Application of Petri-nets in object-oriented environment

摘要

The Petri-net is an effective modeling tool, especially in the area of the concurrent events, and it has a very intuitive characteristic: the non-deterministic simulation. This behavior is hard to be implemented in an imperative language, and its parameters even more difficult to be changed. This was the reason why this paper was selected the aim: embed Petri-nets into an imperative high-level language and use its beneficial properties as clean as it possible. To reach that goal, first of all a well-formed description was needed about Petri-nets, and its extensions, and also need a new extension to model the object-oriented events inside the simulation. We have to identify the moments of the transitions' activation (transition firing), the token player movement and the change of the token distribution. When somebody wants to implement a state machine which able to handle the transitions between the application's states, a Petri-net model creation may be needed to help the understanding of the task. In that case the model should not be only kind of static document of the product. With a new Petri-net extension it can be highlight connections between the model and the real source code events. If somebody changes the model it will take effect "immediately" in the product. In the agile world the IT industry is also changing: the speed of the changes is faster than the chance to generate a modified model of any applications.