Automatic Adaptation of BPEL Processes Using Semantic Rules: Design and Development of a Loan Approval System

Dynamic service composition is an important challenge for many applications. This paper considers dynamic composition where the parties involved and the process they execute changes based on context. In such a setting, creating compositions from scratch is costly. Instead, we advocate automatic adaptation of existing compositions to fit the requested service demands. Our approach starts from existing BPEL processes. The details and constraints on the environment are expressed by semantic rules. We automatically decide which services are necessary to complete a desired service in a given BPEL process and which services can be removed from the process by using a reasoning engine. Based on this information, our flow generator modifies existing BPEL pivcesses to derive effective variations. The newly generated BPEL process is then executed. We study this approach in the context of a loan approval system and show how the modifications can be done on example scenarios.