Goal-Oriented Service Refinement Based on Dynamic Planning

To improve the flexibility and adaptability of the automatic service composition, this paper proposed a goal-oriented service refinement approach to transform the high-level abstract service into a number of service operators to meet a specific request. First, a goal-oriented abstract service model is presented to describe the most relevant solution corresponding to the consumer´s profile and the request context. Second, the mechanism of service refinement was provided to realize data-gathering, service generation and service late binding. Meanwhile, the dynamic planning algorithm was designed to generate abstract services and to select concrete services corresponding to the current environment. Finally, an online shopping case study was used to show that the proposed method can be effectively utilized in practice. The result indicated that this method enabled users to operate it more easily and more dynamically.