Based On Object-Oriented Development Of Graphic Modeling And Analysis Tool For Stochastic Petri Nets

Flexible manufacturing system (FMS), has been developed to improve automation and flexibility of manufacturing systems. Described as a discrete event dynamic system, it is assumed very complex with all forms of Petri nets engaged in the study of FMS as a mainstream technology for modeling and analysis, as well as a powerful tool for performance evaluation and control of FMS.Recent years has found more and more efforts being devoted to research of discrete event dynamic system with stochastic Petri nets.Thanks to stochastic Petri nets'complexity, developing software for stochastic Petri nets, including tools for general and special purposes, is helpful for extending both theoretical research and application of the nets. This study is intended to show how to design and develop SPNSim, a graphic modeling and analysis tool by analyzing the current situation of correlative software for stochastic Petri net home and abroad and the practical need to carry out the performance analysis of the system, and by drawing upon Microsoft Foundation Class Library of the visualized development environment Visual C++in an Object-Oriented approach.At first, the thesis presents a review upon the theories of Petri nets and stochastic Petri nets, and demonstrates how to apply stochastic Petri nets in the system performance analysis, with a concrete stochastic manufacturing unit.And then, the thesis explains functions achieved by the SPNSim software, designs the framework of software, and realizes the functions of software modules. The interface of SPNSim software is friendly to users, who can do visualized modeling with the editor of the Petri net draw, which is very convenient. Also extension and maintenance of the software is more than practical if, by exploiting Petri nets'features, many classes are created by abstracting elements.In addition, Windows Message responses and Doc/View structures are used to make sure that each module has got clear missions and the whole software system is easily maintained. Besides, the software can be an accessible graph of the Petri net models; on such basis can some properties like incidence matrix, reachable graph and steady probability are gained. And last but not least, all performance specifications of the SPN model performance are obtained. Finally,the software's modeling and ananalysis course are demonstrated by one instance, and the correctness of the modeling software is proved.