碩士 / 國立高雄應用科技大學 / 土木工程與防災科技研究所 / 94 / There are a few restraints and drawbacks in existing fuzzy regression models as follows: (1) only deal with the situation which independent variables are crisp data, dependent variables is fuzzy data, and subject to the fuzzy data are symmetric triangular fuzzy numbers. (2) the interval of prediction is too large and easy to be affected by outliers, (3) computational procedure is complexity and difficulty. Therefore, this study built two fuzzy regression models without deficiencies discussed above and name as (1) fuzzy approximate least-squares regression and (2) fuzzy extended least-squares regression, respectively. Besides, previous fuzzy regression models can’t deal with variables which is fuzzy values, crisp values and mixed values ( crisp and fuzzy values exist together) at the same time, but the fuzzy extended least-squares regression provided in this study solves the problems listed above. Numerical examples are used to describe the analytic steps and demonstrate the feasible and practicability of the two models in this thesis. Finally, the bridge’s deck deterioration in the middle section of Jong-Shan freeway will be discussed by using fuzzy extended least-squares regression, and the regression model built by extended least-squares regression can be provided as the decision references of bridge maintenance strategy for bridge engineers.
| Identifer | oai:union.ndltd.org:TW/094KUAS0653026 |
| Date | January 2006 |
| Creators | Kuan-Chih, Huang, 黃冠智 |
| Contributors | Nan-Fei, Pan, 潘南飛 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 0 |
Page generated in 0.0122 seconds