The Behavior Analysis Of Web Service Composition Based On Petri Net

At present, the Web service composition research focuses on demand. It thinks about the function perspective more than the service behavior perspective. This paper, from the service behavior perspective, mainly studies the following two contents:(1) At present, the correctness of the Web service composition is primarily analyzed by soundness, and the soundness has caused the attention of scholars at home and abroad. Now, many researches are mostly based on Petri net, the finite state machine or automatic machine theory. However, it is difficult to determine the soundness, and it is also a high requirement for the correctness of the Web service composition. So the paper considers the weak soundness. The paper finds that the algorithm for judging the behavior relativity is feasible only in the case that T-invariant exists. In the consideration of the fact, Web services are modeled by the Open Petri Net. Then, in order to realize the correctness analysis of the composition, an algorithm for determining the weak soundness of the service composition based on the theory of the service tree is presented. Finally the concrete example shows the effectiveness of the proposed method.(2) A service-oriented system is composed of independent software unit, namely services, that interact with one another exclusively through message exchanges. The proper functioning of such system depends on whether or not each individual service behaves as the other services expect it to behave. Since services may be developed and operated independently, it is unrealistic to assume that this is always the case. So it causes a problem that the actual behavior of a service conforms to the expected behavior? The paper models the expected behavior by Petri nets and transforms the actual behavior into transition sequence teams. Then, the paper checks conformance between Petri nets and transition sequences from fitness and appropriateness. Conformance check, also referred to as conformance analysis, aims at the detection of inconsistencies between Petri nets and its corresponding execution sequences. Fitness means that how many of the execution sequence can be accepted by the Petri net. Appropriateness is used to describe the proportion of the execution sequences in all sequences of the Petri net model. Appropriateness can be evaluated from both a structural and a behavioral perspective. Next, the paper presents an algorithm for determining the behavior correctness based on conformance checking. Finally, the experiment result shows the advantage of the method.