• 论文 •    

Web服务集成的策略规划

曾伟,金杨恒   

  1. 华中科技大学 系统工程研究所,湖北武汉430074
  • 出版日期:2008-02-15 发布日期:2008-02-25

Policy planning of Web services composition

ZENG Wei, JIN Yang-heng   

  1. Institute of Systems Engineering, Huazhong University of Science & Technology, Wuhan 430074, China
  • Online:2008-02-15 Published:2008-02-25

摘要: 针对Web服务存在的业务逻辑不确定性和服务质量不确定性问题,提出了利用马尔可夫决策理论和高级逻辑程序语言DT-Golog来解决Web服务集成中的不确定性规划问题的方法。该方法首先将Web服务调用视为不确定性动作,建立Web服务集成的马尔可夫决策模型。在此基础上,利用情景演算理论及其扩展的马尔可夫决策理论的DT-Golog高级逻辑程序语言,对Web服务的集成逻辑进行编程,同时基于马尔科夫决策理论,获得了Web服务集成的最优策略。通过示例说明,利用DT-Golog解决Web服务集成问题,体现了编程和规划相结合的优点。

关键词: Web服务, 马尔可夫决策过程, 情景演算

Abstract: To deal with uncertainty of business logic and Quality of Service (QoS) in Web services, Markov Decision Processes (MDP) and high level logic programming language DT-Golog were used to solve uncertain planning problem of Web Services Composition (WSC). Web service invocation was regarded as a stochastic action, and MDP formal model was used to describe WSC. Moreover, situation calculus and its high level logic programming language DT-Golog, which was augmented with MDP, were used to give the logic representation of WSC, and the optimal policy of WSC was obtained by planning based on MDP. Finally, programming and planning combination advantages of DT-Golog to solve WSC were also illustrated.

Key words: Web services, Markov decision process, situation calculus

中图分类号: