【作者】 刘延秋；

【导师】 张勤生；

【作者基本信息】 中国海洋大学 ， 管理科学与工程， 2009， 硕士

【摘要】 面对越来越激烈的市场竞争,企业要在世界市场中拥有一席之地,必须建立核心竞争优势,则必须通过深入研发,企业可以通过自主研发、委托研发和合作研发三种研发模式实现研发的目的,在不同的研发模式选择时,技术溢出水平和企业的学习能力是两个重要的影响因素。从本质上讲,企业间的合作研发过程是企业在如何利用他人资源和如何防止被他人利用自己资源之间进行的一系列博弈过程。本文在同时考虑了技术的溢出水平和企业的学习能力情况下,应用博弈理论,建立了一个双寡头企业的完全信息模型。通过分析双寡头企业研发投资和生产阶段两个阶段的博弈,对两个企业在自主研发、委托研发和合作研发之间选择的博弈问题进行了均衡分析,得出三种研发模式下企业有效成本降低幅度和最大利润的均衡解,并对均衡解进行比较,提出了存在子博弈精炼纳什均衡的条件并讨论了该条件与溢出水平和吸收能力的关系,从而得到不同溢出水平和学习能力取值范围对应的最有效的研发模式。通过研究,获得了以下研究成果和研究结论。在企业具有较大的技术溢出水平(特别是技术完全溢出)和学习能力时,合作研发模式无论在成本降低幅度还是最大利润方面都是最有效的研发模式。在技术溢出水平较小时,自主研发可以最有效的降低成本,而合作研发能取得最大的利润;在学习能力较小时,委托研发可以最有效的降低成本,而合作研发能取得最有效的利润。在这两种情况下,企业会遇到模式选择的矛盾,此时企业决策取决于:一是企业符合“理性经济人”假设,会优先考虑利润,会选择合作研发模式;二是企业正扩大市场占有率,采取成本领先战略,会选择自主研发或委托研发模式。企业技术完全溢出和具有较大的学习能力,是指合作研发模式中的合作研发网络组织形式,合作研发网络组织对其成员不是建立在控制关系、合并关系之上,不具有强有力的行政和经济控制权利,正是由于这种合作网络形式的特殊性而导致了它的不稳定性,极易发生网络成员背叛合作承诺的不利情形,造成研发网络的不稳定。本文就这一现象,运用博弈论的方法,分析了合作研发网络的稳定性问题。首先分析了完全信息下合作研发网络企业的决策,分“相互不合作模型”、“相互合作模型”和“单方合作模型”对完全信息下企业的决策进行分析,得出完全信息下合作研发网络研发决策的纳什均衡解。然后本文更加深入的从三个方面分析了不完全信息下研发网络动态稳定性问题,一是,“参与人策略类型”稳定性模型,运用复制动态方程对合作研发网络中参与人的策略类型进行分析,得出影响参与人策略选择的四个因素,从而提出加强合作研发网络稳定性的信任机制、惩罚机制和经济机制:增加相互合作企业的收益、减小相互不合作的收益、增加合作一方的合作收益而减少不合作方的收益。二是,“贴现因子”博弈稳定性模型,量化了参与人对未来收益的贴现因子分析,得出企业贴现因子的取值范围和贴现因子与上述四个因素的关系,得出加强合作研发网络稳定性经济机制和信任机制。三是,“产品关系”博弈稳定性模型,研究了合作研发网络中合作伙伴企业的产品关系对网络稳定性的影响,得到结论产品关系越竞争,网络越不稳定,产品关系越互补,网络的稳定性越好,从而纵向研发模式的上下游企业间更能保持合作研发网络的稳定性。更多还原

【Abstract】 Facing increasingly fierce marketing competition, firms must build up the core competitive advantages and continual completive power through in-depth R&D in order to have a regular place in the world market. There are three models for R&D: independent R&D, consigned R&D and cooperating R&D. Firm’s R&D model is controlled by two factors: technique spill-over and learning ability. Essentially, the process of cooperating R&D among firm is a series of game process during which firm try to make use of others’resources and meanwhile protect its resources from utilizing by others.Given the technique spill-over level and learning capability, this article provides a static model of duopoly firm by game theory. In this thesis, it analyzes duopoly firm’s R&D and puts forward the existence conditions of Nash equilibrium. It also discusses the relationship between Nash equilibrium and absorbing capacity and technique spill-over. It makes an equilibrium analysis about the game of choice between independent R&D,consigned R&D and cooperating R&D, which has a corresponding equilibrium strategy respectively. Lastly, it compares equilibrium solutions of firm’cost-effective decreasing amplitude and a maximum profit in three R&D models.Firm can make full use of internal technical knowledge if they choose cooperating R&D when they have a higher level of spill-over technique and better learning capability, thereby achieving their R&D purpose and gaining a maximum profit.The research net form of researching could fulfill the aim that all firms under the net may share, use, and process their all techniques, at the mean while, the overflow of the firm techniques will get the best level. Partners can learn from each other and absorb the shared-knowledge from the overflow. Integrating with the self-owned knowledge, the firms can strengthen their old, at the same time; they still get their basement for the further research..The cooperating R&D model easily gets the result that its members disobey the commitment and make the research net instability just because the organization of the form not founded from the force measure controlling of the administration and economy.This thesis applies game analysis to point out stability problems of cooperating D&R network. It firstly analyzes firms’strategic decision under complete information, by‘mutual noncooperation model’,‘mutual cooperation model’and‘unilateral cooperation model’obtains Nash equilibrium of D&R strategic decision on cooperating D&R network. And then it analyzes strategy problems of D&R network dynamic stability under incomplete information from three aspects. Firstly,‘participant strategy model’analyzes participants’strategy in cooperating D&R network by replicating the dynamic equation. Secondly,‘discount factor’stability model quantifies the values of discount factor of D&R network so that stability problems on network can be analyzed quantitatively, which the greater discount rate on values received in the future is, the more stable the D&R network. Thirdly,‘product relation’stability model studies the impact generated by product relation of firms in cooperating D&R network and it shows that the more competitive the more unstable the network; in contrast, the more complementary the product relation is, the better the network stability. Consequently, upstream and downstream firms in the chain of vertical D&R model can better ensure the stability of cooperating D&R network.更多还原