English  |  正體中文  |  简体中文  |  Items with full text/Total items : 26988/38789
Visitors : 2357590      Online Users : 31
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search

Please use this identifier to cite or link to this item: http://ntour.ntou.edu.tw:8080/ir/handle/987654321/41798

Title: 新模糊貝葉氏風險評估模型之 發展及其在船舶安全之應用
Authors: 翁順泰
Contributors: 國立臺灣海洋大學:商船學系
Keywords: 海事風險評估;模糊法則庫方法;貝葉氏推論;模糊德菲爾法;Maritime Risk Assessment;Fuzzy Rule Base Approaches;Bayesian Network;Fuzzy Delphi Method
Date: 2014
Issue Date: 2017-03-28T02:01:25Z
Abstract: 海事產業由於尚無建置船舶安全資料庫,多數海事風險評估架構係以模糊法 則庫為基礎建構而成,這是因為該方法具備能將專家質化意見轉為量化數值之優 點。然該類型研究中,各法則前項變數狀況組成不同,卻常以相同程度描述法則 後項結果,因此法則組成之邏輯性遭到質疑;此外,法則多寡需視輸入變數及語 言詞數量而定,隨變數及語言詞之增加,建立模糊法則過程將過於耗時;另外, 推論過程無法考量所有變數歸屬函數值,造成部分資訊遺失,在變數增多的情況 下,其隱藏資訊的狀況會更加嚴重;部分研究為避免變數資訊遺失,以模糊貝葉 氏推論為基礎整合所有變數,然變數資訊結合過程並未考量變數權重,且模糊法 則的組成邏輯性亦受到質疑。 本計畫預計建立一個近似推論風險評估架構,將以模糊德菲爾法將變數資訊 轉為先驗機率,並將以信賴結構方式發展法則庫,其將具備考量不同變數之特性, 並建立後項語言詞之歸屬函數,使法則結果可更符合專家思維與人類經驗法則之 邏輯;且以貝葉氏推論運用電腦軟體將各變數之先驗機率與法則後項之條件機率 整合,快速求得後驗機率,最後以法則後項語言詞之歸屬函數為基礎將後驗機率 轉換為一般機率值。本研究架構將會以文獻以及實際船舶安全案例驗證其實用性。
Due to the lack of failure databases, most maritime risk assessment studies are established based on fuzzy rule base approaches. This is because such methods are capable of transferring qualitative opinions from experts into quantitative outcomes. However, in such studies rules which are equipped with different compositions in the antecedents may have identical consequents. Thus, the logic of the rule development is often questioned. In addition, the quantity of fuzzy rules to be developed for a rule base depends upon the number of variables and linguistic terms adopted. Accordingly, it may be time consuming once the parameters under consideration and the associated terms are numerous. Also, the fuzzy rule base studies often apply min-min-max techniques and this, however, may cause the loss of useful information from experts. Such situations will deteriorate once the number of the variables increases. On the other hand, allowing for the loss of variable information, the studies based on fuzzy Bayesian Network (BN) have been developed to combine factors. However, such approaches do not consider the relative importance of each parameter and the logic of rule combination is also criticized. In this research, an approximate reasoning will be developed. The fuzzy information of each input will be converted into a probabilistic nature using the fuzzy Delphi method. A rule base will subsequently be constructed on the basis of the belief degree concept. It will have the features of considering the relative importance of each variable and the membership functions of the fuzzy sets established for the consequent. The level of consistency with human perceptions of the proposed study will be higher than that of traditional literature. The BN reasoning will then be deployed to combine the factors in a fashion where the input data and the consequent of each rule will be treated as the prior and conditional probabilities using computer software. The outcome will finally be transferred into the property of generic probability based on the membership functions constructed for the consequent of each rule. The developed model will be verified and validated using the ship safety related case studies acquired from literature and the maritime industry.
URI: http://ntour.ntou.edu.tw:8080/ir/handle/987654321/41798
Appears in Collections:[商船學系] 研究計畫

Files in This Item:

There are no files associated with this item.

All items in NTOUR are protected by copyright, with all rights reserved.


著作權政策宣告: 本網站之內容為國立臺灣海洋大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback