Anzeige
Erschienen in:
2013 | OriginalPaper | Buchkapitel
A Hidden Markov Model for Tool Wear Management
verfasst von : Chen-Ju Lin, Chun-Hung Chien
Erschienen in: Proceedings of the Institute of Industrial Engineers Asian Conference 2013
Verlag: Springer Singapore
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Determining the best time of tool replacement is critical to balancing production quality and tool utilization. A machining process would gradually produce defective parts as a tool wears out. To avoid additional production costs, replacing a tool before the yield drops below a minimum requirement is essential. On the other hand, frequent tool replacement would cause additional setup and tool costs. This paper proposes a hidden Markov model (HMM) to study the unknown nature of a tool wear progress by monitoring the quality characteristic of products. With the constructed model, the state of tool wear is diagnosed by using the Viterbi Algorithm. Then, a decision rule that evaluates the yield of the next machining part is proposed to determine the initiation of tool replacement. The simulation analysis shows that the proposed method could accurately estimate the model and the status of tool wear. The proposed decision rule can also make good use of tools whereas controlling yield.