 |
International Society of Science and Applied Technologies |
| Fault Tree Analysis Model for Remanufactured Machine Tools by Using Generative AI and Analytic Hierarchy Process | ||||
| Author |
Kuo-Ping Lin
|
|||
| Co-Author(s) |
Yueh-Ju Kuo; Ting-Yu Lin; I-Jan Wang
|
|||
| Abstract |
This study develops a systematic framework for reliability allocation in remanufactured machine tools by using Fault Tree Analysis (FTA), Analytic Hierarchy Process (AHP), and the application of large language models (LLMs) to optimize reliability distribution across subsystems. The framework leverages Artificial Intelligence-Generated Content (AIGC) models to refine allocation strategies based on expert inputs and predefined criteria, enhancing predictive accuracy. A case study on an NC gear hobbing machine demonstrates the proposed method’s effectiveness. The LLM-generated reliability values exhibit comparable accuracy to those obtained from expert surveys, while significantly reducing the time and resources required for data collection and processing. This approach not only streamlines the reliability allocation process but also mitigates potential biases inherent in expert assessments, providing a scalable, consistent, and cost-effective alternative for reliability analysis in remanufactured systems.
|
|||
| Keywords | Remanufactured machine tools; Fault tree analysis; Large language models | |||
| Article #: RQD2025-122 | ||||
Proceedings of 30th ISSAT International Conference on Reliability & Quality in Design |