| Summary: | 碩士 === 義守大學 === 工業工程與管理學系碩士班 === 97 === To maintain a good relationship with customers, a quick response to corrective action request is of utmost important to airplane maintenance companies. System Equipment Maintenance Status Record (CAFM 0020 form) is what they mainly use to deal with customers’ requests. However, the probem-solving engineers can only provide relating analysis and solutions using their past experiences and knowledge.
This research centers on the development of a quality diagnosis system for airplane maintenance industries, with a practical experience of some airplan repair hangar. The knowledge provided here is based on finding technical data to diagnose various discrepancy. As is CAFM 0020 form, it documents all the corrective actions and faults, then uses SOM neural networks to draw a sequence chart of diagnosis, which will be used to address the cause of discrepancy and solutions. It’s from this outcome, the percentage of discrepancy get much lower than brfore through SOM, and the chart. The diagnosis system will be comprehensive and more helpful to largely control the timing of such failures. Further, it can reduce big impact on production scedules, safty costs and training costs for an untrained, unskilled staff, and providing airplane industries with a complete guide to effectively analyse production quality.
|
|---|
