基于精炼贝叶斯均衡的制造网格资源交易研究

在制造网格资源市场中，买卖双方有关产品信息是不对称的，一些资源服务提供者（RSPs）利用其拥有的信息优势制造混同均衡，使得资源服务消费者（RSDs）在信息不完美情况下无法辨别资源质量高低的情况。本文提出基于精炼贝叶斯均衡的制造网格资源交易，在制造网格系统架构的支持下，RSPs通过承诺赔偿价格的信息作为传递手段，使RSDs获得更多的资源质量信息，提高其判断、辨别资源的质量能力，使RSPs达到分离均衡的目的，从而防止低质量产品的RSPs模仿高质量产品而向RSDs发送错误信号，以诱使RSDs购买、使用其低成本低质量产品，从而获得高利润。仿真实验结果显示此博弈模型有一个稳定、合理、资源卖方不会主动偏离的精炼贝叶斯分离均衡。

Study on the Resource Trading in Manufacturing Grid based on Perfect Bayesian Equilibrium

In the manufacturing grid (MGrid) resource market, some resource service providers (RSPs) with resource information superiority, tend to make the pooling equilibrium. Therefore, resource service demanders (RSDs) can’t recognize high-quality resources due to the imperfect information. This paper proposes an approach based on perfect bayesian equilibrium, in order to prevent the resource trading fraud in the MGrid resource market. In the support of the MGrid system architecture, RSPs will promise a compensation as the display of resource quality. In order to make RSPs achieve the separating equilibrium, the paper focuses on preventing low-quality RSPs from sending out a higher compensation which high-quality RSPs would promise. The simulation results show that the game theoretical model has a stable, reasonable and perfect Bayesian separating equilibrium, from which RSPs wouldn’t deviate initiatively.