Adaptive Evolutionary Genetic Algorithms on a Class of Combinatorial Optimization Problems

Abstract

This paper investigates an adaptive evolutionary genetic algorithm on combinatorial optimization problem, where the solution space can be organized in form of a subset tree. A kind of genetic gene uniform encode scheme and adaptive evolution idea are used before proceeding crossover operation, and crossover is achieved between the current and previous generations individual. The orthogonal table approach is utilized to produce initial population, which can satisfy the multiplicity of the initial population. Two examples are provided to illustrate the effectiveness of the proposed methods.