Average Gene Method for increasing the convegence speed of discrete genetic algorithms

One of the most popular techniques in evolutionary computation research is the genetic algorithms (GAs). The challenge to increase the convergence speed of the GAs has been always one the most interesting area for researches. The need to boost the convergence speed of the GAs is sensed more obviously when the number of variables in the cost function increases, where it takes a long time to compute the value of the cost function. This paper presents a new method, called Average Gene Method, to increase the speed of the GA. The average gene method is discussed in two different scenarios: with respect to the chromosomes and with respect to the costs. A set of benchmark cost functions is used to compare the results of proposed method with some other known algorithms such as original GA, Jumping Gene method and Particle Swarm Optimization. The simulation results show that when the average gene method is used simultaneously along with the sequential mutation and circular gene methods, the number of generations and cost function evaluations, as two criteria for comparison different algorithms, reduces significantly.