贝叶斯决策网在智能决策系统中的应用研究

面向复杂的决策问题，提出基于贝叶斯决策网在智能决策支持系统中的应用，通过决策网可以直接计算不同行动方案下的效用，直接为决策者服务。文章深入阐述了贝叶斯决策网的语义、语法、效用函数的构造和详细探讨了在求解复杂决策问题中决策模型构建和推理算法实现的方法和途径。最后建立了供应链管理中零售商预定冰激凌的实例，实际运行表明把基于贝叶斯决策网的智能决策系统用于复杂决策问题是有效的，尤其适用于多角色参与的决策问题。

Study and Application of The Bayesian Decisions Network in Intellingent Decisions System

With the complex problem of decision, the application of Bayesian decisions network in inteiligent decisions system is studied. The networks of making decisions can directly presents the efficiency of differem plans, serving policy makers. This study deeply explains the semanties, grammar and utility function structure of the Bayesian decisions ntework. It also explores the methods and channels of making decisions models and realizing algorithm. Take icecream dealers for example in a supply chain management. It shows that it is efficient and effective to apply the Bayesian decisions network in solving the complicated problems, especially decision-making of many parts involved.