Efficient tools for analysis and implementation of manufacturing systems modelled by Petri net with objects: a production rules compilation-based approach

Abstract

An attempt is made to show how well-designed rule-based systems can be used for analysis and implementation of real-time manufacturing systems modeled a by Petri net with objects (PNO). After defining the PNO model, the authors outline some problems that one can encounter when implementing it. On account of the PNO similarity with rule-based systems, it is shown how translating a PNO specification into an equivalent rule-based program can narrow the gap from a PNO specification to its implementation. However, PSs run quite slowly and it is shown how a rule compilation-based technique can decrease the PNO execution cost by improving the computational efficiency of the rule interpreter at the pattern-matching level. The authors introduce a novel approach for this end in light of specific features provided by the PNO model. They present analysis tools built from this approach, underlining solutions for invariant searching and state graph generation problems. The distributed implementation issue is addressed, and a solution is proposed.<>