关键词: 分布估计算法, 紧致遗传算法, 混合流水车间, 概率模型

Abstract: According to the characteristic of Hybrid Flow-shop Scheduling Problem (HFSP) and Estimation of Distribution Algorithm (EDA),a new method of building probabilistic relationships between HFSP variables was proposed.In this new method,the variable was grouped by process,and the variable probabilistic relationship in a high level was built through the probabilistic constraints in the same group and the probabilistic coupling in different groups.The mechanism of population generation and probability updating was improved in the standard compact Genetic Algorithm(cGA)which accelerated the convergence rate of algorithm.The effectiveness and robustness of the algorithm was proved by simulation and the comparison with other algorithms and the verification of large-scale practical problem.

Key words: estimation of distribution algorithm, compact genetic algorithm, hybrid flow-shop, probability model