A Partheno-genetic Algorithm for Combinatorial Optimization
Genetic Algorithms (GA) Using ordinal strings for combinatorial optimization must use special crossover operators such as PMX, OX and CX, instead of general crossover operators. Considering the above deficiency of GA using ordinal strings, a Partheno-Genetic Algorithm (PGA) is proposed that uses ordinal strings and repeals crossover operators, while introduces some particular genetic operators such as gene exchange operators, which have the same function as crossover operators. The genetic operation of PGA is simpler and its initial population need not be varied and there is no “immature convergence” in PGA. The schema theorem of PGA was analyzed. Similarly with TGA, by genetic operators processing schemas, the individuals in the population continually move towards optimal individual in PGA, finally the optimal solution can be gained. The global convergence of PGA was studied. It was also proved that optimal maintaining operation is the key operation to make the algorithm global convergent.
KeywordsGenetic Algorithm Global Convergence Crossover Operator Combinatorial Optimization Problem Genetic Operator
Unable to display preview. Download preview PDF.
- 2.Xiaoyuan, H., Sihan, X., Shulin, W.: Application of Genetic Algorithm in Train Line Holding. Information and Control 25(1), 58–63 (1996) (in Chinese)Google Scholar
- 3.Bo, W., Qun, Z., Fei, W., et al.: Quantitative Analysis of Infeasible Solution to Job Shop Scheduling Problem. Control and Decision 16(1), 33–36 (2001) (in Chinese)Google Scholar
- 6.Maojun, L., Tiaosheng, T.: A Partheno-genetic Algorithm and Its Applications. Journal of Hunan University 25(6), 56–59 (1998)Google Scholar
- 7.Maojun, L., Taoye, Z., Tiaosheng, T.: Comparison Between Partheno-genetic Algorithm and Traditional Genetic Algorithm. System Engineering 19(1), 61–65 (2001)Google Scholar