敏捷制造中面向盟友选择问题的遗传算法

A genetic algorithm focusing on the problem of partner selection in agile manufacturing

摘要:

为解决敏捷制造中组建动态联盟的盟友选择问题,在分析传统盟友选择方法局限性的基础上,设计求解敏捷制造中面向盟友选择问题的遗传算法,建立算法模型,论述遗传算法的个体编码及初始群体产生方法、概率淘汰与轮盘赌相结合的选择方法以及编码循环取值的变异算子设计方法,实例分析验证了算法的稳定性、可靠性及高效的收敛速度.验证结果表明算法能给相关研究提供参考,亦可用于动态联盟盟友选择问题的指导.

Abstract:

To solve the problem of partner selection in establishing dynamic alliance in agile manufacturing,a genetic algorithm based on the analysis to the limitation of traditional selection method was designed and the corresponding algorithm model was constructed.The method for individual coding and initial population generating,the selection method for combining probability elimination with roulette and the design method for mutation operator about coding cycle value were discussed.Example analysis verified the stability,reliability and high convergence speed of the algorithm.It is concluded that the algorithm is beneficial for solving the partner selection problem in dynamic alliance.