發(fā)布時間：2018-12-18 15:40

【摘要】：粗糙集理論是波蘭數(shù)學家Pawlak在1982年提出的，是概率論、模糊集理論之后又一處理模糊性、不確定性數(shù)據(jù)的數(shù)學工具。該理論的特點是不需要任何先驗知識或附加信息，在多屬性決策問題中的指標篩選和排序選優(yōu)等方面有很好的應(yīng)用前景。 經(jīng)典粗糙集主要針對完備信息系統(tǒng)，然而在現(xiàn)實生活中由于數(shù)據(jù)測量的誤差，對數(shù)據(jù)的理解或獲取的限制等原因，使得在知識獲取時往往面臨著不完備的信息系統(tǒng)，即可能存在部分對象的一些屬性值未知的情況，本文首先在完備信息下模糊決策可變精度粗糙集模型得基礎(chǔ)上，根據(jù)隸屬度函數(shù)給出了處理不完備信息多屬性決策下的粗糙集屬性約簡算法，并通過算例分析檢驗?zāi)Ｐ偷目尚行浴?另外，，針對多屬性決策問題中的排序選優(yōu)問題，以往研究大多默認屬性之間是可以相互補償?shù)�，而現(xiàn)實生活中也存在著屬性之間不能完全補償?shù)那闆r。針對這一情況，本文提出了粗糙集層次權(quán)重確定法和相對信息熵改進的信息熵擴展粗糙集排序模型，綜合考慮方案的整體性和均衡性，能夠在一定程度上提高排序結(jié)果的準確性。 最后將排序模型應(yīng)用于供應(yīng)商選擇，構(gòu)建了適合化工設(shè)備零部件供應(yīng)商選擇的指標體系，并采集數(shù)據(jù)通過實例分析說明其應(yīng)用價值，比較線性加權(quán)模型和改進后模型的評價值，計算結(jié)果表明改進后的模型更加符合企業(yè)的實際選擇結(jié)果，證明方法是科學有效的。
[Abstract]:Rough set theory, proposed by Polish mathematician Pawlak in 1982, is a mathematical tool for dealing with fuzzy and uncertain data after probability theory and fuzzy set theory. The characteristic of this theory is that it does not need any prior knowledge or additional information, and it has a good application prospect in the field of index selection and ranking selection in multi-attribute decision making problems. The classical rough set is mainly aimed at the complete information system. However, in real life, because of the error of data measurement, the limitation of data understanding or acquisition, etc., it is often faced with incomplete information system when acquiring knowledge. That is to say, there may be some unknown attribute values of some objects. Firstly, based on the fuzzy decision variable precision rough set model under complete information, According to the membership function, a rough set attribute reduction algorithm for multi-attribute decision making with incomplete information is presented, and the feasibility of the model is verified by an example. In addition, in order to solve the problem of sorting and optimization in multi-attribute decision making, most of the previous researches have shown that the default attributes can compensate each other, but in real life, there are some cases in which the attributes can not be fully compensated. In this paper, a hierarchical weight determination method based on rough set and an improved information entropy extended rough set sorting model are proposed in this paper, considering the integrity and equilibrium of the scheme. It can improve the accuracy of sorting results to a certain extent. Finally, the ranking model is applied to supplier selection, and the index system suitable for the supplier selection of chemical equipment parts is constructed. The application value of the model is illustrated by a practical example, and the evaluation values of the linear weighted model and the improved model are compared. The calculation results show that the improved model is more in line with the actual selection results of the enterprise, and proves that the method is scientific and effective.
【學位授予單位】：南京航空航天大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：F274;F224

【參考文獻】

相關(guān)期刊論文 前10條

1 滕書華;周石琳;孫即祥;李智勇;;基于條件熵的不完備信息系統(tǒng)屬性約簡算法[J];國防科技大學學報;2010年01期

2 菅利榮;劉思峰;方志耕;黨耀國;朱建軍;吳和成;姚天祥;;基于優(yōu)勢粗糙集的教學研究型大學學科建設(shè)績效評價[J];管理工程學報;2007年03期

3 吳立云;楊玉中;;基于粗糙集-熵理論的綠色供應(yīng)商選擇模型研究[J];工業(yè)工程與管理;2011年02期

4 楊善林,劉業(yè)政,李亞飛;基于Rough Sets理論的證據(jù)獲取與合成方法[J];管理科學學報;2005年05期

5 王國胤;Rough集理論在不完備信息系統(tǒng)中的擴充[J];計算機研究與發(fā)展;2002年10期

6 徐偉華;張曉燕;張文修;;優(yōu)勢關(guān)系下不協(xié)調(diào)目標信息系統(tǒng)的下近似約簡[J];計算機工程與應(yīng)用;2009年16期

7 滕書華;魏榮華;孫即祥;譚志國;胡清華;;基于不可區(qū)分度的啟發(fā)式快速完備約簡算法[J];計算機科學;2009年08期

8 魏大寬;黃兵;周獻中;;不完備模糊目標信息系統(tǒng)粗集模型與知識約簡[J];計算機工程;2006年08期

9 徐偉華;張曉燕;張文修;;優(yōu)勢關(guān)系下不協(xié)調(diào)目標信息系統(tǒng)的上近似約簡[J];計算機工程;2009年18期

10 陸秋琴;和濤;黃光球;;多重變精度粗糙集模型[J];計算機應(yīng)用;2011年06期

本文編號：2386062