【作者】 周勇；

【导师】 郑丕谔；

【作者基本信息】 天津大学 ， 管理科学与工程， 2007， 博士

【摘要】 动态联盟作为一种新的组织形式,其组建的目的和范围超出了其最初起源时在制造领域对单一敏捷性的要求,已扩展到知识创新、市场开发等诸多领域。为了解释这种经济现象和利用其组织优势,本文系统研究了有关动态联盟的合作伙伴选择、利益分配、组织学习的理论和管理机制的一些问题。首先,动态联盟中重要的决策是联盟伙伴的选择问题。在目前研究中很少考虑专家在评价过程中对某些不确定性指标值估算的差异和评价结果的置信度,而且在方法上局限于单一伙伴选择,造成对伙伴网络缺乏优化。本文建立了一套考虑质量、技术、绩效等要素的合作伙伴选择评价体系。鉴于一些指标很难给出准确的评价值,以及决策者对指标重要性的主观评价、指标在决策中的作用和指标评价值的可靠程度对不同目标动态联盟指标权重的影响,使用了集值统计方法。在确定了联盟候选伙伴后,还根据它们之间的网络关系对目标进行了优化,选择最终的合作伙伴组合,从而确保参与成员不论是从个体还是从整体角度衡量都是有效的。其次,目前动态联盟利益分配过程博弈研究主要集中在Shapley解的讨论和应用中,缺乏从讨价还价方向的系统性分析。本文从5条Nash谈判公理出发,讨论了两企业利益分配的纳什解、联盟解和鲁宾斯坦解,并将两企业分配的情况推广到多企业利益分配。为了动态联盟利益分配有效进行,给出了动态联盟设计的原则,保证联盟分配过程有可行的各成员都认可的结果。在动态联盟管理机制方面,针对目前研究中理论和方法的不足,重点研究了绩效评价、核心能力保护、信任和关系资本的发展。为了改善联盟失败率高的问题,将平衡记分卡的概念引入到动态联盟的评价中。除了财务指标外,还对顾客满意度、内部程序和组织的学习和成长活动的业务指标进行测评,并给出了评价的二级指标体系的示例。分析了构成动态联盟伙伴企业核心能力知识资产的流失渠道,及其面临的风险的类型、来源和后果,并给出了风险防范的方案体系。利用博弈论的方法论证了信任的作用机制,提出了信任和关系资本的培养路径,以及联盟文化对它们的影响。动态联盟的组织学习是维系动态联盟发展的重要环节,但目前对其学习过程及其对知识创造的影响还缺乏系统研究。本文从渐进式学习和突破式学习两方面进行研究。利用Nonaka的SECI模型和“Ba”的概念,构建了动态联盟的渐进式学习模型;论述了通过知识场和动态学习循环,动态联盟可以进行渐进式的、螺旋上升的知识创造循环。在Argyris的单循环学习和双循环学习概念的基础上,构建了动态联盟的突破式学习模型,有效地描述了联盟成员间不断发展的知识传递和基于冲突的创造性思维和知识创新。最后,本文在结论与展望中总结了主要研究结果及其局限性,特别是根据目前的研究结果,提出了有待进一步研究的几个关键问题。更多还原

【Abstract】 A dynamic alliance (DA) is a new form of organization of which the function and applicative scopes have gone beyond the single requirement for agility within manufacturing field compared with it’s original purposes. It’s function has been extended to a number of domains such as knowledge creation and marketing. In order to interpret this economic phenomena and take a good advantage of it, such problems as selection of alliance partner, benefit allocation, organization learning and management mechanism of DA are explored systematically in this dissertation.Firstly, selection of alliance partner is a critical decision in formation of DA. Several problems including consideration of variation in estimates of some indefinite items in the course of evaluation by experts, confidence level of evaluation results and limitation of selective methods to single partner resulting in lack of optimation of partner’s network are seldom investigated in the existing researches. An instrument for selection of partner with consideration of such dimensions as quality, technology and performance, etc., is constructed. In view of difficulty in precise estimate of some items and the effect of such factors on the weights of items of DA with different targets as subjective evaluation of item’s weight by appraisers, importance of items in decision-making and reliability of estimate, a set value statistical method is put forward. In order to determine the final combination of alliance partners, objective optimization is made based on the network relationships among candidate partners, making sure validity from the perspective of individual partner and as a whole.Secondly, Researches on benefit allocation in DA by means of game play theory are focused only on discussion and application of Shapley solution, and are lack of systematic analysis from bargaining perspective.Based on 5 items of Nash axioms, this dissertation presents a game model for benefit allocation between partners of a two-partner DA and analyzes Nash solution, alliance solution and Rubinstein solution, and extends a solution into a case of multi-partner DA. A mechanism of design of DA is proposed for benefit allocation agreed upon by all members in the alliance.As for the management mechanism of DA, in view of deficiency of current theories and methods, the focus of research is put on performance evaluation, protection of core competence, development of trust and relationship capital. In order to reduce high failure rate in setting up a DA, a balanced scorecard is introduced into evaluation of DA, which integrates customer satisfaction, internal process, organizational learning and growth activities into evaluation instrument besides financial dimension. A sample set of secondary items for evaluation is given. In addition, losing channels of knowledge assets forming core competence of members of a DA, and classification, source and consequence of risks faced in partners’intellectual property right are analyzed with a proposal of risk–preventing measures. Functional mechanism of trust is proved in virtue of game theory with some measures suggested to develop a trust system and social relation capital and some valuable discussions of effect of alliance culture on them given.Although organizational learning plays a key role in maintenance and development of DA, there still exists lack of systematic research on learning process and it’s effect on creation of knowledge. This dissertation carries out research from such two perspectives as incremental learning and critical learning. Incremental learning model of DA is constructed on the basis of SECI model and concept of“Ba”brought forward by Nonaka. A DA is able to perform incremental and spiral cycle of knowledge creation by means of knowledge field and dynamic learning cycle. Critical learning model of DA is constructed based on concepts of single loop and double loop put forward by Argyris, who effectively depicts increasingly developed knowledge transfer and creative mind and knowledge creation derived from conflicts among alliance partners.Finally, at conclusion and outlook, major research results and limitations are summed up, and several key issues are put forward for further research based on present conclusions.更多还原