Genetic algorithms are widely used in business applications, engineering and in scientific fields.This algorithm can be used to obtain a proper solution to the problem of optimal than one variable or multiple variables. Before the algorithm is run, what problems to be optimized it must be stated in the objective function, known as the fitness function. If the fitness value of the greater, then the system generated the better. Although at first all possible fitness values are very small (because these algorithms produce random), some will be higher than the other. Chromosomes with high fitness values will provide a high probability to reproduce the next generation. So for every generation in the evolutionary process, fitness function that simulates natural selection, will push the population towards an increased fitness.
Very precise genetic algorithm used for solving optimization problems are complex and difficult to be solved by using conventional methods. As the process of evolution in nature, a simple genetic algorithm generally consists of three operators: reproduction operator, crossover operator (crossover) and mutation carriers. The general structure of a genetic algorithm can be defined with the following steps:
Generating the initial population, initial population is generated randomly so we get the initial solutions. The population itself consists of a number of chromosomes that represent the desired solution.
Forming a new generation, used in forming the three operators mentioned above the carrier reproduction / selection, crossover and mutation. This process is done repeatedly so we get a sufficient number of chromosomes to form a new generation which is a representation of this new generation of new solutions.
Evaluation of the solution, this process will evaluate each population by calculating the fitness value of each chromosome and to evaluate it until the stop criteria are met. If the stop criteria have not been met then a new generation will be formed again by repeating step 2. Some stops are frequently used criteria include:
- Stops at a certain generation.
- Stopped after several successive generations obtained the highest fitness value does not change.
- Stops when the next generation n not get a higher fitness value.
Other Algorithm: