Optimization of a Can Size Problem Using Real Encoded Chromosome in Genetic Algorithm

Abstract

AbstractOne of the major drawback of Genetic Algorithm (GA) based solutions to many optimization problems is the difficulty to obtain convergence to an optimal solution. One of the possible reason for not obtaining good convergence is due to the improper encoding of chromosomes. Many techniques were proposed in some previous researches for improving the convergence of GA based solutions. However, no consideration regarding the role of chromosome encoding in achieving convergence and optimality both has been discussed in the past. In the present work, a can volume optimization problem is solved with the help of two types of chromosome encoding techniques that are proposed and evaluated in GA environment. First, based on single random gene selection and second based on mean value of genes of the encoded chromosome. A numerical example with an objective function and constraints has been solved and the results for each of the scheme is being discussed.