Learning and reasoning method using fuzzy coloured Petri nets under uncertainty

Petri nets have been widely used to model computer systems. Manufacturing systems, robotics systems, knowledge-based systems, and other kinds of engineering applications. Further, to present complex real-world knowledge fuzzy Petri net models have been proposed to perform fuzzy reasoning automatically. However, in Petri nets one has to represent all kinds of processes by separate subnets even though the process has the same behavior as another. Real-world knowledge often contains many parts which are similar, but not identical. This means that the total number of Petri nets becomes very large. Therefore, it becomes difficult to see the similarities and the differences among the individual subnets representing similar parts. The problems may be annoying for a small system, and catastrophic for the description of a large-scale system. To avoid this kind of problem the authors propose a learning and reasoning method using fuzzy coloured Petri nets (FCPN) under uncertainty. For the correction of rules of the knowledge-based system a hand-built classifier and empirical learning method based on domain theory have been proposed as machine learning methods, where there is a significant gap between the knowledge-intensive approach in the former and the virtually knowledge-free approach in the letter. To resolve such problems simultaneously they propose a hybrid learning method which is built on top of the knowledge-based fuzzy coloured Petri net and genetic algorithms.