基于任务的虚拟企业组织设计研究  

The research on organizational structure design for virtual enterprise base on mission

作　　者：余胜阳[1,2] 赵强[3] 

机构地区：[1]华中科技大学,湖北武汉430074 [2]中国石油化工股份有限公司催化剂分公司,北京100011 [3]武汉科技大学,湖北武汉430081

出　　处：《计算机与应用化学》2011年第4期490-494,共5页Computers and Applied Chemistry

摘　　要：虚拟企业构建具有明显的任务导向特征。当某企业面临资源缺乏或能力限制而导致无法独立完成某项生产任务时,就可以以盟主企业身份通过构建虚拟企业来完成该任务。组织设计关系到虚拟企业运作的成败,因此组织设计问题成为虚拟企业研究一个关键问题。本文以注塑模开发任务为例,探讨了虚拟企业组织设计问题,提出虚拟企业组织设计框架,并对组织设计中的关键问题进行了研究。本文提出的虚拟企业组织设计框架包括任务分解及归并、伙伴集构造及初选、伙伴集优选、组织结构设计等关键环节,文中对上述各阶段的主要内容及实现技术进行了详细分析与说明。针对虚拟企业组织设计中的关键问题—伙伴集优选,建立了基于完工时间约束的,以任务总成本最小化为目标的数学模型,并采用限制法,借助0-1背包问题证明了该伙伴优选问题是NP难题。为求解上述伙伴优选模型,提出了粒子群求解算法。在上述研究基础上,引入爬山算法对粒子群优化算法进行改进,以克服采用粒子群算法求解整数规划问题,容易因为取整而漏掉较优解及易于陷入局部最优等弊端,提高优化求解效果。通过注塑模开发任务的伙伴优选实例验证了粒子群求解算法的实用性及有效性。本文研究对虚拟企业快速、有效地组建具有一定指导意义。Virtual Enterprise（VE） has obvious characteristics of task-oriented.Because of lacking of its resources or of its ability,an enterprise may not able to independently complete a task.As the chief corporate identity,the enterprise can build a VE for completing ／ the task.The success of the operation of VE depends on its organizational design.In this paper,with actual example of injection mold development tasks,the problem of the organizational design of VE is discusses and the organizational design framework for the virtual enterprise proposed and investigated.The key aspects in the proposed framework of the organizational design of VE include task decomposition and merging,set construction and primary selection of partners,partners set optimization.The main contents and the implementation techniques involved in the various stages are explained and analyzed. For partners selection and optimization a mathematical model is established based on minimizing the total cost of task,using task completion time as constraints conditions.With the 0-1 knapsack problem and constraint method,the partner＇s selection and optimization problem is proved to be a NP problem.For solving the above optimization model of the partner＇s selection,a particle swarm algorithm is proposed and climbing algorithm is used to improve particle swarm optimization in order to overcome the defects of easily missing the optimum solution and easily falling into local optimum in solving the problem.This examples demonstrates that particle swarm optimization algorithm is practical and effective.The results presented in this paper give some guidance for the building of VE quickly and efficiently.

关 键 词：任务 虚拟企业 组织设计 伙伴优选 粒子群优化 

分 类 号：TQ015.9[化学工程] TP391.9[自动化与计算机技术—计算机应用技术]