碩士 / 國立成功大學 / 工業與資訊管理學系碩士在職專班 / 103 / While researchers are paying more attention to production capacity in production management, less attention were paid to facility maintenance and supplement of spare parts. A maintenance department, aiming to maintain a high volume of output, should ensure sufficient quantities of spare parts are available to repair machines when malfunctions occur. This is to minimize any adverse effects on the production process and production capacity. The aim of this study is to investigate the demand forecasts for spare parts for repairing packaging machines in the artificial fiber manufacturing industry. Real data is used to compare differences among the time series methods, regression, an artificial neural network, and the rules of thumb in case company. The accuracy of the model is assessed by the Root Mean Square Error. The results show that the artificial neural networks method provides the best level of accuracy and goodness of fit. On the contrary, a multiple regression method performs poorly as far as accuracy and goodness of fit concerned. The time series method and the heuristic rule provide the worst level of accuracy and goodness of fit.
| Identifer | oai:union.ndltd.org:TW/103NCKU5041065 |
| Date | January 2015 |
| Creators | Jia-TingChang, 張嘉庭 |
| Contributors | Tai-Yue Wang, 王泰裕 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 60 |
Page generated in 0.0294 seconds