A study on Solving Unbounded Knapsack Problem Based on Adaptive Genetic Algorithm

• Main Authors: Cheng-Huei Jian, 簡程輝
• Other Authors: Rung-Ching Chen
• Format: Others
• Language: en_US
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/16649717016724242659

碩士 === 朝陽科技大學 === 資訊管理系碩士班 === 95 === The Knapsack problem is an NP-Complete problem. Unbounded Knapsack problems are more complex and harder to solve than the general Knapsack problem. In this thesis, we apply the genetic algorithm using adaptive mechanism which includes greedy method to arrange the chromosomes and automatically adapt the runs to solve the unbounded Knapsack problem. In reproduction procedure, we use elitism strategy to select new offspring. The elitism strategy is utilized to overcome the defect of the slow convergence rate of the general genetic algorithm. The elitism strategy retains good chromosomes and ensures that they are not eliminated through the mechanism of crossover and mutation, while ensuring that the features of the offspring chromosomes are at least as good as their parents. The system automatically adapts the number of the initial population of chromosomes and the number of runs of the genetic algorithm. It will obtain the best value from the chromosomes of each run and retain the best values into an elitism set. The best value is then taken from the elitism set and adapted as the real solution. In addition, we use the strategy of greedy method to arrange the sequence of chromosomes to enhance the effect of executing. Experimental results showed that our method could find the best solution of the problem in evidence space.