An Expandable Petri Net Framework Based Method for Behavior Evaluation of Reconfigured Equipment

• Main Authors: Wan-Ling Li, 李宛玲
• Other Authors: none
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/9w2367

碩士 === 國立高雄第一科技大學 === 系統資訊與控制研究所 === 96 === Efficient analysis of equipment scenarios can reduce time to confirmation of equipment scenarios in the equipment developing phase, and it also can enhance feasibility analysis of scenario reconfiguration in the equipment usage phase. However, regardless of developing, reconfiguration, and extension of equipment scenarios, these results of scenario analyses were limited from the conventional models which should be reconstructed and re-evaluated by change to answer whether the reconfigured scenario is feasible. Consequently, this paper will consider how to effectively construct an equipment behavior model, especially behavior sequences, to efficiently analyze the scenario with a model framework. We propose a method for evaluating feasibility of Reconfigurable Equipment Behavior (REB) based on the Extendable Petri Nets Framework (EPNF). In the EPNF, a variety of REB components were constructed to specify various equipment behaviors. The specific scenarios by equipment are described by UML based notation use cases and sequence diagrams, transformed into Petri Nets using the proposed mapping rules and forward transformation method, and synthesized with the corresponding REB components to construct the specific Petri Nets behavior model. After the model constructed in developing phase, the feasible behavior sequences of the equipment can be evaluated by the proposed hierarchical method and the partial reachability tree. As a case study on a semiconductor cluster tool, the results indicated that the EPNF can efficiently integrate and unify the scenarios reconfigured into the behavior model to evaluate variety combinations of the equipment behaviors. Especially, on acquiring the feasible sequences, the REBs can be derived by the partial reachability tree that avoids the explosion problem in searching sequences.