Blending Roulette Wheel Selection & Rank Selection in Genetic Algorithms

Both exploration and exploitation are the techniques employed normally by all the optimization techniques. In genetic algorithms, the roulette wheel selection operator has essence of exploitation while rank selection is influenced by exploration. In this paper, a blend of these two selection operators is proposed that is a perfect mix of both i.e. exploration and exploitation. The blended selection operator is more exploratory in nature in initial iterations and with the passage of time, it gradually shifts towards exploitation. The proposed solution is implemented in MATLAB using travelling salesman problem and the results were compared with roulette wheel selection and rank selection with different problem sizes. Genetic algorithms are adaptive algorithms proposed by John Holland in 1975 (1) and were described as adaptive heuristic search algorithms (2) based on the evolutionary ideas of natural selection and natural genetics by David Goldberg. They mimic the genetic processes of biological organisms. Genetic algorithm works with a population of individuals represented by chromosomes. Each chromosome is evaluated by its fitness value as computed by the objective function of the problem. The population undergoes transformation using three primary genetic operators - selection, crossover and mutation which form new generation of population. This process continues to achieve the optimal solution. Basic flowchart of genetic algorithm is illustrated in Figure 1.