基于遗传算法的虚拟企业伙伴选择问题研究.docx

基于遗传算法的虚拟企业伙伴选择问题研究摘 要在全球经济飞速发展的今天，传统企业的弊端愈发明显，面对企业日益增多的问题，虚拟企业的产生对于企业的规划发展有着重要作用。虚拟企业是一种新型的生产模式，一种具有代表性的合作企业，它实质上是暂时性的，可能是仅存在于一个或几个的项目周期内的企业网络结合。虚拟企业省略部分功能，仅保留企业中最为关键的功能和职能部门，运营机制的灵活可以使其降低市场风险，充分地利用资源。但虚拟企业在帮助企业解决问题的时对于伙伴的选择是极为重要的。本文利用计划评审技术(program evaluation and review technique ,PERT)来描述各个子项目的时序约束，并建立相应的非线性整数规划模型，将遗传算法嵌入PERT中，来对问题进行求解。利用遗传算法的基本原理，构造问题的解的初始种群，经过选择、重组、交叉和变异来产生新的最优的解集种群。研究结果表明，遗传算法对于虚拟企业的伙伴选择问题在时间和结果方面是非常理想的，是一种行之有效的管理工具。关键词：虚拟企业；遗传算法；伙伴选择：PERTResearch on the partner selection of Virtual Enterprise Based on genetic algorithmAbstractWith the rapid development of the global economy, the disadvantages of the traditional enterprises are becoming more and more obvious. In the face of the growing problem of the enterprises, the production of virtual enterprise plays an important role in the planning and development of enterprises. Virtual enterprise is a new mode of production, a representative of the cooperative enterprise, it is essentially temporary, may be only in one or a few of the project cycle enterprise network combination. Virtual enterprise omitted part of the function, only to retain the most critical functions and functional departments in the enterprise, the flexible operation mechanism can make it reduce market risk, make full use of resources. But the virtual enterprise is very important to the choice of the partner when helping the enterprise to solve the problem. The PERT (program evaluation and review technique, PERT) to describe timing constraints for each sub project, and establish the corresponding nonlinear integer programming model, genetic algorithm is embedded in PERT to solve the problem. Based on the basic principle of genetic algorithm, the initial population of the solution of the problem is constructed, and the new optimal solution set population is produced through selection, recombination, crossover and mutation. Research results show that the genetic algorithm for the virtual enterprise partner selection problem in terms