扩展粗糙集模型研究及其在供应商选择中的应用
发布时间：2018-12-18 15:40
【摘要】：粗糙集理论是波兰数学家Pawlak在1982年提出的，是概率论、模糊集理论之后又一处理模糊性、不确定性数据的数学工具。该理论的特点是不需要任何先验知识或附加信息，在多属性决策问题中的指标筛选和排序选优等方面有很好的应用前景。 经典粗糙集主要针对完备信息系统，然而在现实生活中由于数据测量的误差，对数据的理解或获取的限制等原因，使得在知识获取时往往面临着不完备的信息系统，即可能存在部分对象的一些属性值未知的情况，本文首先在完备信息下模糊决策可变精度粗糙集模型得基础上，根据隶属度函数给出了处理不完备信息多属性决策下的粗糙集属性约简算法，并通过算例分析检验模型的可行性。 另外，，针对多属性决策问题中的排序选优问题，以往研究大多默认属性之间是可以相互补偿的，而现实生活中也存在着属性之间不能完全补偿的情况。针对这一情况，本文提出了粗糙集层次权重确定法和相对信息熵改进的信息熵扩展粗糙集排序模型，综合考虑方案的整体性和均衡性，能够在一定程度上提高排序结果的准确性。 最后将排序模型应用于供应商选择，构建了适合化工设备零部件供应商选择的指标体系，并采集数据通过实例分析说明其应用价值，比较线性加权模型和改进后模型的评价值，计算结果表明改进后的模型更加符合企业的实际选择结果，证明方法是科学有效的。
[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.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F274;F224
本文编号：2386062
[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.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F274;F224
【参考文献】
相关期刊论文 前10条
1 滕书华;周石琳;孙即祥;李智勇;;基于条件熵的不完备信息系统属性约简算法[J];国防科技大学学报;2010年01期
2 菅利荣;刘思峰;方志耕;党耀国;朱建军;吴和成;姚天祥;;基于优势粗糙集的教学研究型大学学科建设绩效评价[J];管理工程学报;2007年03期
3 吴立云;杨玉中;;基于粗糙集-熵理论的绿色供应商选择模型研究[J];工业工程与管理;2011年02期
4 杨善林,刘业政,李亚飞;基于Rough Sets理论的证据获取与合成方法[J];管理科学学报;2005年05期
5 王国胤;Rough集理论在不完备信息系统中的扩充[J];计算机研究与发展;2002年10期
6 徐伟华;张晓燕;张文修;;优势关系下不协调目标信息系统的下近似约简[J];计算机工程与应用;2009年16期
7 滕书华;魏荣华;孙即祥;谭志国;胡清华;;基于不可区分度的启发式快速完备约简算法[J];计算机科学;2009年08期
8 魏大宽;黄兵;周献中;;不完备模糊目标信息系统粗集模型与知识约简[J];计算机工程;2006年08期
9 徐伟华;张晓燕;张文修;;优势关系下不协调目标信息系统的上近似约简[J];计算机工程;2009年18期
10 陆秋琴;和涛;黄光球;;多重变精度粗糙集模型[J];计算机应用;2011年06期
本文编号：2386062
本文链接：https://www.wllwen.com/jingjilunwen/jingjiguanlilunwen/2386062.html
