows particular patterns, now referred to as the laws of Mendelian inheritance.

However, GA became popular through John Holland's work in the early 1970s, and particularly after his book “Adaptation in Natural and Artificial Systems” (1975) was published. He is an American scientist and Professor of psychology and Professor of electrical engineering and computer science. Holland presented a mathematical model that allows for the nonlinearity of complex interactions. Further, he demonstrated the model's universality by applying GA to economics, psychology, game theory, etc.

Nowadays, GA is widely used as an optimization technique.

The GA optimization process starts with a randomly created candidates (Figure 1). It is called the initial population. The population size should be large enough to effectively create the next generation, but, small enough to handle practically. The characteristics of individuals in the population are different from each other. Their performance or suitability is measured in terms of fitness. Therefore, the objective is to select the best candidate (fittest individual).

Secondly, any pair of candidates (individuals) in the generation are selected randomly. They are called parents. A pair of parents generates a pair of children. In the reproduction, parents go through crossover and mutation. Therefore, children have some common characteristics as parents, but they are different from parents.

Characteristics of individuals are defined in their chromosomes. Chromosomes consist of many genes. Genes can be represented as “1”s and “0”s. Therefore, a chromosome can be defined as a string of bits. During crossover, genes of a randomly selected pair of individuals are exchanged in order to create the individuals of the next generation. Mutation is another important genetic operator that randomly changes a gene of a chromosome. In binary representation, mutation operator changes a "0" to "1" and vice-versa.

Once the parents generate children, the population size is doubled. Therefore, the fitness of all the individuals are checked and the best individuals are selected as the next generation. Likewise, the GA optimization process is repeated until finding the best solution.

The GA optimization method is quick as it finds the best solution without checking each individual. Further, no human interaction is needed when running the GA optimization process in a computer. Once feeding the GA parameters to a computer, it runs until finding the optimized solution. Due to these advantages, GA optimization is used in different fields such as mathematics, scheduling, data analysis, economics and finance, networking and communication, circuit design and antenna engineering.

Dr. (Mrs.) J. M. J. W. Jayasinghe
Senior Lecturer
Department of Electronics
Wayamba University of Sri Lanka.